make it look Iike a fish's dorsd fin. In adciition.

nre hastet1 to point out that some of the illustrations dep~ct tadpole with .I almost no tall fins, toothlike strtictures on the fisM1ke lower jm; front leg that looks very much like a pectoral fin, and ~vhat appear to be scales on tllc tail. Mqo, the sequence of e i ~ h~lluarations t $haws a recently transformed metamorph progressing Into a fislihke tadpole! Our review sugpeas that proqess on understand~ng nature of tadpoles n.as belng the made but slowlv m d rel~ictantl!: if this paper is any inciication. The fact that the tadpole was d r a ~ ~ n right latcral in view, t l ~ u s precluding the illustration of the spiracle on the left side. also 15 of note. Recause of thi5 common structural asvmmetn: it has become conventional to ill~15trate tadpole5 in-left late'ral vlen: The earliest figure and descript~on a tadplrv=i mouthof parts were those bv Swamn1erd.im (1737-17.78, pl. 40, fig. 1). Nearlv a cennlnf elapsed before Saint-Arise (183 1) described tadpoIe mouthparts and Di1gGi.s (1834) described and pamally ~llustrated a n d m o u t h p . 1 ~ ~h ~ comparative l In s studv of the o ~ t c o l o p m v o I o p of ,imphrbl~nc.These and cla5s1cd works \trere t'ollo\ved bv the fi rst hlstologlcal s n ~ d v of the de\~clopmcntof 1ahia1 teeth and j.11~. 4ieaths (Vogt 1842). Van Bambeke (1863) pioneered the use of lama1 traits, zncluding rnouthpnns. as rayononilc charac~enand n.as the first to note lnterspecific ditference5 among European specles. There Followed sc\,eral s n ~ d i e ~ tadpoles of on Europe,m frogs. Amon9 the more noteworth\. are papen hv Lrvdig (1876) on structural 1ari.ition in morrthparts and Lamste ( 1879) on structural differences in inchvidual labid teeth of L)ircqflfo.c~~ts pictrrs; L a ~ s t c l ~ o a generated farnil id names based on chancteri~tics the s p ~ ~ i c(e.g., Medioof l~s gninidae [midventral spiracle] m d L c v o ~ r i n i d a e [s~nirtral spiraclc]). Ke~ffcr (1888) d~scussed development and disthe tribunon of labla1 teeth on the oral d ~ s c f A h ~ cobs~rfi-iurrzs, o s and Schulze (1888) did the same for Peloltntrrfi~scrrs. Gutzeit (1889) expanded on these comparative \vorh and studied the development of 13wsheaths. Papers by Hiron-Koyer and V J ~ Bmbeke (1881, 1889) described and compared the formation and structure of the mouthparts of tadpoles of 22

Ftg 1.2. An early rendition of rhc tadpole of Px~snrdispnmdn.vn from Hutchinson 1797, nr ,\Tnhtm[ Hist? o rlrr Fmq Fir/) ofStrrirmrrr, t f PI. 3.

kinds of f r o p (15 currently recognized European and 2 African species) and built on the earlier work of Van Bmhekc by recognizing that terrain morphologicd traits were uscful in distinguishing among tadpoles of different species. They d s o pointed o~rt those 1nn.d characteristics usefill at the generic and E~miliallevels. Roulcnger (1892) recognixci the value of being able to identi& tadpoles and presented a synopsis of u s c f ~ ~ a n d characters for distinguishing amon% ll species. He darifiecl and expanded the results of previous work, affimied the value of the position of the spiracle and vent tube in distinguishing among groups, argued for including patterns of pigmentation in species descriptions, was the first to p r o p s c a labial tooth r m r formula, discussed the relative value of the distribution of neuromasts in species characterizations, and wrote a key to the tadpoles of 1 Y species of Europcan anurans. This paper had a significant influence a n future workers. Tllc cxccllent illustrarions of the tadpoles m d mouthparts, some of the best elverpublished, nrerc reissued in his nionograph on thc tailless batrachians of Europe (Roulenger 1897-1898). Roulenger described only what he cnllcd "matlire' tadpoles, and only recently have we started looking at all stages of dc\rclopment in our delibcrations o n tadpole morphc~lop. Interest in thc tadpoles of North h e r i c a n frogs also was de\reloping in the later pare of the nineteenth ccntury. At almut die same time that Hkron-Royer and Van Ranibcke (1881) published their iniprt.int paper on European tadpoles, M a n Hlnckle!. independentiy had beLgun to study the lanae of anurans in AIassachusetts (Hitickley 1880). Her subsequent paper describing the mouthparts oitadpoles of seven species was the first compantive work on North American muran l a n w (H~nckley1881) and marked the advent of comparative devclopmentd studies of tadpoles in the Nelv \170rld. She quickly foIlo~vxiwith dc\~elop~iiental studies of nvo other species (Hlnckle!r 1882, 1884). Mre have observed that older literature, even though niformarive, often is slighted m d that cunous obsenrations reported therein too oi'ten are missed. For example, ~ i n c k l e y (1 882:95) reported that Rnnn gdraricn occasionaUy has dual spiracles, a condition not known in rmid tadpoles. In his cInssic stirdies o n North American .kura, A. H. Wright (lC?14, 1932) published detailed descriptions and photop p h s of taciplcs and clrawings of mouthparts of species occurring near Ithaca, New York, and from the Okefenokec region o southern Gcorgia. In a separate svnopsis, A. H. f \\'right (1529) described, illusmted, and presented a key to tadpoles of 38 species mostly from the eastern and southern Cnlted States. Some information on tadpoles appeared In the first (1933) and second (1942) edinons ofdie Hn~rdltaoX. o Frqgs nnd Ends by A. A. Wright and Wright, but it wasn't f unnI the third edition (1949; rcpr~ntedIn 1995) by A. H. IVright and tjTrisht that a kev and photographs -of tadpoles 2nd their mouthparts appeared. Unfortunatel!; A. H. \lrnsht continued to use only "niature" larvae because "halfgro\ipnlarvae . . . are ofien quite aknormaI in the usual characters used In I,zn.al descriptions." U'hilc the early w o r k by A. H. Ct'r~ghtwcre uulv tlseful, his preoccupation with vari\ln\!~ expressing relof ation ~d measurements m d c o m p l e ~

28

RONALD ALTIG AND ROY W. McDIARMID

usages may be inappropriate. If tadpole tails were classified by the same terminology as fish tails, a tadpoles wouid be diphycercai, aithough presumably not homologous (in diphycercal fishes, a mediai fins fused into one unit). Terms suggested by Van Dijk (1966) apply oniy to curvature of the tad axis. Fins of lotic tadpoles are lower, thicker, and stiffer than the taer. thinner more flexible ones of lentic forms. The amount and distribution of connective tissue in the taiis of lotic versus lentic forms have not been studied. Kinematics comparing swimming between the extremes of taii morphology surely wouid be worth investigating, but oniy pond forms have been studied (chap. 9). Van Dijk (1966) and Lambiris (1988a, b) noted that the basai ~ortion the tail muscle and adiacent fins of Hemisus of tadpoles has a sheath of thick connective tissue visible in live Tail Morphology and preserved specimens (see figures in chap. 12). This tail Some aspects of tad size and shape have been discussed pre- sheath is more common than previously suspected but when viouslv. Here we ex~lore taii structure in more detail. present, is visible oniy under specific, but undetermined, cirthe The tad muscle is composed of bilateral myotomic muscle cumstances of preservation. We know of no explanation of masses divided by V-shaped myosepta, the points of which what the structure is, even if it is conneaive tissue, or of face anteriorly; the epaxiai portion is smaer than the hypax- how it funaions, but it does occur in other taxa (e.g., O. L. iai portion. The unsupported fins are composed of loose Peixoto, personal communication, Hyh mamuwata group; connective tissue covered by epidermis (Lebedinskaya et ai. I. Grfiths and De Carvalho 1965, StereocycI.ps; personal ob1989; Rhodin and Lametschwandtner 1993; Yoshizato servation, Gartropbvyne carolinensis and Rana sphenocephah). Anvone who works with tadooles from different habitats 1986). Fins extend from various points on the body and taii posterioray to slightiy beyond the tip of the notochord and soon questions the influente of environmentai factors on tad muscle. Dorsai fins originate at or near the tad-body morphology. Unfortunately, few definitive data are available. junction (most common), at various points anterior to the Tadpole color and coloration obviously vary in different dorsai tail-body junction as far fonvard as the plane of the environments, and Van Dijk (1966) questioned if tadpoles eyes (e.g., &sina senegalensis and Scinm mbra group), or raised in shaow water have lower fins than normal. Tadat various points aiong the taii muscle (see fig. 12.1 and poles of Rana chiricahuensis captured in flowing water difsketches of examples of tadpoles in various genera within the fered in the degree of development of abdominal musculataxonomic accounts). The margins of low taii fins usuay are ture, pigmentation, and shape compared with conspecifics nearly parde1 with the margins of the taii muscle. Extremes from still water (R. D. Jennings and Scott 1993). Martinez in this case are represented by suctorial species, Poyntoniapa- et ai. (1994) reported skeletai deformities in reared tadpoles ludicoh, and various semiterrestriai forms (e.g., Arthrolep- that were likely nutritionai. tides, Cycloramphus, Nannupbvys, and Thoropa). The anterior part of the dorsai fin of the tadpoles of Occidozy~a (sensu Limbs stricto) rises abruptiy (M. A. Smith 1916b) and then slopes The hind limbs of tadpoles are exposed during much of tadeveniy to the tip. The same descriptors apply to the ventrai pole development, and their pattern of development (see fin, which most commoniy originate at the ventrai terminus Alberch and Gaie 1983; B. I. Baiinsb 1972) is of primary of the body but may originate anywhere on the tail or ante- importance in comparing growth rate among tadpoles of rior to the vent tube at various points on the abdomen (e.g., different sizes and taxa under differing conditions (e.g., phyliomedusine hylids, pipids, some Rhaqhoms, and Znu- Gosner 1960; see fig. 2.1). Even though the front limbs are covered by the opercuium, they foliow a similar developpus in which a distinct lobe may be present). The taii tip takes many shapes: broadly or narrowly mental pattern. Based on our unpublished data, developrounded to various degrees or extended into a flagelium. A ment of the front limbs lags b e h d that of the h d limbs flagelium comprises the terminal part of the taii muscle and by about one stage (i.e., length-diameter changes in limb reduced fins; it is more or less distincdy separated from the buds in Gosner stages 26-30 and toe differentiation in anterior portion of the ti by a prominent reduaion in fin stages 33-37) in spite of the fact that oniy four digits occur ai height. ~ h flagelium ofien iBpigmented less and may undu- on the front limb. The apparent differentiai in limb develope late independentiy of the remainder of the tail. A flagelium ment increases greatiy in semiterrestriai tadpoles, but the admay be present in tadpoles of severai ecological types with justments in timing and pattern of the growth trajectories low (e.g., Hyh leucophylhta group, phyliomedusine hylids, required to reach proper metamorphic conditions are not and Znoptts) or high (e.g., some Hyh, Ihsina, and S c i ~ ) known. Drewes et ai. (1989) found a difference of about five fins. D u e h a n (1978) used xiphicercal to describe a taii that stages between comparable development of hind and front narrows abruptiy to a distinct flagelium. Comphentary ter- limbs ofArthroleptides, and Lawson (1993) saw tadpoles of minology for other forms of fins is not available, and such Petropedetes with oniy fuliy developed hind legs hopping
i

sophyne and Stephopaedes, sma African bufonids that commoniy breed in phytotelmata, have a circular, fleshy "crown" that encircles the eyes and nares (Channing 1978, 1993; Grandison 1980). We suggest that the hnction of this structure remains conjectural because none of the suggestions (e.~., - attachment, floatation, and respiration) seem feasible. Even t h o u ~ h dorsai and laterd ~rofiles the tad~ole of " the snout vary considerably and their shapes appear to be phylogeneticdy and ecological correlated, no descriptive terms have been proposed. Viewed from above, the snout varies from rathe;ac;te to broadly rounded. In lateral profiie (eyes to base of upper labium) the snout varies from uniformiy rounded to sloping, the trajectory of which turns abruptiy near the oral disc.

BODY PLAN

29

of elv~ia.a of whch are assumed to shade the eve from ' intense light. The umbraculum (e.g., Rana vertebralis) is a projeaion of the dorsal pupiilary margin. Elygia may occur as a hemispherical area of melanophores extending lateray from the kddorsal margin of the ;ris (ocular) or i i the dorsal cornea (epidermal).She corneas of an uddentified hyiid tadpole (personal observation) are mostly pigmented; if not a case of teratology, this rnay be an extreme form of epidermal elygium. Wassersug et ai. (1981a) noted similar pigmented corneas in Thelodemza carinensk. Recently we have observed an ocular elygium in a number of other taxa (e.g., some Buf, Hyh, Mantidactylus, Pseudamis, and Rana), although often it is visible only with bright dumination and at rnagnification of iive specimens. The eyes of Stauroisnatator are covered with "thick skin . . . but in stages 36 onward Eyes and NasolucrimalDuct there is a small, clear window over the eyeball" (Inger and Eye position and eye orientation are different concepts (fig. Tan 1990:4). 3.2A). Position indicates where the structures are located on The tadpole iris cornmonly has bright, metdic pigments the head and orientation describes their manner of place- arranped in comviicated Datterns that seem to be influenced " ment or aiignment (e.g., facing direction). Eye position is by the topography of underlying blood vessels. Interspecific generay described as dorsal or lateral, although a contin- pattems in the arrangement and distribution of color probuum exists. Dorsal eyes may face dorsay (upward), dorso- ably exist, but these features (e.g., D u e h a n 1978; Gaardo laterally, or lateray, while lateral eyes always face laterally. 1961; Scott and Jennings 1985) have seldom been examThe eyeba of tadpoles is covered by a two-layered cornea ined. Aithough the entire eye can be rnoved, the round pupil that is continuous with the body epidermis. Dorsal eyes have has no abiiity to adjust its size relative to iight intensity. no part of the eye or cornea in the dorsal siihouette, and During &id- to late metamorphosis, either dark pigmenlateral eyes have some part of the eye or cornea included in tation appears or the surroundmg pigmentation decreases in the dorsal siihouette (fig. 3.2). Dorsal eyes e x e m p e benthic intensity to accent a shallow groove that extends in an arc tadpoles in both lentic and lotic systems, whde lateral eyes from the anterior corner of the eve toward the naris. This are rnost common in lentic forms that spend considerable nasolacrimal duct develops in association with the Harderian time in the water column (e.g., many hyiids and microhy- gland (Schmalhausen 1968; see Baccari et al. 1990) that lulids). Relative positions of dorsal or lateral eyes can be quaii- bricates the eye. fied verbay or by measurements. Lateral eyes are typicay larger and have more cornea curvature and lenticular protru- Integument and Chromatuphores sion than dorsal eyes. Eye size ranges from minute (e.g., Numerou uniceiiular glands throughout the epidermis of centrolenids) to quite large (e.g., some nektonic hylids). The tadpoles produce mucus; chapter 5 has more details of integexceptionay sma eyes of centrolenids are located quite umentary histology. Small serous glands often are scattered close to the midline, and because of their orientation and an generally over the body surface (e.g., Asca.phus and Poyntonia apparent late closure of the choroid fissure, they often ap- paludzcoh) or concentrated along the dorsal fin, w i t h the pear C-shaped in dorsal view. The extent that the eye pro- ventral fin, along the dorsum of tad muscle, and dorsolatertrudes frorn the surface of the head, the curvature of the cor- ally and ventrolateray on the body (e.g., Inger 1966; Liem nea, and the lenticular protrusion also vary. We detect some 1961; S. J. Richards 1992; some Amolops, Huza, and Meriscorrelations with ecology, but definitive data to support togenys, Hyla, Hyhrana, Litoria, Rana, Phyllomedusa, and these observations are lacking. Physahmus). Tadpoles of someAmolops have keratinized spiVan Dijk (1966) discussed an umbraculum and two types nules on the dorsum that may abate turbulence as water flows over the body; they also have various patterns of small, densely arranged papiilae and iight keratinization in the roof of the beiiy sucker. Short papiilae occur on the snout of the tadpoles of Hoplophryne rogemz (Noble 1929). Close exarnination of certain tadvoles reveals a circular area of siightly contrasting color anterolateral to the base of the vent tube. This slightly raised area appears glandular, but it mav be associated with the lateral line svstem. It is most J obvious in darkly pigmented, neotropical, lotic hylids. Bright coloration and distinaive patterns are important sexual and species recognition features for many adult vertebrates but apparently are of little importance to larval or imFig. 3.2. Dorsal views of styiized tadpole bodes showing eye positions. (A) Lateral. (B) Dorsal. mature ones. Dorsal coloration in tadpoles is mostly of about on the forest floor. Definitive data are lachg, but it appears that finger and toe webbing of metamorphs is equal to or less than that of adults. Inhibitory versu stimulatory modulation of iimb growth by hormones of the pituitary and adrenal hormones is stage dependent (M. L. Wright et al. 1994; also Elinson 1994). L. Muntz (1975) recognized four behavioral stages in leg mobiiity and muscle and nerve development during leg myogenesis (also Dunlap 1967; Hughes and Prestige 1967; L. Muntz et ai. 1989). Comparisons of her data on Xenopus to those from species of Merent metamorphic patterns and ecology would be informative. &o, the ontogeny of pigment patterns during limb development, although largely unstudied, may be mefi in species identifications. u
i"

30

RONALD ALTIG AND ROY W. McDIARMID

muted tones of black, brown, gray, and green arranged in various patterns; ventraily most tadpoles are coppery, silver, or white. Altig and Channing (1993) described the major components of tadpole color pattern; compared to adults the coloration of tadpoles is somber and presumably functions in some aspect of crypsis (i.e., camouflage). Altig and Channing (1993) also identified common patterns that they associated with background matching, disruption, and countershadmg. Other colors (e.g., gold, orange, and red), when present, usudy are comparatively bright or contrasty. After considering the coloration and the behavior of the tadpole ofXenohyla, Izecksohn (1997) considered it to be a leaf mirnic, and the coloration is also surprisingly convergent with young tadpoles of Pseudisparadoxa. Specific pigments sequestered in chromatophores provide color and pattern (Bagnara et al. 1978). Melanophores contain black or brown melanins derived from tyrosine; xanthophores and erythrophores contain synthesized yellow and red pteridines plus dietary carotenoids; and iridophores contain refleaive purine platelets. Chromatophore type can be identified in TEM sections by the structure of the pigment-containing organelles (S. K. Frost and Robinson 1984). Filiform, foliose, punctate, rodular, and stellate are terms that are useful for describing the shapes of chromatophores, especidy of melanophores, that remain in formalinpreserved material. The presence and apparent shape and sue of chromatophores differ when viewed with incident, transmitted, dark-field, or polarized iliurnination. The type of pigment and its distribution within the ceil and the density, location, orientation, shape, and size of the chromatophores m o w the perceived color and pattern. Much of the color and pattern of tadpoles commonly is derived from chromatophores positioned in subintegumentary areas (e.g., w d s of the giii chambers, peritoneum, blood vessels, nerves, and gut). Pubiished figures often show these drfferences without mention in the text. Other components of coloration probably serve to delay (e.g., yeiiow oceh at the base of the tail in Rana alticola) or misdirect (e.g., black tail tips of someAcris tadpoles) a predator attack. Caldwell (1982) published one of the few quantitative studies on the distribution and occurrence of color pattern in tadpoles. She argued that a polymorphic pattern involving black tail-tips in A& is maintained by disruptive seleaion and functions as a defleaion mechanism to misdirect predatory attacks by certain dragonfly naiads from the head of the tadpole to the tail. Integumentary glands that presumably secrete noxious substances occur in some tadpoles (e.g., Liem 1961, Rana chalconota; persond observation, Physahmuspetersi) and their location on the body or tail often is highiighted by bright contrasting coloration (see photographs of Rana altimla and an Indian ranid in figs. 1 H and I in Altig and Channing 1993). Such patterns of aposematic coloration are well known among other organisms and their scattered occurrence among tadpoles is not surprising. Brodie and Formanowicz (1987) argued that the intense black coloration and cutaneous toxicity of many bufonid tadpoles and their tendency to aggregate in slow moving schools are traits indicative of aposematic coloration (see Wassersug 1971, 1973; see chap. 9). The bright red color-

ation of many centrolenid tadpoles is attributable to blood capdiaries and not pigment in the skm (Vilia and Vailerio 1982; personal observation). There also is some evidence that the bright golden band on s m d tadpoles of Rana heckrcheri may serve as a visual cue to promote spatial orientation of schoohg individuals. Altig and Channing (1993) reviewed generalizations about the associations of color and pattern with habitat, behavior, and ontogeny. Metachrosis (i.e., change in color via changes in the distribution of pigments within chromatophores) is iirnited, although considerable blanching and darkening relative to diel light cycles occur. The body of many tadpoles becomes paler at night, whde the tail, particularly the distal third, often gets darker. The tail of the tadpoles of Hylagratwsa has punctate melanophores near the base and steilate ones distally; it is the stellate population of melanophores that dilates at night that produces the intensely black tail of this species. Detailed studies of the distribution and morphology of melanophores are needed to serve as an aid to identification and to help explain certain aspects of behavioral ecology. For example, a chainlike pattern of melanophores visible only at magnification occurs in discoglossid tadpoles (Bytinski-Salzand Eiias 1938). Also, the tadpoles of some species of Leptodactylus have melanophores arranged in rodular bundles that are oriented at various angles to each other. As tadpoles approach metamorphosis, adult pigments begin to appear. Tadpole coloration is known to vary ecologicdy and geographically but detailed studies are lacking. Conspecifics inhabiting turbid water often appear unpigmented whereas those in clear or tannin-stained water even a few meters away often are quite diflerentiy and frequentiy prorninentiy colored. The reddish or yellowish colors that occur in the h s of tadpoles of certain species (e.g., Hyla chlysoscelis, H. cinevea, H. femoralis, H. versicoh, and a s s i n a senegalensis) develop most intensely in individuals that develop in tanninstained waters and seldom in those from turbid water. Experiments by McCollum (1993) and McCollum and Van Buskirk (1996) seemed to indicate that the reddish or yellowish pigments developed only in the presence of predatory odonate naiads. Understandmg details of how color pattern develops during hind limb ontogeny would be usem in distinguishmg arnong species and probably provide interesting data on pattern development in earlier stages as well. For example, hatchhgs of a number of North American hylids have dorsaily banded tail muscles. This pattern subsequently is lost in most species but usually retained by tadpoles of species of A&, Hyla at?ivocu, and Pseua'mi-is crucifer. A similar case occurs in young salamanders of various species ofAmbyrtoma with only A. talpoideum retaining the bands throughout larval ontogeny. Unfortunately, such studies are uncommon (e.g., Altig 1972). Ontogenetic changes in tadpole coloration usudy are not profound after about stage 25. Tadpoles of Hylagratwsa between stages 25-28 have a single black saddle on the tail muscle that disappears later in ontogeny when tail metachrosis appears (see above). Smail tadpoles of Rana vaillianti have bands across the body and tail, and smail tadpoles of Pseudis are prominently banded in contrast to the unicolored, larger ones. The black rim on the tail fins of

BODY PLAN

31

Rana heckscheri tadpoles begins to develop at about the same stage (28) that the prominent golden band of smaller tadpoles begins to become less prominent; whether there is an ontogenetic change in the orientation of individuals in schools is unknown. The oifactory sacs and both external and internal nares usuaiiy are present during ontogeny, but the external nares of most microhylid tadpoles do not open until metamorphosis; Metaphrynellu (Beny 1972), Ramanella (Kirtisinghe 1958), and Uperodon (Mohanty-Hejmadi et ai. 1979) are reported to have open nares during tadpole ontogeny. Berry (1972) also stated that the nares of Ranaglunduhsa do not open until just before metamorphosis. Embryos of Gastroplnyne mlznensis and Micvobylu heymonsi have external narial openings (personal observation), but the canal is blocked; t h s external opening subsequently closes at the surface for most of larval ontogeny, and, for a while, one can detect where the opening was on SEM photomicrographs. The narial openings presumably form via induction from the oifactory ectoderm (Zwilling 1940), but the mechanism for closure of the external opening and subsequent reopening late in ontogeny is not known. The external nares of the nidicolous cophyhe microhyiids are open throughout ontogeny, although it is not known if they are blocked within the tube during at least early stages. External nares are situated at various points on the snout. Tadpoles with cylindrical bodies and long snouts (e.g., Hylu leucophylluta group) have the nares much closer to the tip of the snout and more lateral than seen in most tadpoles; the nares of Xenupus are distinctly parasagittal and near the mouth (see iustrations in chap. 12). Most often the narial aperture is rounded, but it may be oval, roughly trianguiar (e.g., Bufo), or transversely ehptical (e.g.,Xenupus). Interspechc differences in snout shape and surface topography cause the aperture to face anterolateraiiy, dorsaiiy, dorsolateraiiy, or lateraiiy. The nares of Bufo are quite large for the size of the tadpole, and comparisons of relative size among species often can be used for identification (e.g., Ranapipiens vs. R. catesbeiana groups). Other detads of narial structure have not been exarnined in detaii (e.g., Channing et al. 1988; G. F. Johnston and Altig 1986). For example, the narial opening may be flush with the surrounding surface, countersunk, or marked with a marginal rim that varies interspecificaiiy in prominence and shape (fig. 3.3F). Other types of narial ornamentation range from a single, medial papilla to various arrangements of several papiae. Some descriptions suggest that the authors have looked through the transparent skin and mistaken pigmentation around the nasal canal for a tubular external naris. The narial rims ofRrcaphus and certain other suctorial forrns extend as distinct tubes (e.g., fig. 3.3F). Operculum and Spiracle "Operculumyy the giii covering of tadpoles and a term that is has different meanings in other groups; it is not homologous with the gii covering of fishes or the ear element in frogs (chap. 4). "Spiracle," the exit for respiratory water in tadpoles, forms as a result of the growth and fusion pattern of

the operculum with the body waii; it is another borrowed term and not homologous with the hyoid pouch derivative in elasmobranchs. We retain both terms because thev have been used for a long time for tadpoles and chang& the names Iikely would produce more confusion than that caused by the multiple meanings that exist in dderent groups today. The number, location, and morphological configuration of the spiracle(s) depend on where and how the o p e r d a r fold ses with the bodv waii (Starrett 1973). Starrett observed a frequent association of spiracular position with feeding mode and n shape, especiaiiy of midwater and surface feeding tadpoles. She commented that an excurrent spiracular flow from a single sinistral spiracle would tend to rotate and thereby decrease the efficiency of a tadpole feeding. in the water column and observed that most midwater L7 iiter-feeding tadpoles have either dual lateral or a single midventral spiracle. The position and number of the spiracle(s) generaiiy are consistent enough among related species and within certain hizher taxa to be useN for svecies identifica" tion and as phylogenetic traits. Exceptions include a midventral spiracle in FhctonutusfEssilis, rather than the sinistral condition typical of hylid tadpoles and a spiracle that opens on the left side of the taii muscle into the same tube as the vent in Stereocyclops, rather than midventraiiy as in most microhyiid tadpoles (I. Grifliths and De Carvaiho 1965). . By f i t h e most common spiracle morphology is a single midiateral opening on the left side (= sinistral or laevogyrinid) but at various positions along the body. At least phyllomedusine hylids, pelobatids, and some myobatrachid and leptodactylid tadpoles have a single spiracle situated far below the midiateral area in a parasagittal position. G. F. Johnston and Altig (1986) coined the term paragyrinid for this condition. A single spiracle situated anywhere from the chest (e.g., Rrcaphus) to near the vent (most microhylids) is termed midventral, medioventral, or mediogyrinid. Dual, lateral spiracles (= amphigyrinid) occur in &e tadpoles of rhinophrynids, pipids, and species of Leprdobatrachus (Leptodactyiidae),although the mode of formation and therefore homology in the latter case (Ruibal and Thomas 1988) is different from the first two. Laviila and Langone (1991) described the ontogenetic changes in the spiracle of Elachistochis ovalis, and Inger and Frogner (1979) cautioned workers about using spiradar characteristics of smaii microhyiid larvae because of ontogenetic changes. For example, the edge of the ventral waii of the spiracular tube of Paraduxophyla palmata is entire, but older individuals have four streamers extending posteriorly (personal observation). The unique spiracle of Otoplnynepybumi (l'yburn 1980) includes a free tube that is the longest spiradar tube known (i.e., extends wei past the body) and a seemingly sinistral position unique withn microhylids. Wassersug and l'yburn (1987) showed that the spiracle in Otophryne starts out in the expected medioventral position and ontogeneticay moves to the left side. These authors suggested that the tadpole feeds passively whde embedded in the bottom of sandy streams. ~ a t e flowing over the orifice of the long spiracle traiiing r upwardiy in the water acts as an aspirator to p d water into the mouth. Other tadpoles with rather long spiracular tubes

RONALD ALTIG AND ROY W. McDIARMID

Fig. 3.3. Representative anatomical and morphological features of select tadpoles. Typical arrangement of the recrus abdominis musde (origin uppermost in all frames, viewed with polarized light) in (A) Gastivphye caroliincnris (Microhylidae), (B)Agalycbnk callidym (Hylidae), and (C) L c p i k b d u ( hmk (Leptodactylidae).(D) Adhesive gland of Scupbbpu holhkii (Pelobatidae). (E) Abdominal flap (dorsal

view, transmitted light) of T h m o p a ~ l i t a n (Leptodactylidae).(F) a Narial aperture of HercOphyepumIli (Helwphrynidae). (G) Head flap of b c S c a m (Bufonidae). (H) Belly ''finger" (right side, ventral view, head at top) of Hopluphye r o (Microhylidae). ( I ) ~ ~ Belly sucker (anterior to right, longitudinal section) ofAtelopw &Mum (Bufonidae).

BODY PLAN

33

genetic link exists between the variations in spiracular tube length and middorsal stripe.
Vent Tube An anus is the posterior opening of the mammalian digestive tract. Amphbians have a common collecting chamber -a cloaca-into whch the dgestive, urinary, and reproductive tracts empty. The aperture through which feces exit to the outside is not the anus in the adult and probably not in the tadpole either. Without knowing more details of the development and origin, the use of proctodeum in association with the vent is also debatable. For tadpoles, we suggest the terms vent and vent tube in deference ;o the samelotation (i.e., snout-vent length) used for frogs. See previous dscussions of why body and tail length measurements should not involve the vent or vent tube as a landmark. The digestive tract of a typical tadpole exits the abdominal cavity through a tube, the aperture of whch usually is located in the sagittal plane and commonly associated with the ventral fin (fig. 3.5). The location of the aperture usually is consistent enough within taxa to be usefid for identhcation. The primary character states, recognized more than

Fig. 3.4. Basic patterns of spiracular tube arrangement in anuran lar(Dendrobatidae). (B) vae. (A) Single, sinistral, Dendvobates tiwononus Single, sinistral with long spiracular tube, Otophynepybumi (Microhylidae). ( C )Dual, lateral, Lepidabatruchw LlanennS (Jiptodactylidae). (D) Dual, lateroventral, Rbinophynus dorsalis (Rhinophrynidae).(E) Single, posterior ventral, KahlapuLhva (Microhylidae). (F) Single, midventral (on chest),Acaphw tmei (Ascaphidae).

includeA r s and Heleophvyne. Scaphzophvyne (Scaphiophrynicr nae) has many odd features (e.g., jaw sheaths) for a microhylid tadpole, including a spiracle that is between midventral and sinistral. Representative spiracle arrangements are shown in figure 3.4. Because of differences in the pattern of fusion between the opercular fold and body wall, the spiracular aperture may face various drections and have different shapes and tube codgurations. The different developmental patterns that produce dual, lateral spiracles, a single medial spiracle at various sites, or a sinistral spiracle are understood in general, but details of which tissues are involved and the Datterns of formation among taxa are unclear (Inger 1967; Nieuwkoop and Faber 1967; 0.M. Sokol1975; Starrett 1973). The inner or centripetal wall of a sinistral, paragyrinid, or midventral spiracle may be absent (e.g., Rrcaphm; phyllomedusine hylids), present only as a partial or complete ridge, or present as the wall of a tube of various lengths that posteriorly or terminally is free from the body wall with the posterior tip free from the body wall (fig. 3.5A-D). The lateral wall may end anterior to, at the same plane as, or posterior to the insertion of the medal wall. The spatial limits of the spiracular walls determine the shape and orientation of the aperture. In certain cases the aperture faces laterally and is commonly smaller than the bore of the s~iracular tube. Functional explanations to account for the Merent spir a d a r arrangements are few. Starrett (1973) suggested that midventral (e.g., microhylid) or dual lateral (e.g., pipid) spiracles in suspension-feeding, nektonic species may allow the tadpole to avoid disturbance of planktonic or particulate food items. At least some of these forms can maneuver solely by the force of the spiracular jet (Starrett 1973; personal obm a t i o n , several taxa). Wassersug and Viertel (1993) sugp t e d that folds in the inner walls of the long spiracle of OtopiCnyne may provide elasticity t o modulate water flow. Based on tadpoles raised from a single clutch of Stereoyclups eggs,I. Griffiths and De Carvalho (1965) suggested that a

Fig. 3.5. Schematic drawings of major configurations of spiracle and vent tubes. (Uppw)kft-lateral views of spiracular tubes that essentially face posteriorly: (A) inner (= cenmpetal) wall absent; (B) inner wall present as slight ridge; ( C )inner wall free from body; and (D) inner wall free and formed such that aperture opens laterally instead of posteriorly. (Mtddk)Lateral views of the right side of supine tadpoles showing vent tubes (arrowsindicate points of attachment of right wall): (E) right wall displaced dorsally; (F) right wall displaced anteriorly; (G) right wall displaced dorsally and anteriorly; (H) medial vent tube. ( h e r ) Schematic cross-sectionsthrough the base of the d fn of i supine tadpoles showing various placements of the vent tube: (I)right wall displaced dorsally, as in (E) above; (J) medial vent tube with lateral displacement; (K) medial vent tube with web between tube and fin; and (L) meha1 vent tube with both walls attached directly to ventral fin.

34

RONALD ALTIG AND ROY W. McDIARMID

100 years ago (see Camerano 1890, 1892; Willey 1893), involve the position and orientation of the opening; apertures open medially (in line with the plane of the ventral fin) or dextrally (to the right of the plane of the ventral fin). Some authors (e.g., Cei 1980; Duellman and H-illis 1987; Kehr and Basso 1990) reported sinistral vent tubes in Pleuro&ma, Gastrotheca, and Lysapsus, respectively, and we have seen aperture variation within a single lot of Ceratuphlys cornuta. Numerous other variations w i b each basic type are known (G. F. Johnston and Altig 1986), but no significance has been attached to the different configurations. We recommend that the character state (medial, dextral, sinistral) of the vent be determined relative to the position of the aperture, not the position of the tube. A vent tube that parallels the ventral margin of the fin may still have a dextral or sinistral aperture. We know of no data supporting the notion that the vent tube is abdominal in origin but we assume that llkely it is. All vents are medal duri& early development, and the ontogeny of vent tube formation has not been studed well. Those phyllomedusine hylids in which the entire vent tube and ape&re is displaced to the right and not associated with the tail fin in any way have not been examined as embryos. The dextral condition in Phyllobates lugubris (Domelly et d . 1990b) is not attained until later than normally expected (i.e., by stage 25; also see Lavilla and Langone 1991). Davies and Richards (1990) noted that until stage 32 the limb buds of Nyctimystes h y i are enclosed in a membranous sac associated with the vent flap (also see Van Dijk 1966; Hemisw). Variations on these themes are found among a number of rheophilous forms (see below). A medial vent may or may not be associated with the ventral fin in several different ways (fig. 3.5). The tube may be long or short, free or attached directly to the f n or attached i, via an intervening, fleshy web. The ventral wall may be shorter than, equal to, or longer than the dorsal wall. Differences in wall lengths produce distinctly shaped apertures that usually face in different drections. The right wall of dextral vent often is displaced anteriorly, dorsally, or more commonlv anteriorlv and dorsallv. A ventral wall.that is longer than the other w d s causes the aperture to face in a number of drections, includmg dorsally. In some cases, dextral vents are com~letelv l se~arated from the tail fin and musl i d e (e.g., Phasrnahylaguttata). In a number of suctorial forms (e.g., Ascapbus, Heleuphly, H y h bogotensis group, Telmatobufo) a flap of tissue lies below the vent; these vent flaps take In on several dfferent sha~es. Asca~hw.the vent tube is attached to the dorsal surface of this kapYJand vent aperthe ture is even with the dstal margin of the flap. The margin of the aperture normally is smooth, but B. T. Clarke (1983) noted that in tadpoles of Nannuphlys ceylonensis it was fluted and appeared similar to marginal papillae on the oral disc. We know of no data that indicate a correlation between vent tube morphology and position and tadpole ecology. Transitory Embryonic Structures Adhesive glands (fig. 3.3D), external gills, integumentary ciliated cells (visible in fig. 3. lo), and hatching glands occur

for various periods during embryology. The adhesive gland usually is visible between stages 18-25; it can be detected much earlier with proper techniques and may persist at least as a pigmented spot for several stages later in ontogeny. The term "sucker" should not be used in reference to the adhesive gland. The mode of attachment is via sticky secretions; there is no suction comparable to that found in the belly or abdominal sucker o f gastromyzophorous tadpoles and the oral d s c of suctorial forms. Hatchlings use the sticky secretions (Eakin 1963, Pseudacris regilla) of these transient glands to provide stabhzation prior to the further development of the oral d s c and tail whch afford more coordinated adhesion and locomotion for active tadpoles. Znopzw and perhaps other hatchlings may hang from a mucous strand from this gland immediately after hatchmg (Bles 1905; also see Sive and Bradley 1996). Adhesive glands inxinupus l& first appear at about Gosner 18 (Nieuwkoop/Faber 15, early neurula), reach fl development at stages 22-23 (35-36; ul see Drysdale and E h s o n 1993), functionally degenerate by Gosner 25 (49), but persist as pigmented remnants posterolateral to the oral d s c as late as Gosner 27-28. Lieberkund (1937) and Thiele (1888) surveyed the gross and histological structure of these glands in Burope& tadpoles. External gills (see chap. 5) are present in most tadpoles for a short ~ e r i o d (Gosner 19-24'1 in earlv develo~ment. Ex, ternal gills appear as short nodules or long, branched fimbriae and their morphology may reflect the developmental environment. Lsvtrup and Pig6n (1968) discussed the influence of 0, and CO, concentrations on their develo~ment. and there is some evidence of similar gill morphology among related species. Gdls usually are on two visceral arches. Tadpoles of Boophis, Manti+lus (BlommersSchlosser 1979a) and a number of microhylids lack external f l s . Ciliated cells scattered throughout the epidermis of anuran embryos (Steinman 1968; see fig. 3.10) may have the following functions: removing micro6rganisms i d impurities from the epidermis, transporting newly hatched embryos, providing respiratory ventilation of the body surface wide & the egg as well as .after hatching, and redicing the drag coefficient (Nachtigall 1982). Embryos move frequently within the egg jellies, and hatchlings move at surprising speeds via these cilia while lying on their side. Ciliated epidermal cells produce a cephalocaudal current along the embryo in the egg and after hatchmg (Kessel et al. 1974). The hatching gland is a Y-shaped array of unicellular glands on the 6 p o f the head; the open end of the series faces anteriorly (Drysdale and Elinson 1993; Ohzu et al. 1987). Presumptive cells of the gland are present by the end of gastrulation, and differentiated cells start to form after n e d a t i o n is completed. These glands produce an enzyme that dissolves or weakens the egg jehes and facilitates hatching (Carroll and Hedrick 1974; K. W. Cooper 1936). Yoshzaki (1973, 1974, 1975, 1991), Yoshzaki and Katagiri ( 1975),, and Yoshizaki and Yamamoto ( 1979) discussed the , ontogeny of these glands and their enzymes (also see Gollman and Gollman 1993a; Noble 1926a).
I I
A

BODY PLAN

35

describing: relative locations of structures on the face of the oral dlsc, we prefer "anterior/posteriorn relative to the longiTemzimlogy t u h a l axis of the body and "proximal/dlstal" relative to a Terminology used to describe certain features of the oral ap- landmark (usually the mouth or jaw sheaths) and reserve abparatus (fig. 3.6) is confusing and sometimes inappropriate. oral (= opposite) and adoral (= same) for denoting the back The terms "mouth" and "oral" have multiple usages, and and face of the oral disc, respectively. Orientation of the oral clarification is required to avoid confusion and apparent disc relative to the longitudinal axis of the tadpole varies contradictions. Anatomically, "mouth" refers specifically to among tadpoles and is-correlated with habitat- and feedthe opening that forms by the invagination of the stomo- ing mechanics; unfortunately, this variability complicates dedeum, anterior growth of the archenteron, and eventual rup- tailed descriptions. For example, the oral dlsc is positioned ture of the oropharyngeal membrane (Cusimano-Carollo (see fig. 12.1 and other examples in chap. 12) ventrally et al. 1962; Watanabe et al. 1984). The mouth provides ac- (about 180") in suctorial forms, anteroventrally (about 40") cess from the outside to the buccal or buccopharyngeal cav- in many tadpoles, terminally (90") in some carnivores and ity. Use of "mouth" to refer to the oral dlsc is incorrect, and some forms with reduced oral discs. and uvturned or umits use to refer to the buccal cavity is colloquial. We use belliform (more than 90") in certain surface feeders. As a "oral" as a general term to refer to the oral disc and its associ- result, the leading edge of the anterior labium may be anteated keratinized and soft structures that are situated adjacent rior (oral disc ventral), dorsal (oral dlsc terminal), or posteand external to the mouth. Collective terms for these struc- rior (oral disc umbelliform); and the trailing edge of the tures are oral apparatus and mouthparts. Restricting usage lower labium would be posterior, ventral, and anterior, reto these definitions means that papdlae on the oral dlsc exter- spectively. In addition, a tadpole can attain many positions nal to the mouth have incorrectly been called buccal (Cei in the water. Understanding the orientation of the oral disc " 1980) or peribuccal (Lamotte and Lescure 1989) papillae. is an essential component of what we consider to be a satisL~kewise, papillae on the roof and floor of the buccal cavity factory tadpole description and essential to interpreting data confusingly have been called oral papdlae (Wassersug 1980). from other studies (e.g., feeding studies). To facilitate comIn the latter case we recommend that "buccal" or "buccopha- parisons and refer to the location of structures of the oral ryngeal" are more appropriate terms to refer to those pa- dlsc without appearing contradictory, and certainly confuspillae and other structures on the roof and floor of the ing, we have adopted an unspoken convention of &dung buccopharynx. of the oral disc as if it were a published figure in face view The infrequently used terms "aboral" (opposite, away, or with anterior (= dorsal, upper) at the top and posterior (= distal from the mouth) and "adoral" (same, toward, or prox- ventral, lower) at the bottom, regardless of the actual locaimal to the mouth) can be interpreted in several ways. For tion of the dlsc on the tadpole or of the position of the tadpole in the water. We strongly recommend that the term "denticle" be forever struck from the tadpole lexicon because of confusion "2 A-2 GAP resulting from its multiple usages for different structures of tadpoles and other organisms as well: labial teeth (e.g., Inger 1983), cusps on labial teeth (e.g., Faier 1972), serrations on jaw sheaths (e.g., Ruibal and Thomas 1988; M. H. Wake 1993a), and other small pointed structures. The preferred term is labial teeth and their structure and form'ation are much more complicated than the simple structure implied by "denticle" (e.g., Altig 1970; Boulenger 1892). By extension. the serrations on the head of an individual labial tooth are termed "cusps." Because a consistent terminology decreases the likelihood of confusion, and when coupled with long-term usage provides stabhty and understanding, a unique terminology (e.g., Van Dijk 1966) seems unwar'P-3 ranted or is premature pending better understandmg of the OD EMARGINATE NOT EMARGINATE distribution of morphological traits and their homology. The jaws cartilages of tadpoles (chap. 4) are without basal Fig.3.6. Oral apparatus of a tadpole showing emarginate (&side) articulation and significant adult derivatives and thereby and not emarginate (lightside) conditions of an oral disc. Abbreviadiffer from those of all other vertebrates. The sheaths that nons for descriptive terminology: AL = anterior (upper) labium; A-l overlie the cartilaginous jaws of tadpoles are transitory and and A-2 = first and second anterior (upper) tooth rows; A-2 GAP = have a columnar histological structure dtfferent from that of medial gap in second anterior tooth row; LJ = lower jaw sheath; similarly keratinized structures in birds, turtles, and some LP = lateral process of upper jaw sheath; M = mouth; M P = marsalamander larvae. In contrast. the beaks of other vertebrates ginal papilla; OD = oral disc; PL = posterior (lower) labium; P-l, (e.g., turtles and birds) are a&t structures that overlie basP-2, and P-3= first, second, and third posterior (lower) tooth rows; SJI = submarginal papilla; and UJ = upper jaw sheath. ally articulating jaws comprised of dermal bone. Because the

Morphology and Ontogeny: Oral Apparatus

36

RONALD ALTIG AND ROY W. McDIARMID

sheaths that cover the jaws of tadpoles and the keratinized shape, as seen in tadpoles ofMegophvys. In tadpoles of the coverings (= rhamphotheca) of the jaws of other animals hylid Scarthyla ( D u e h a n and De Sa 1988) and members of surely are not homologs, we believe they warrant a distinct the Scinm ccowata group (e.g., McDiarmid and Altig 1990), term and recommend "jaw sheath." The so called jaw sheaths the exceptionally short P-3 lies at the end of an armlike modfound in certain salamander larvae (e.g., Ambystoma and Si- ification of the medal part of the posterior labium that rests ren) have not been examined hstologically, but they appear in a gap in the ventral papillae. The lower labium commonly to differ structurally from either of the previous two cases; is larger than the upper labium and more of its area is free although they overlie dermal bones, the specific bones in- from the body wall. Tadpoles of species with many tooth volved are not entirely the same in larval salamanders as in rows have larger oral discs than those with fewer rows. The other taxa. Also, the jaw sheaths of anuran larvae typically lower labium particularly may extend a considerabledstance are not pointed and do not project, as is implied by the usual posterior to the last tooth row in some tadpoles (e.g., some connotation of beak. Finally, the term "beak" has been used Ansonia, some Litoria, and Stauois). The labia consist of to denote the prominent medal convex projection of the loose connective tissue covered with an epidermis of varied margin of the upper jaw sheath of certain carnivorous tad- thtckness. Although the disc is more rigid in lotic than in poles; this usage adds addtional confusion, and we recom- lentic forms, the structural modifications that contribute to mend incised as more appropriate descriptive terms for this these differences in rigidity have not been determined. Aljaw sheath morphology. though the cores of papdlae and bases of tooth ridges someThe conventional use of length to refer to the longer and times are pigmented, the face of the d s c usually lacks pig-, usually longitudinal dimensions of a structure poses a con- ment, and mucous glands are absent. An inverted U-shaped flict when describing jaw sheaths because their greatest di- depression or transverse slit commonly, perhaps always, ocmension is transverse. For convenience we retain length for curs within the medial gap of the anterior tooth row adjacent the transverse measurement. This dunension is usually mea- to the upper jaw sheath. Neither the function nor developsured or described without considering curvature because it ment of this structure has been studied, and its occurrence is the only measure that can be obtained without dissection. in those species with a complete proximal, upper row is Jaw length as appropriate for a mammal (= gape, the basal not known. Most commonly the oral disc has all but its anterior marterminus of the jaw sheath in a straight line to the peak of the front curvature) has never been suggested as a measure- gin free from the body wall. In tadpoles with a dorsal gap in ment for a tadpole, again perhaps because it could not be the marginal papillae, the upper labium essentially is a conmeasured on an intact specimen and because the jaw carti- tinuation of the snout; in species with complete marginal lages do not articulate basally. By retaining length for the papdlae (e.g., suctorial forms) the entire margin is free. In longest but transverse dunension and width for the shorter, some rheophilous Litoria from the Australopapuan region, longitudinal dimension, jaw sheaths may be described and the snout and anterior margin of the oral disc are separated measured as follows: long/short (transverse measurement), by a crevice that appears to be covered by a thin membrane widelnarrow (basal boundary of keratinization to serrated (personal observation). The function of this morphology is edge), and thicwthin (front, oral or adoral to back, buccal unknown but surely it signals a novel structure in streamor aboral surface). Thickness has not been examined in dwelling pelodryadine tadpoles as compared to streamdwelling tadpoles of neotropical hylids. The back of the free many tadpoles. The jaw sheath is only the keratinized part of the jaw portion of the oral disc, which is primarily in the lower lacovering and is usually pigmented. Descriptions should be bium, is typically smooth. In tadpoles of Ceratupblys, this based only on the sheath and should not include jaw and surface is often fluted. associated soft tissues. In those rare cases where the sheath An umbelliform or upturned oral disc appears as a conis not pigmented (e.g., Scaphiophlyne), determining the lim- vergent trait in tadpoles of some arthroleptids, dendrobatits of the sheath is diicult. If the excised jaws of tadpoles ids, hylids, mantelline rhacophorids, megophryids, and miare placed in water, the jaw sheaths will detach (soon if speci- crohyline microhylids (fig. 9.5). Most umbelliform tadpoles men not preserved, several days for preserved specimens) occur in the backwaters of lotic systems. Tooth rows and jaw from their underlying cartilages, thereby facilitating de- sheaths are reduced to absent in these forms, and large tailed observations. ridgelike papillae that project radially from the mouth are common. When present, marginal papillae are complete but Oral Disc and P a p i l h greatly reduced in size. Morphological comparisons among The oral disc of a typical tadpole is round to transversely umbelliform tadpoles have not been made, but cursory elliptical. The poorly delimited upper (anterior) and lower examinations show that both labia or primarily the lower (posterior) labia of a M y formed d s c in nonsuctorial tad- labium form most of the disc. The bi-triangular disc o f M g poles are delineated by a fold that during closure (as usually ophvys is the most bizarre; when the tadpole submerges, the occurs in preserved specimens) develops along a line that lateral, pointed flaps of the oral disc by their own elasticity approximates the angles of the jaw cartilages. The degree to fold medally; when the tadpole rises to the surface to feed, whch parts of the upper and lower labia contribute to the the flaps unfold to form a bi-triangular oral d s c with the entire disc determines the disc shape. For example, enlarge- longer dimension oriented transversely. Even though funcment of lateral portions of the disc produce a bi-triangular tions other than feeding have been suggested for an umbelli-

BODY PLAN

37

form disc (e.g., stabilization among vegetation, float, etc.), observations on certain forms suggest to us that feedmg on materials caught in the surface meniscus seems to be the most likely functional explanation for all umbelliform discs. The umbelliform disc tadpole of Mimhyla heymonsi has a large, infolded semicircular structure at each corner of the mouth that appears derived from the lower labium. The oral disc of some suctorial tad~oles does not fold shut, even in preserved specimens, and the discs of rheophllous forms commonly have lateral pleats (see emarginate below). These pleats presumably permit the seal between the disc and substrate to be maintained whde the disc margin " is expanding during the extension-retraction cycles of oral locomotion (C. L. Taylor and Altig 1995). Other folds and specimens, primarily flexures commonly seen in along the disc margin, are caused by differential contractions of muscles and different types of connective tissues and may not occur in living tadpoles. Emarginations, which are marginal indentations and not simply folds of a uniform margin, occur at various sites on the oral d s c (fig. 3.6). Lateral emarginations are the most common (e.g., most Bufi and Rana) but emarginations may occur dorsally, ventrally, and or ventrolaterally. The dsc often must be opened to distinguish an emargination from a pleat or fold. Margins of the labia usually are defined by papillae that vary in density (number/dstance), basal diameter, length, orientation, pigmentation, and shape (pointed vs. rounded tip). These marginal papillae occur in four basic patterns on the disc: complete (no gap), dorsal gap only (taxonomically and ecologically common), ventral gap only (rare), and dorsal and ventral gaps (most bufonids and a few hylids, mantellines, ranids, and rhacophorids). The condition of complete marginal papillae is most common in tadpoles that attach to rocky surfaces in fast-moving streams; it occurs uncommonly in a few nonrheophllous forms (e.g., Anotheca). The closely spaced, short, and numerous papillae presumably aid in forming a tight seal between the oral disc and the irregularities of the substrate. Van Dijk (1981) suggested that the ventral gap in the marginal papillae of bufonid tadpoles serves as a weirlike c o n t r o h g device during feeding. Other explanations to account for gaps in marginal papillae have not been offered, but a large phylogenetic component seems obvious. In some species, the size of the ventral gap is associated with the length of the adjacent tooth row, but notions that these patterns may be developmentally linked have not been pursued. We have noted obvious aualitative differences in the size (e.g., smaller in hylids than ;anids) and density of marginal papillae among tadpoles but have found few published notices (e.g., I. Griffiths and De Carvalho 1965) of this variathe tion. ~Cnerally marginal papillae are relatively similar in size around the oral disc, but some ventral papillae may be greatly elongated (e.g., in Hemisus, some Phlynobatrmhus, and some Rana). Elongate ventral papillae are rarely bifid (Charming 1978), and Van Dijk (1966) noted that the length of the long ventral papillae in Phlynobatrmhus may decr ase ontogenetically.Marginal papillae typically occur in

'%

one (uniserial) row but dscs bordered by two (biserial) or several (multiserial) rows are not uncommon. Sometimes the bases of ventral paplllae are slightly offset toward the back of the labium. The paplllae forming a single (uniserial) row are properly aligned, but those that occur in two or more rows have their bases slightly offset and appear double, or the bases are in one series but alternate papillae project in different directions (also emulating a doubl; row): ~ h e s e alternative patterns seemingly represent different solutions to the same problem and for clarity and understanding, they should be ex~lained descri~tions. in Submarginal papillae, which occur on the face of the disc and sometimes grade centripetally from the marginal papillae, vary in density, distribution, shape, and size. The bases of submarginal ~ a ~ i l l usuallv are Sircular but sometimes ae " (e.g., Njctimystes ahyi, Phasmahylaguttata, and several umbelliform types) are elongate in a longitudinal or radial direction; S. J. Richards (1992) suggested that these types of papillae may be secretory. A patch of submarginal papillae commonly occurs near the lateral ends of the lower tooth rows, particularly in lotic tadpoles, and single rows of widely spaced, large papillae may occur dstal to the first upper and last lower tooth rows in other forms. Some species of Cardi&ma and Hyalinobatrmhium euqv~natha have a row of paiillae that crosses the ~roxirnal of the lower labium &d face is continuous with the lateral marginal papillae. In these situations, papillae lie in front of the recessed jaw sheaths. It is tempting to speculate that these papillae may be a vestige of a tooth ridge(s). The oral disc and its presumed derivatives are reduced in various ways in tadpoles in several different lineages (e.g., some species groups of neotropical hylids, microhylids in general, pipids, and a few ranids) and feeding modes (see fig. 12.2). Some tadpoles have few (e.g., Hyla mamzorata group) to no (e.g., leucuphylluta, miwocephala, and parpiceps groups of Hylu spp., Occiduzy~asp.) labial teeth but have normal or unusual (e.g., Otophrynepyburni) jaw sheaths. In these forms, the oral dsc is reduced to differing degrees, typically to a U-shaped yoke around the mouth, and the marginal papillae appear as slightly differentiated mounds or are absent. The iaw sheaths. which sometimes are massive for the size of the tadpole, often are deeply recessed. During feeding, the snout is deformed and the jaws protruded to bring the sheaths in contact with the substrate. This configuration presumably is what prompted Lavilla (1990; Hylu nana) to describe what he called an "oral tube" that was visible only during feeding. The report (I. Grifiths 1963) of minute keratinized "denticles" in the oral amaratus ofPseudhymenochirus needs verification. A few tadpoles other than microhylids, pipids, and rhinophry~llds both labial teeth and at least keratinized jaw lack sheaths. Tadpoles of Litoria subglandulosa (Tyler and Anstis 1975) have complete marginal papillae, abundant papillae over the face of the disc, and small, unpigmented jaw sheaths. Six large papillae occur in a transverse row above, and a flat, white structure lies between the papillae adjacent to the midline. The tip of this structure is divided into 4-7 toothhke structures, each bearing 1 4 black, hairlike filaI
I I

I I

38

R O N A L D A L T I G A N D R O Y W. M c D I A R M I D Fig. 3.7. Representative photomicrographs of labial teeth and jaw sheath of various tadpoles. (A) Multiserial tooth row (A-1) and biserial tooth rows (A-2 and A-3) inhcaphw w e t (Ascaphidae). (B) Tooth series in an intact tooth row of H y h chysoscelis (Hylidae). (C) Cross section (viewed with polarized light) of two posterior tooth rows (mouth a t bottom) of Hyhgratiasa (Hylidae) showing two tooth series (black structures), each surrounded by patches of connective tissue @ale areas) within the tooth ridge (P-2 and P-1) and with mitotic (toothforming) cells (dark areas) near their bases; slips of the mandibulolabialis muscle (pale structures) also visible (her&). (D) Labial teeth in nonnormal hillocks of tissue and on marginal papillae on oral disc of Scaphiopusholbrookii (Pelobatidae). (E) Lateral view of tooth series of about 15 teeth (series dissected from posterior tooth ridge, erupted tooth to h e r right) ofRrcaphtrs tmzi showing lengthy extensions of anterior base of each tooth sheath with associated connective tissue (white in polarized light). (F) Part of a posterior tooth row ofkcpphus twei (distalperspective, mouth to @ht) showing extensive connective tissue (white in polarized light) in tooth ridge. (G) P-2 and P-3 tooth rows of Hylagratwsa (Hylidae) showing (proximdperspective) fibers of the mandibulolabialis inferior muscle (white in polarized light) inserting on tooth ridges and marginal papillae lateral to tooth rows. (H) Serrations on jaw sheath (buccalsuYfke to left) of Ram catesbeianu (Ramdae). (I) Lateral view of five labial teeth (distal suYfke or bad to r2ht) from last posterior (multiserial) row ofhcaphw tmzi.

ments, some of which are branched. Below k s structure there are 2-6 papillae with lightly keratinized tips. Our unpublished data show some surprising similarities between the mouthparts of the tadpole of Mantida~tylussguttulatus ( m a n t e h e rhacophorid; jaw sheaths absent) and those of L. subglandulosa. kn unidentified Boophis has an immense oral d s c that lacks keratinized structures and consumes exclusively clean-looking pieces of sand. spehes without keratinized mouthparts often have different or unusual soft structures. The paired, semicircular labial flaps of many microhylids presumably are homologs of the upper labium of a typical oral dsc. These labial flaps, with or without papillate edges, are suspended in front of the mouth and separated by an inverted U-shaped medal notch (see Donnelly et al. 1990a for review). The flaps may be strongly semicircular, short and essentially without the medal notch (i.e., upper lip is transversely straight), excessively long and almost rectangular, or reduced to small protuberances around an extremely wide medial notch. A few microhylid tadpoles (e.g., NelsolMphlyne from the New World and some Mzcrohyla and Scaphtophlyne from the Old World) have other structures that may be derived from the upper, lower, or both labia (Donnelly et al. 1990a). The unconventional microhylid tadpoles of Nelsonophyne (microhyline) and ~cupbzophlyne(icaphiophrynine with nonpigmented jaw sheaths) have oral discs with papillae. Except for vestigial structures that presumably are derived from typical jaw sheaths, the macrophagous carnivorous tadpole of Lepidoba~rmhwhas an immense, slit-shaped mouth and no other mouthparts. The su~erficialsirnilaritv of the soft mouth~arts rhiof nophrynids and pipids diminishes upon closer examination. and Both Rhil~phlynus &nopus have slight folds at the corners of the mouth that may represent vestiges of an oral disc (Orton 1943; Thibaudeau and Altig 1988), but the barbels of the two tadpoles are not homologous (Cannatella and Trueb 1988a). The single long barbel at each corner of the mouth of Xihoiws has muscle-fibers attached to an internal cartilaginous support and is mobile. The shorter and more numerous barbels in tadpoles of RhilMphlynus lack these characteristics; they also seemingly vary in number and length ontogenetically and certainly between cohorts at one site and between sites. Tadpoles ofRhinophynw also have a at the symphisis of the lower single papilla-like jaw cartilages that probably is not a barbel.
L

Tooth Ridges and Labial Teeth Transversely arranged tooth ridges form on the face of the oral d s c during early ontogeny (Thibaudeau and Altig 1988: see Cherdantseva and Cherdantsev 1995). The number, length, position, shape (curved vs. straight), and spacing of these ridges on the disc vary interspecifically and ontogenetically. Darkly keratinized labial teeth that develop from mitotic sites in-the tooth ridge are arranged linearliin one or more rows along the ridge. The tooth ridges of pond tadpoles are tall and have narrow bases and wide interridge valleys that are situated about equidstant between ridges; the ridges are quite flexible, and teeth generally can be removed

easily. Rheophilous and particularly suctorial forms have shorter, flat-topped ridges with narrow interridge valleys that lie immediately behind the next proximal ridge. I n all cases, the ridges become shorter and more closely spaced in 1 the distal rows, and this is particularly pronounced 11species with many tooth rows. Judged by the dstribution of polarizing birefringence, the tooth ridge contains considerable connective tissue proximal and distal to a mitotic zone in the base of the tooth ridge that produces the labial teeth (figs. 3.7C, F). Although known for some time (e.g., Schulze 1892; Weber 1898), the extrinsic musculature of the oral d s c has not been well documented (e.g., Carr and Altig 1991; Gradwell 1968, 1972b, c; Noble 1929; Starrett 1973). Undoubtedly many unknown variations remain to be described, and several errors or omissions need to be clarified. Fibers of the m. mandbulolabialis (fig. 3.7G; see chap. 4) originate on the basal, ventrolateral surface of Meckel's cartilage. The m. mandibulolabiahs superior serves the upper labium, and the m.m. inferior serves the lower labium (Carr and Altig 1991; McDiarmid and Altig 1990). When present, fibers of the muscle insert at the bases of the lateral quarters of the valleys between tooth ridges and at the bases of marginal papillae lateral to the tooth rows. Fibers of the m. mandibulolabialis usually do not extend distal to the most distal tooth row in either labium, although they do so in some suctorial Litoria. Examination of 24 species (Carr and Altig 1991) demonstrated little ontogenetic but considerable interspecific variation in these muscles. Typically, slips of both muscles are present, but in some species only the m. m. inferior has been detected. Contractions of these muscles apparently change the shape particularly of the lateral portions of the oral d s c and of the tooth ridges. Distortion of the tooth ridges ~resumablvrotates the labial teeth dstallv and effects a better contact with the substrate. Although the sequence

BODY PLAN

39

40

RONALD ALTIG A N D ROY W. M c D I A R M I D

and timing of these muscle contractions during a feedmg cycle are not clear, it seems that the muscles relax soon after maximum oral dsc excursion and as closure begins (C. L. Taylor et al. 1996). The m. intermandibularis posterior (Gradwell 1972b) also may aid in tooth reorientation. Gradwell (1968) noted that the m. mandibulolabiahs is innervated by the trigeminal nerve. We suspect that lymphatic spaces in and near the oral disc (chap. 5) also may function in the operation of some of the mouthparts. The number and arrangement of tooth rows on the oral disc of tadpoles is species specific. The labial tooth row formula (LTRF) is a synoptic representation of this arrangement. A number of systems have been devised for numbering labial tooth rows and designating the rows with medal gaps. Any system is clear to an experienced user, but many systems are not obvious to a neophyte. To us, less cumbersome schemes seem to be more useful. Dubois (1995) and Dubois and Ohler (1994) present the most recent variations. Accordingly, we continue-to use the fractional designation (e.g., Altig 1970) to spec@ the number and gross morphology of tooth row. Thls system accommodates all row configurations easily, does not use Roman numerals, has been used for quite a long time and is familiar to many workers, and takes less space than formulas that are written in spatial order. Rows on the anterior labium are numbered distal (labial margin) to proximal (mouth). The notation "A-1" denotes the first anterior (most dstal from mouth) row, and designations extended sequentially through A-n to the row adiacent to the mouth. Rows on the vosterior labium are numbered proximal to distal. The first row adjacent to the mouth is P-1, and more distal rows are numbered sequentially through P-n. Notations such as "PR-1" and "AR-1," denoting the tooth ridge for tooth row P-1 and A-1, can be extended for any anterior or posterior row if in developmental studies one wishes to refer to tooth ridges dstinct from tooth rows. Rows with medial gaps are-designated with parentheses, and rows that vary between individuals (gap present or absent) are placed w i t h brackets. A gap in a tooth row is a physical break in the tooth ridge and therefore expressed in the tooth row; teeth occur only on tooth ridges. Even apparent atavistic teeth are positioned on hillocks of tooth ridge tissue (see below and fig. 3.7D). If the gap is quite narrow, as it often is in row P-1, staining (see chap. 2) may be necessary to discern if the ridge in fact is broken. The absence of teeth on part (or all) of a tooth ridge should be noted, but these breaks do not constitute gaps as defined here. Also, sharp bends in tooth rows caused by flexures of the d s c in preservation are sometimes interpreted as gaps; one should open the oral disc manually to ver@ that a gap probably is not present in most of these cases. The functional and developmental considerations of gaps in tooth rows have not been examined. These gaps may allow larger excursions of the jaw sheaths during feeding; adjacent tooth rows that have only narrow or no gaps perhaps signal smaller jaw excursions during feeding. We assume that most gaps are neither gained nor lost ontogenetically w i t h rows and that in most cases gaps can be used to - evaluate row homology.

In summary, a labial tooth row formula (LTRF) of 5(2-5)/3[1] indcates a tadpole with 5 upper tooth rows with medal gaps in rows A-2 through A-5 and 3 lower rows with or without a gap in P-1. An advantage is that this system apparently numbers homologous rows the same in different species and does so more often than other systems. In fact, no system will always number supposed homologs the same because of different patterns of ontogenetic and phylogenetic row alterations (Altig and Johnston 1989). Some tadpoles, particularly of species of ranids and pelobatids, have accessory tooth rows situated in the lateral areas of the oral dsc. The functional importance, developmental association, and homologous nature of these short rows compared to the more typical rows are not known. Two types of these accessory rows are distinguishable. In ranids, accessory rows essentially parallel the lateral ends of the posterior tooth rows; in pelobatids the rows tend to lie off the ends of and are oriented at nearly right angles to the posterior tooth rows. The formulation ofAltig (1970) was modified by R. G. Webb and Korky (1977) by placing the number of accessory rows between solid - 5 (2-5)/4/3 [l]. Other than notations concerning ontogenetic change, few data (e.g., Bragg and Bragg 1958) illustrate the natural variations in tooth formulas. Echeverria and Filipello (1994) found that 77.4% of the tadpoles Odontopbvnus occzdentalis had one formula and the remaining 22.6% had several other formulas. The most deformities (26.7%) were in A-1, while P-1 showed the least (6.7%). Grdhtsch and Grillitsch (1989) and C. L. Rowe et al. (l996,1998a, b) reported similar data on ranid tadpoles. Tadpoles of Phq~nohym,Tvachycephalus, and some species of the Hylageogvaphicu group share the uncommon feature of having A- 1 with a wide medan gap but with A-2 entire. In such cases, this row is probably not homologous with A- 1 (but A-2 is) of other tadpoles. Certain structures on the oral disc of tadpoles with reduced mouthparts may be nascent tooth ridges without teeth. Bresler (1954) showed that the incidence of abnormal tooth rows (and jaw sheaths) increased with temperature, but all tooth rows were not affected equally. Jaw sheaths seemed to be less affected than tooth rows. Bresler and Bragg (1954) reported variations in tooth rows of five species in five genera, including Spea bombzj?ons. It is interesting that variation in tooth rows appears to be greater among individuals and populations of species of Spea than for almost any other tadpoles. Perhaps this is a reflection of the relatively warm aquatic environment and shorter developmental period characteristic of these species, but a phylogenetic d u e n c e cannot be ruled out. A review of tooth row variability in other pelobatids and an experimental test of the effect of different rearing temperature on tooth row variability would provide some insight. It also is tempting to speculate that this variation somehow is tied to the drastic ontogenetic changes that occur in cannibal morphotypes of this group. Altig and Johnston (1989) proposed a balance value (i.e., the difference between the number of upper and lower tooth rows) to aid in understanding and comparing suspected functional Merences in tooth row formulas among species.

BODY PLAN

41

They assumed that differences in balance values reflected Merences in feeding mode. A tadpole with more tooth rows on the upper labium than on the lower labium has a positive imbalance (e.g., LTRF of 513, BV = +2); one with more rows on the lower labium than the upper has a negative imbalance (213, BV = - 1); and one with equal number of rows on both labia is balanced (515, BV = 0). A balance value of - 1 in the formula 213 is by far the most common. The extremes are + 7 (1013, Bouphis sp.) and - 11 (4115,Heleuplnyne sp.), and not all intermedate values occur. The distribution of values shows that 88% of 627 species surveyed (51% had a LTRF of 213) had balance values between + 2 and -2. A tadpole with a strong negative imbalance (e.g., LTRF 4/15; BV = - 11) surely must use their labial teeth Merently than one with a strong positive value (LTRF 101 3; BV = +7). A comparison of foraging behavior and feeding efficiency among tadpoles with the same total number of tooth rows (e.g., 12) but Merent balance values (LTRFs of 418 vs. 616 vs. 814) would provide needed data. It is interesting that the tadpole with the largest number of tooth rows (LTRF 17/21; Hyla sp.) has a nearly balanced formula. Except for the pronounced ontogenetic changes characteristic of tadpoles of species with many tooth rows, the number and arrangement of labial tooth rows (LTRs) after about Gosner stages 25-26 are stable enough to be definitive of the species. A LTRF does not provide information about the position on or spacing between tooth rows on the face of the oral dsc. Row spacing has not been examined, perhaps because the flexible oral disc makes it dfficult to standardize the measurements, but dfferences apparently exist. An unidentified suctorial hylid tadpole from Papua New Guinea with a LTRF of 213 has an unusual arrangement of tooth rows on the lower labium; the space between P- 1and P-2 is over twice the distance between the other two rows. Large areas of the d s c in certain tadpoles (e.g., Duellmunubyla, some Tauductylus; species with large oral discs and short tooth rows situated close to the jaw sheaths) are without teeth. Tadpoles of some centrolenids (personal observation) and suctorial forms of Litoria and Staurois (Inger and Wassersug 1990) have a large expanse of oral d s c beyond the dstal posterior row. Although we have a poor appreciation of the nature of and limits to intraspecific variation, tadpole morphology at the generic level often is remarkably consistent. Interspecific variation in some taxa (e.g., most bufonids) is much less than in others (e.g., hylids, leptodactylids). Other than the cannibalistic morphotypes of Spea, true polymorphism, not just pronounced variation, is rare or poorly documented in tadpoles. Annandale (1917, 1918), Berry (1972), E. R. Dunn (1924), Khan and Mufti (1994), and M. A. Smith (1917) talked about notable dfferences within species, but the causes of the differences are unclear. I. Grfiths and De Carvalho (1965) used some of these cases as examples of dimorphisms that must be considered with caution when using larval features in systematic studes. Although the premise is valid, the underlying assumptions that the tadpoles were correctly identified and that the correct number of species was involved in all cases went unquestioned. Likewise,

M. Kaplan (1994) suggested that variations of labial tooth rows in Hyla mznuta prohibited the use of these features in phylogenetic studies; the problem here most certainly centers on the variation representing multiple species. A smattering of reports (e.g., Annandale 1917; Korky and Webb 1991; Lavilla 1984a, b; J. M. Savage 1960; M. A. Smith 1917) described apparent geographic variation in mouthparts, and a few others (Annandale 1917,1918; Berry 1972; Caldwell 1982; E. R Dunn 1924; Khan and Mufti 1994; M. A. Smith 1917) have documented notable examples of other types of morphological variation (e.g., coloration and pattern) within species. The paucity of data on both the nature of and limits to intraspecific variation raises questions about the reliabilitv of some t a d ~ o l e identdications. We he, quently ask ourselves if observed differences are an indcation of geographic variation w i h a species or a reflection of dfferences between species. Tadpoles of s~ecies oi's~ea notorious for having variare Lose induced by cannibalis; ( q . , able mLuthpartsLbeyond Bragg 1956; Bragg and Bragg; 1958; Bragg and Hayes 1963; Bresler and Bragg 1954; Hampton and Volpe 1963; Orton 1954; Potthoff and Lynch 1986). Yet, the relationship between a reported missing tooth row or patch of teeth and the presence of the underlying tooth ridge (see Bresler 1954') seldom has been explored. A missing Goth row mav " be the consequence of a missing tooth ridge (i.e., developmentally induced) or merely reflect the loss of teeth on the ridge (i.e., attributable to some postembryonic response). Too often, the condition is masked by an imprecise or incomplete description. If we are to make progress in understanding the nature of presumed geographic variation, this needs to be corrected! Although intraspecific variation is poorly documented, the morphology of closely related species often is remarkably consistent. We have suggested (e.g., chap. 12) that this consistency occasionally signals the need for taxonomic reevaluation. Exceptions to this generic uniformity are known; the dramatically Merent tadpoles of Litoria citrupa and L. sub&ndulosa (Tyler and Anstis 1975) and Hyla chlysoscelis and H. avivoca are examples. In some tadpoles with an exceptionally short P-3 row, such as Pseudacvis trisertertata P s e u h s crucifer (Gosner and and Black 1957a), the third row is frequently absent, and a gap of equal size in the marginal papillae below the expected position of P-3 may be present. A LTRF of 213 is typical in most members of the Ranapipiens group, but a LTRF of 3/ 3 is not uncommon; the occurrence of the unconventional a formulas of 214 and 314 in R m bevlandieri (Bresler and Bragg 1954) needs further study. Cannibal morphotypes of Spea have larger, differently shaped jaw sheaths and few to no teeth compared to the conspecific omnivorous morphs. Aberrant and dsplaced tooth rows are not uncommon; parts of rows may be absent, rows may merge at any point or connect laterallv. and teeth mav form small circles within a row. Gosner (1959) attributed these alterations to wear, but thls seems unlikely. Documentation of wear is not as common (e.g., Tubbs et al. 1993) as one might expect from structures that are being replaced continually. ~ e m ~ e r a t u r e ,
i'

RONALD ALTIG AND ROY W. McDIARMID

nutrition, disease, and the substrates on which the teeth act may contribute to aberrant patterns. Cultured tadpoles often have higher incidences of abnormal tooth rows (and in specific patterns; Grdlitsch and Grillitsch 1989), including a complete lack of keratinized structures, than field-collected specimens. Some taxa (e.g., Bu.) seem more likely to show such alterations than others (e.g., Rana). In 8 of 10 simultaneously field-collected specimens of Heleophryne purcelli aberrant teeth appear in the two proximal upper and three proximal lower rows. Apparently teeth were not shed as replacement teeth were produced, and the result was a remarkable array of long flexible strands of up to 40 teeth extendng above the surface of the tooth ridge at the site where a single tooth would normally be. This condition surely prevented the teeth from working properly, but the tadpoles appeared healthy. Interpretation of some kinds of anomalies may provide hypotheses about the formation, evolution, and functions of mouthparts. Each keratinized labial tooth (fig. 3.8) is derived from cells in the base of the tooth ridge and consists of three indistinct regions: a distal head, an intermediate body, and a basal, hollow sheath (Gosner 1959). A fracture plane in the body often is evident. At this point, the tooth head can break free of its body and expose the head of the next replacement tooth interdgitated in the sheath. The head of a tooth may be straight or variously curved and spoon-shaped (i.e., bluntly rounded and flattened; fig. 3.71), thin with a sharp point (noncusped), or of various shapes with 2-18 terminal cusps. Even though the head often is flattened front to back, the body is the reverse -wide in lateral view and narrow in anterior view. Strands of connective tissue anchor the hollow sheath of the tooth within the tooth ridge; attachment is most prominent at the front (proximal to the mouth) base of each tooth. In most teeth, the front and back bases of the sheath in cross section are of comparable shape, but in suctorial forms, the front base of the sheath often extends considerably toward the mouth. This extension protrudes (fig. 3.7E) into the tissue of the tooth ridge proximal to the row and presumably acts as a brace to keep the tooth series and row from folding backward. The imprint of the nucleus of the original cell can be seen in the wall of the sheath of a tooth cleared in potassium hydroxide. Tips of the labial teeth on both labia face the mouththat is, in opposite directions. Usually two to three f d y formed replacement teeth are successively interdigitated in the sheathof the p r e c e h g tooth below ;he erupted tooth. The amount and style of interdigitation appear to vary, but data are laclung. An erupted tooth and all its replacements are termed a tooth serieL(figs. 3.7B, E), and the population of tooth-forming cells is renewed continuously in a mitotic zone near the base of a tooth ridge (fig. 3.7C; Beaumont and DeunfT 1958; DeunfT and Beaumont 1959; Luckenbdl 1964,1965; Tachibana 1978). As the presumptive tooth cell moves toward the top of the tooth ridge, a replacement tooth slowly forms from the head toward the sheath. A tooth row comprises numerous tooth series aligned side by side within a tooth ridge. If the tissue of the tooth ridge is sufficiently transparent, the tooth series can be seen through

Fig. 3.8. Photomicrographs showing variations in labial teeth and their marks. (A) Short-cusped, moderately curved teeth ofMgaelosia Joeldii (Leptodactylidae). ( B ) Long-cusped, strongly curved teeth in single row/tooth ridge of Hyla bwbeba (Hylidae). (C) Cusped, relain tively straight tooth (emw~ent upper left) and replacement series of Lepto+lus chagmnsis (Leptodactylidae). ( D )Noncusped (pointed), short-sheathedtooth of Spea bombcfions (Pelobatidae). ( E ) Negative cast of mark (uppwjawsheath entered on left) made on agar-based food by feeding tadpole of Rana sphenouphala (Ranidae). (F) Cusped, biserial teeth of Bombina van'gata (Bombinatoridae).From Altig and Johnston (1989);reprinted by permission ofHetpetologzcalMonogvaphs.

the tissue. What appears as a single long tooth with sequential light and dark areas are actually the replacement teeth stacked in a series. Sometimes an older tooth sticks to a newly erupted one and makes it appear longer than others in the row, and other times a tooth looks longer because of the dfficulty of interpreting the extent of the soft tissue of the tooth ridge on the shaft. However, closely adjacent teeth are comparable in size and shape. Tooth size may vary in different parts of a row or in dfferent rows (e.g., fig. 3.7A, I; Gosner 19-59), but the usual pattern is a gradual shlft to smaller size laterally within a row and dstally among rows. Smaller specimens have smaller teeth than larger specimens. Meager data suggest that rows increase in length without changing tooth density; where and in what pattern teeth are added remains a mystery. Most tadpoles have a single tooth row per ridge (= uniserial), but tadpoles with bi- (e.g., dscoglossids) and even tri- or multiserial (fig. 3.7A) rows are known. Tooth densities of 30-100/mm, with higher counts in lotic tadpoles, are typical, and the number of cusps range from 0 (simple spike) to at least 18. Biserial sections in typically uniserial rows are not uncommon. The teeth of Bombina orientalis are uniserial at first appearance but become biserial during development. Noble (1929) noted biserial tooth rows in Hopfobatrachw rttgufosus, but the text and adjoining figure leave doubt if these rows actually are biserial. Tadpoles ofXenopus lamis develop true calcified teeth precociously (J. E Shaw 1979). True teeth usually appear late in metamorphosis as the maxlllae and premaxillae ossify, but in X. laevts, tooth buds are present at about stage 33 and start to calcify at stage 37. True teeth begin to form at about stage 36 in tadpoles of Lepidobatrachw (Starrett 1973). Jaw Sheaths Jaw sheaths are formed by the fusion of palisades of keratinized cells along their lateral margins (Kaung 1975; Kaung and Kollros 1976). Sometimes iaw sheaths have a striated appearance to at least their basal parts. As the serrated edges of the jaw sheaths wear away or break off, mitotic cells in the base of the sheaths continually produce new cells that keratinue in transit to the dstal edge. Typically, jaw sheaths have a serrated edge, and the serrations vary in density, orientation (straight medially and angled laterally), presence (rarely absent, e.g., Osteopilusbrunneus group; Noble 1929), shape (broad-based and short vs. narrow-based and long), and sue. In a few cases, 1-3 enlarged serrations occur withtn

44

RONALD ALTIG AND ROY W. McDIARMID

a normal graded series on the upper (e.g., Plectrohyla ixil and

l? matudai) or both (e.g., Leptodactylodon ventrimamratus)

sheaths. Individual serrations lackmg a common sheath occur as transitory structures (e.g., Visser 1985, some young Heleophyne) or for the duration of larval life (e.g., Cei 1980, Ruibal and Thomas 1988, Lepzdobatrachus; personal observation, Mantiaktylus lzgubris; Pyburn 1980, Wassersug and Pyburn 1987, Otophlyne). Jaw sheaths with 30-80 serrations/mm are typical, and higher counts are known in lotic forms. In a few cases (e.g., some Amlops and Boophis), serrations on the edges are only 5-8/mm. Among suctorial forms of some Boophis, the sheaths appear to be composed of a series of fused columns; the irregular surfaces of these sheaths are in stark contrast to the smooth surfaces of typical forms. Some Amlops (Ranidae), Ansonia (Bufonidae), and Litoria (Hylidae) have one or both jaw sheaths divided medally; the functional significance of such arrangements is unknown. A comment bv Menzies and Zweifel(1974) that older specimens of a species of Litoria with broken sheaths have a single sheath suggests ontogenetic changes, but this needs to be examined more closelv. The umer and lower sheaths are usually similar in si7~, the U-shaped lower but sheath ofRrcaphus is minute relative to the large, flattened, upper sheath. Tadpoles of Heleophryne lack one (upper) or both sheaths, and the upper one is absent in an unidentified tadpole of Boophis (personal observation). As with tooth rows and ridges discussed above, the apparent absence of jaw sheaths must be evaluated carefully. Absence of pigmentation does not corroborate the lack of keratinization (e.g., Scaphiophlyne), so a judgment of absence should be based on an actual lack of a sheath on the labial cartilages. Taw sheaths " usually co-occur with labial teeth but may occur alone (e.g., hylid: Hyla leucophyllata group; ranid: Occidozyja; microhylid: Scaphiophlyne). Rarely do tadpoles have labial teeth and no jaw sheaths (e.g., some ~deophlyne). The upper sheath of some suctorial forms may be M-shaped, and some species (e.g., Hylapiccipes) appear to have a scouring surface iust to on the front face of the umer sheath , ~roximal the serrated edge. Based on the various accounts, the structures of the upper sheath of Hoplobatrachzts tigerinus (surely confusion exists in dstinguishing between the tadpoles of Hoplobatrmhus tigerinus and Hoplobatrachus rttgulosus) apparently are unique, although it cannot be determined whether one or two features are involved. Khan and Mufti (1994:28) stated that there is "A pair of long cylindrical thick papillae, tipped with keratinized plates . . . at the angles of the mouth opening. . . ." These structures, perhaps keratinized buccal papilla< do not appear associatedwiththe upper jaw sheath. Kirtisinghe (1957) illustrated the lateral margins of the upper sheath with convexities but made no mention of them in the text. Khan and Mufti (1994) also noted that a ". . . median-buccal keratinized shikld is'observable through the mouth." This structure must be similar to the keratinized boss on the roof of the buccal cavity of Spea and some Hyla. Various iaw sheaths are included in fig. 3.9. shapds of jaw sheaths are difficultyo describe because of their complex curved shapes. The curved jaw sheaths and trajectory of the cutting edge vary tremendously among taxa,
L L L L L

but neither have been described well. Wide infra- or suprarostral cartilages accompanied by narrow keratinized sheaths occur, and the difference between this situation and one where narrow sheaths cover most or all of the cartilages needs to be qualified. To visuahze jaw sheath shapes better, imagine a rectangular piece of paper with the opposite short ends placed on a table so that the paper forms an arch that represents the typical smooth arc of the upper jaw sheath. The same concepts pertain to the lower sheath, although it is more difficult to simulate because the sheath is most often V-shaped. With proper pressures on and repositioning of the ends, the paper can be fashioned into various uniform (e.g., tall and narrow vs. short and wide) or nonudorm (e.g., less curvature in medial section) arches that effectively illustrate the known variations in the shapes of jaw sheaths. The facing edge, which equals the serrated edge of the sheath, can be cut to form various uniform or nonuniform (e.g., incised sheath with a medial convexity) borders. Also, various (uniform and nonuniform) serrations that usually grade from the largest medally to the smallest laterally can be cut into the facing edge. The ends of the arch can be modified in shape and lateral flexure to form the indstinctly differentiated lateral processes. Last, the roof (= front, oral, or adoral surface of sheath) and ceiling (= rear, back, aboral, or buccal surface of sheath) of the arch could be nearly flat and parallel, similar to the paper, or could curve to various degrees. The line of keratinization on the front face of the upper sheath typically describes a smooth, either abrupt or graded, arc, but the line of dark keratinization on the buccal face of the upper sheath often takes various shapes (Altig and Johnston 1989) with graded or abrupt margins. The lower sheath passes inside the upper sheath during a feeding stroke, and through scissorlike interactions serrations on the two sheaths cut or gouge surfaces for food removal. The degree of keratinization varies from weak (yellow to light brown) to strong (dense black); other than color differences, there is no system for even subjective evaluation of these differences. The cause or functional significance of light keratinization in sheaths that normally are heavily keratinized is unknown. Structures with melanic pigment are stronger than those that lack h s pigment, but the strength and resistance to wear of sheaths that differ in degrees of keratinization have not been evaluated. Because of the way jaw cartilages articulate with their supportive structures, jaw sheaths are oriented and operate at several angles. A wide variety of sheath shapes can be seen in buccal view in Wassersug and Heyer (1988). Other keratinized structures are associated with the oral d s c and buccal cavity. A rounded, heavlly keratinized boss occurs on the ceiling of the buccal cavity just posterior to the buccal base of the upper jaw sheath in tadpoles of Spea. Tadpoles of Kassina sengalensis, T~chobatrachus robustus, and several members of the Ranapipiens group, mostly with robust lower jaw sheaths, commonly have areas of thin keratinization lateral to the oral face of the lower jaw sheath. Although unclear, a comment by Annandale and Rao (1918) suggested a unique feature of the tadpole of Euphlyctis hexadactylus: "There is a deep groove, with its sides and base

ME[

pun

pun d d e IOU dem SMOJ ylm~ ~ u a u q d m a q u a ayl uupspy se luasard a n aepded p @ n m ~ u m s e u ' 2 2 jo o~ pur! Inq '(0661 r a 8 q pue gsuazpo~) dua801uo u r a g m nadde a k s dq uaas aq m %mum%aq sa! 5qp ayl uo luamrd sy ! 'SUIJOJ ~ I O P I I d p a d s a 'SMOJ ylm~ S duem q q s ~ p d s j o u o u e @ m a pale1 e JI -1Turapun a m o q ped a y l p sean a s a m r u l s p o ayl 'salodpa p!dh aroux a p a n d m - m a n p u a A o p m pue dpaaqaa1ue ayl uayl pue s a w pale1 aSppuou t p paredm03 h y d p $ g s ~eadde ~ pue r a 8 q -oauaA ayl ' a x p n s 8 q u n o ~ mayl a A q 8 q . 1 JT u r n s a n sa+u ayl uo q m pgaq!da a u - s m 7 uf dpm@!s 1Ga~ .aJarl mrno os@ m a p o m dpoq ayl mno@no.np nw r a m IOU o p re^ snyl pauyexa sapads ayl Jnq ' x - 2 sa%as -30 ~ e yq m pa1ag13 33e3.m~ ~ u n o m ayl aAoqe dp@gs l 8 s mqel yloq jo &I 8-P (12 'Z/Z '2/1 '111 ' o f 130 W @ ~ u a n b5s-d '2-'6 s a y 'umapomo~sayl s p u n o m s ~ e y l 7 s '1-a '2-J ' 1 - v '-%a) muanbas e q readde saSpu y l o o ; ~ -m a1!sodu103 ayl 'ped p o ayl 'aurp a m s arp jnoqe q r -luamdo1amp auo raddn arp &q -12 a k s lnoqe 1e auwqmam p%idoJo ayl jo a m d m ayl a b ~ % yleaz~s ~ M O ayl ' A w n ' ( 9 ~ 6 s1a r l l q PW %we)]: m ~ p s d p~u u a m uoraluarpre a y l j o quid pue mnap s J T ayl f ~ L 6 &me)]: ' ~ 6 F1q u a ~ n 12-02 ~~s l n W e -ouro~s 30 q d a p ayl q saswmq Iuanbasqns -3sp p o 1 Is961 ~) le a z p a a q 01 m s dayl aropq la& luandde sy stpeags ayl jo luama8m.m dq p a w a m p SE uoqsod ~ u a n d d e sq ayl 30 UUOJ a u epanadda aAeq saSpu y l m r a p ueyl ratpar @nue a v lunome o ~ aq p lsnm auo ylnom ayl ULIOJ IOU o p a q d e d p@~f!mqns J S O .suouepuarj ~ JO uopejo1 ayl Srqupmavp UI .dp lnous ayl anoqe dp$gs r o
Sup~npucn s)uauodufo3 (a) D'pazymq L p w y XOUI .uo!sspmad dq p a a ~ d a ~ q w s ME[ '13qsyp awded p 6 m . u 'unpsyp s-d ada~xa -1.1 !'TI' ~ O 3 ~ P LW J O ~ P ! S ~ " S ' T S U ' 'w-tl!! '8861 U104 W (3)'2J a w ' 2 6 Pm 1-V 103 O I1 l a ~ d - d J. ym e '3 e ! q q (8) s a@&dw '(8861)~ J pm n q n e 4 U I O J ~ V ~ .spxpla ayl uo s a a d a s SU@JVUI p q a yap "ruasardq w y s MV[ 1 d q q ! q =aau= qayl a sep a w j o s p d y qp p a a v pm 'sp '12n q 'auwquxaoup i h h q d w o ayljo d w ~aap p p v l v -md lP3O S J ~ U JJaMOIFa19!V ( 3 ~ ) a A ! W P 4@ @ P O (v) ( a ~ ! l d ~ ) J~ L H poleJ1~ll! 'dXT3 &/Z ~ SP F ' W w dq se '92 mu9ot) Fn r adma p a @ ! q q'92 nus00 d i m '~uau~do~anap a~odpm ! h e 30 m d d e PO ayl jo auamdopua -OT-E a. 1 qapi pd

BODY PLAN

Fig. 3.11. Photomicrographsshowing development of the oral apparaars in three tadpol~es. o : Hyh s a y d (Hylidae), a tadpole with Tp jrw sheaths, reduced oral disc, and no labial t e h (A) Jaw sheath visiet. blc, mvginapapillae starthg to appear, Gosna 24. (B) Distinct seratioason jaw sheaths and marginal papillae arc obvious, Gosna 25. W: G&qhtync & d (Mimhylidae),a tadpole with oral Saps and no keratinized mouthparts. (C) Oral disc relatively un&a&ad, conical adhesive gland visible in lowa lefi coma, Gosna 24.
@) Labial flaps beghmng to difkentiate, lower jaw (= idkhbial pornincnce) visible as rounded projection in notch between labial

flaps, Gosna 25. Bottom: R b i ~ d o ~ a J j s (Rhinophrynidae),a tadwle without keratinized mouthparts. (E)Oa disc u n e n t i rl a&, without jaw sheaths or & , line& tranmrse adhesive gland visible below mouth, nares visible dorsal to mouth and still plugged with cdls,Gosna 22. (F) Barbels bcghnq to differentiate around slitlike mouth, floor of the buccal cavity d i d y inside mouth visible becaw o slight downward flexure of lower jaw, medial papilla on f lowa jaw visible, Gosner 25. From Thibaudeau and Altig (1988), copyright 0 1988, Wdey-Liss, Inc., a subsidiary of John Wiley & Sons,Inc.; qrinted by permission.

48

RONALD ALTIG AND ROY W. McDIARMID

somewhat later in ontogeny compared with more typical tadpoles. Heyer (1985, Hyalinobatrachium uranoscopa) and Inger (1966, Leptolalux gracilis) noted cases where tooth rows are lost and the tooth ridges seemingly are transformed into a series of papillae. Some data (Altig and Johnston 1989) suggest several patterns of tooth row appearance once a formula of 213 (see above) has developed. Some tadpoles add rows proximally to the rows of the 213 formula on the anterior labium and-distallvon the lower labium. Others add rows distally on both labia: Teeth first appear near the medal parts of tooth ridges and then laterally. The sequence of tooth row appearance among ridges is the same as that for ridge formation. Hemispherical bulges in the epithelium can be seen where teeth are forming underneath, and the first teeth that appear may have a dfferent form than subsequent ones (Thibaudeau and Altig 1988; Tubbs et al. 1993). From their first appearance and throughout tadpole ontogeny, as labial teeth are lost after wear or breakage, replacement teeth are continually formed from mitotic zones in the bases of the tooth ridges (e.g., Luckenbill 1964, 1965; Kaung and Kollros 1976). In most tadpoles, several replacement teeth are interdigitated in a series (figs. 3.6B, E); in a few cases replacement teeth apparently are absent (e.g., dstalrnost rows in Ascupbus). Atrophy of the oral apparatus has not been well studied, but apparently it occurs in the approximate reverse order of ontogeny and more haphazardly. Presumably, mitoses that produce teeth cease and the last teeth are lost in patches. It is not known if a Dattern to tooth loss within rows exists. The initiation, duration, and perhaps pattern of metamorphic atrophy varies among taxa and ecological types. For example, row P-3 (LTRF of 213) is the first to dsappear, and A - i usuallv is the last. Taw sheaths are lost after or most tooth row;. ventrolatek marginal papillae are the last structures on the oral disc to atrophy.

Functional and Evolutionary Aspects of Tadpole Morphology

Too often we have indicated the paucity of information regarding the functional or evolutionary aspects of various larval structures; although bothersome, this reflects the inadequacies of our knowledge. Morphological features and their dstributions among taxa are stdl being described, and often we have relatively little information on the development of such features and less on their functions. Until more of these holes in our knowledge are Ned, probing discussions on the evolution of morphological structures in tadpoles will remain speculative. Body Motpbology Exemplified by the hypotheses on ecomorphological guilds proposed by Altig and Johnston (1989), further observations suggest strong correlations among the following traits: behavior, body and fin configurations, ecology, metamorphic pattern (e.g., Nodzenski et al. 1989), and taxonomic position. Seven generalizations abstracted from Altig and Johnston's (1989) guild hypotheses involve associations

among features of the body, eyes, fins, tail muscle, and mouthparts. (1) Lentic forms have less massive tail muscles than lo& forms, and the smallest muscles are associated with the largest fins. (2) Benthic forms have depressed bodes, dorsal eyes, and low fins whether in lentic or lotic environments. (3) Lentic (pond), nektonic (rarely lotic) forms have compressed (e.g., many hylids), depressed (e.g., microhylids and pipids, sometimes with a tail flagellum), or equidimensional (e.g., macrophagous feeders with jaw sheaths but lacking labial teeth, some Hyla, often with a tail flagellum) bodies and live in different parts of the water column. (4) Burrowers (e.g., centrolenids) and those tadpoles that live in confined spaces (e.g., bromeliad cisterns; Hyla bromeliacia, H. dendrosma) are vermiform with depressed bodies, dorsal eyes, and low fins. (5) Semiterrestrial tadpoles have elongate bodies, narrow tail muscles with abbreviated fins, large eyes that bulge above the surrounding body surface, and hmd legs that develop precociously (Drewes et al. 1989). (6) TadDoles that live in slower reaches of streams resemble lenticbenthic forms except for a typical increase in the number of tooth rows; ranids make up a large component of this assemblage, and the number of upper tooth rows is often larger than the lower. (7) Lotic forms that use the oral dsc to maintain position and feed in fast-moving water have complete marginal papillae and frequently have more tooth rows and higher tooth density; larger numbers of tooth rows presumably signal faster water habitats. The largest LTRF .of 17/21 is from an unidentified Hyla from a high-energy stream in the Venezuela. In these tad~oles fins are low and the eves are dorsal. There probably are two poorly understood patterns of streamlining. Almost no data (see Gradwell 1971a, 1973,1975b) exist on the actual functional proficiency of the basic morphological categories described above. The functions of other features (e.g., vent flaps, dextral vs. medial vent tubes and their many variations, tubular nares, naris size, and long, free spir a d a r tubes) have not been suggested or are seldom supof portable by data. Based on obse~ations where tadpoles occur in the water column (e.g., Alford 1986b; S. L. Mitchell 1983: see Abelson et al. 1993 and Lancaster and Hildrew 1993) and their swimming abilities, associations between tail f n shapes and tail muscle configurations surely signal i different ca~abilities.Unfortunatelv. the data on the swim,' ming kinematics of tadpoles pertain only to a few species of lentic tadpoles (e.g., Hoff and Wassersug 1985, 1986; Wassersug 1989a; Wassersug and Hoff 1985).
\ ,

Oral Morphology The functional roles that have been proposed for oral papillae fall into two basic categories: chemosensory and tactile receptors and structures that control water flow (Van Dijk 1981), enhance attachment to substrates (Altig and Brodie 1972; Gradwell 1971a, 1975b), mod@ the shape of the oral disc during feeding, and manipulate food and substrate particles. As is usually the case, few data are available. The size and density of oral papillae in many cases seem to reflect lineages and habitats. For example, ranid tadpoles have larger papillae arranged more sparsely than hylids, and stream hy-

BODY PLAN

49

lids have smaller, more densely arranged papillae than pond h!lids. The number and prominence of marginal papillae \=aria concordantly with observed reductions in the size of d x oral disc among tadpoles w i h n the same lineage. A comprehensive survey of labial tooth structure has not been made, and the few descriptions that have been published vary in many details (e.g., Altig 1973; Altig and Pace 1974; Faier 1972; Gosner 1959; Heron-Royer and Van Bambeke 1889; Hosoi et al. 1995; Inger 1983; Korky and Webb 1994; R. J. Nichols 1937). Little can be said about functional diierences in tooth morphologies (e.g., Altig and Johnston 1989; Faier 1972; Gosner 1959; Inger 1983). Some suggested functions include acting as a broom, serving zs a current generator, breaking up mucilaginous layers, sieving pamcles, combing strands into alignment for easier cutting (Altig and Johnston 1989), piercing plant cells, rasping hd particles from a substrate (R. M. Savage 1952), holding food (Luckenbill 1965), attachmg to a substrate (Altig md Brodie 1972; Gradwell 1975a, b), and trapping food iT:-Ier 1963). Few data exist to support any of these suggesdons, and the immense variations remain enigmatic. Mertrns (1960) and Nachtigall(1974) compared the morpholoejes of tadpole teeth to those of structures in other groups ;bar presumably serve similar functions. Publications that m u a s t tadpole teeth to the snail radulae (e.g., Steneck and mahg 1982) provide other pertinent examples. Likewise, only a few workers (e.g., Altig and Johnston 1%9) have attempted to place the sizes, shapes, and serration patterns (e.g., fig. 3.6H) ofjaw sheaths into a functional mm.Even though the upper and lower jaw sheaths in a tl~pical tadpole are similar in size and probably serve as gougiug'bitingjscraping structures, their striking differences in dupe and the extensive array of morphologies across taxa qgat an innumerable variety of performance abilities. The immense, flattened upper sheath ofAscaphus is unique and must shave periphyton from various substrates during the mraaion phase of the oral dsc: the small lower sheath is @ably inconsequential. The robust, narrowly arched jaw s&arfis (i.e., short surface for actual biting) of semiterrestrial n l p o l a presumably provide strength and may be essential Q feedmg on tough materials or unusual substrates characd c of these thin (water film) habitats. E-xcept in the most general sense, the functional morpholog-of the oral apparatus of tadpoles is largely unknown. Enensive work has focused on the structural components d functional design of the food-trapping mechanisms in rbc buccopharyngeal area (e.g., Kenny 1969b, c; R. M. Savq c 1955; Viertel 1984a, c, 1987; Wassersug 1972, 1980; V k s m u g and Heyer 1988), but insufficient attention has haen p d to the mechanics of the external feedmg strucmaa. Their small size and the rapid operation of these nmudpm pose several challenges to be overcome in desiping a detailed study of the functional components of kcdug by tadpoles. These problems are compounded by the pucedural difficulty of producing standard food substrates &xcomparative testing. When these complications are conridaedin light of the enormous amount of interspecific varixim eluciaating the functional and evolutionary aspects of

the workings of the oral apparatus during feedmg has proved to be especially difficult. Even so, a brief review is possible and worthwhile. The upper supra- and lower infi-arostral cartllages (chap. 4) that are the jaws of a tadpole have almost no adult derivatives, are not articulated at their bases, and are operated in a unique mechanical fashion. The suprarostral cartilage articulates and is movable against the chondrocranium. The lnfrarostral cartilage articulates and is movable against Meckel's cartilage, which articulates and is movable against the chondrocranium. Therefore, two joints occur in the lower system and one in the upper. In addition, the suprarostral cartilage is not closed through muscular contraction; rather, this cartilage with its attached upper jaw sheath is closed through a tendon-and-pulley system via movements of the lower, infrarostral cartilage (Gradwell 1968, 1972b). This arrangement leads one to conclude that the jaws must operate together and that both sheaths are involved in food removal. Two sets of unpublished data (G. F. Johnston 1982, 1990) and other observations confirm that we do not understand the exquisite mechanics of h s system. Carr and Altig (1992) also found interspecific differences of the rectus abdominis muscle that apparently correlate with feeding mode (figs. 3.3A-C). Among others factors, species, stage, position of the food in the water column, and density of agar-based food alter the following generalities. It also is important to remember that only a small group of ecologically similar tadpoles has been compared. Feeding trials in the laboratory indicate that the lower sheath alone or the two sheaths acting together can remove agar-based food. In the latter case, the mark of the lower sheath on the agar is usually longer than the mark made by the upper sheath. Even though the two sheaths must move simultaneously, both do not have to contact the substrate. Substrate contact may be avoided by changing the angle that the tadpole takes relative to the substrate or perhaps by adjusting the longitudmal relationship of the two sheaths. The angle between the two infrarostral cartilages probably can change during a bite, and the depth of penetration of either jaw sheath can change during a bite. The lower jaw sheath may strrke several times while the upper sheath takes but a single gouge from agar-based food. On agar that is too dense to penetrate, both jaws may strike repeatedly without closure. The labial teeth may leave no mark, leave a mark but show no movement, or show rasping movements with marks left on top of those made late in a feedmg stroke by the jaw sheaths. Insufficient information has contributed substantially to an inadequate understanding of how tadpole mouthparts work. A better appreciation of the following issues would add considerably to understandmg their evolution: those factors important to the evolution and maintenance of the tadpole stage in the life history of a frog; a consensus on primitive or derived state of various larval traits; an appreciation of the full compliment of morphological variations and their functions; baseline data on geographic variation; and further facts on ontogenetic variation, developmental information, and the ecological significance of various larval morphologies. For these and other reasons, oral structures have

50

RONALD ALTIG AND ROY W. McDIARMID

been used infrequently in systematics analyses (e.g., Duellman and Trueb 1983, 1986; Grandison 1981; I. Griffiths 1963; I. Griffiths and De Carvalho 1965; Inger 1967; Kluge and Farris 1969; 0. M. Sokol1975) and debated less often. The lack of a clear phylogenetic hypothesis of some groups surelv has exacerbated the situation. but cursorv examinations suggest to us that tadpole morphology commonly is concordant with perceived taxonomic boundaries and perchance signals the need for closer examination of relationships (e.g., B u . debilis group vs. other B u f , Hyla leucophyllata group vs. other groups of hylids, Scinm yostrata vs. S. rubva groups, Otophrynepyburnivs. other microhylids). Boulenger (1897:110) made a knowing statement in this regard: "The structural differences which separate the genera and species in their tadpole condition reflect, on the whole, pretty accurately the system based on the perfect animals. . . ." Yet, even heuristic attempts to gain ideas on evolutionary patterns by overlaying larval traits on cladograms derived from analyses of adult traits have not been done (see chap. 4). Degree dfcommonness, at least as presently known (e.g., a 213 LTRF in many g d d s and taxa), coupled with a bewildering array of morphological diversity, makes the situation aooear chaotic. Even so. inclusion of larval traits in I I phylogenetic analyses of anuran groups is essential. We are not sure whether external or internal characteristics will prove to be more informative, so we recommend investigators consider both. A few examples adequately illustrate the need for much more work. J. D. Lynch (1971) considered a LTRF of 313 or greater as primitive within leptodactylid frogs, and Drewes (1984) noted that larger LTRFs belong to the more primitive tava in hyperoliids. Noncusped teeth have been considered primitive by Noble (1931) and derived bv Gosner ( 1959). Al&ough devklopkental and functional data may help define characters and their states, their inclusion in phyloeenetic analvses should be with caution. We are convinced V that understandmg the developmental morphology of a character often provides insight into character homology. For example, determining the developmental history of a trait across lineages may enable one to distinguish between instances of what appears to be the independent occurrences of the same trait (homoplasy) and those that actually represent the occurrence of G o traits (apparent homoplasy).~ Altig and Johnston (1989) pointed out that patterns of early development and metamorphic atrophy were informative'and suggestive of tooth row homoldgi. Based on data from several sources (e.g., Dutta and Mohanty-Hejmadi 1984; Echeverria et al. 1987; Fiorito de Lopez and Echeverria 1984; Limbaugh and Volpe 1957; R. J. Nichols 1937; Thibaudeau and Altig 1988; Tubbs et al. 1993),tooth rows in species with a LTRF of 213 commonly appear in the developmental sequence of A-1, P-2, P-1, A-2, and P-3 (sequential formulas 110, 111, 112,212,213). Rows A-1 and P-2 are closely linked during development and may appear slightly reversed in time. Likewise, the first five rows to develop in species with LTRFs larger than 213 commonly appear in this sequence. In species with more than five tooth rows, subsequent rows apparently are added alternately on
i

the upper and lower labia, but in at least four different patterns two of these are apparently common. In ranids, tooth rows are added proximally on at least the upper labium, whereas in ascaphids, heleophrymds, and hylids, additional rows are located distally on both labia. Species with many tooth rows (e.g., suctorial forms) attain the full compliment much later in ontogeny and retain tooth rows further into metamorphosis than species with fewer tooth rows (Menzies and Zweifel1974; Nodzenski and Inger 1990). A thorough evaluation of row homology is hindered by the paucity of developmental series and the lack of understanding of how medial gaps in tooth rows develop and evolve. The rows in a 213 formula are designated the prime formula. In species with a formula greater than 213, the pattern differences that were proposed to account for additional tooth rows reauire that the orirne rows be renumbered once the entire compliment of tooth rows is present. Based on this observation, certain LTRFs should not exist. Of 124 species with fewer than 213 rows, 88 fit within the expected categories of 212 (loss of P-3), 112 (loss of A-2), 111 (loss of P-1), and 110 (loss of P-2) or 011 (loss of A-1). If homology is assumed, LTRFs such as 012, 211, and 113 (common in hyperoliids and African ranids) indicate the presence of other patterns. The genetics of tooth row formation, and for that matter of most other morphological characteristics of tadpoles, essentially is unknown. Fortman and Altig ( 1973) showed that F, crosses among species of Hyla from the southeastern United States resulted in hybrids that more closely resembled one parent than the other; seemingly many traits acted as simple dominants. Crosses of Rana sp. with LTRFs of 213 and 314 produced hybrids with a formula of 214 (Haertel and Storm 1970; also see Kawamura and Kobayashi 1960, Michalowski 1966). Although the relative sizes of various mouthparts remain nearly constant throughout large parts of larval ontogeny (Gates 1983), the patterns of growth of oral structures as they relate to feeding ecology are poorly known.

Summary
The biphasic life cycle of anuran amphibians comprises two quite different body plans on which selection can operate: tadpoles and adults. The products of this selection especially among the larval forms are amazing! As discoveries are made and papers are published, one is coerced into believing that the total morphological diversity of anuran tadpoles is nearly endless. Except for the pelagic areas in a continually flowing river or stream, there seems to be no freshwater habitat of any duration that does not harbor tadpoles. Even though we do not understand them well, numerous adaptations allow tadpoles to live and feed in these various habitats. The distinctions between ecological forces and phylogenetic restrictions remain muddled in part because of the shortage of phylogenetic hypotheses within major clades of advanced frogs. Throughout much of tadpole diversity, a tadpole is a tadpole is a tadpole: a simple, depressed body shape, dorsal eyes, sinistral spiracle, marginal papillae with a dorsal gap, and 213 tooth rows describes many anuran larvae. W i t h

ARCHITECTURE
Cranial and Axial Musculoskeleton
David Cannatella

Introduction
Ovid's epic poem Metamorphoses chronicles miraculous changes of all sorts-a boy is transformed into an eft for offendmg Ceres, a statue carved by Pygmalion becomes a woman, the eyes of the dead musician Argus are made to adorn a peacock's train, and tadpoles metamorphose into frogs. The Roman poet's view of tadpoles as ephemeral stages is not unexpected, for even amphibian biologists do not M y appreciate the significance of tadpoles as free-living organisms. Tadpoles use the transient source of energy available in aquatic primary production, which is not available to adults (Wassersug 1975); the larval stage provides the advantage of rapid growth in unstable environments. Being a tadpole means that cranial morphology is canalized into unique feedng structures at the expense of all else. The body cavity contains mostly gut. The tail is emphasized over the h b s as the instrument of locomotion. The reproductive tissue lies outside of the body cavity, so reproduction (and thus neoteny) is not possible until metamorphosis, when the tad is resorbed, the lunbs develop, and the body cavity enlarges. Thus, except for sex, tadpoles engage in most of the same activities as adult frogs- feeding, respiration, and locomotion- but in very dfferent ways. This chapter emphasizes the structure and function of the musculoskeleton as it relates to these activities.

Semina limus habet virides generantia ranas

Et generat truncas pedibus, mox apta natando

Crura dat, utque eadem sint longis saltibus apta, Posterior superat partes mensura priores. Ovid, Metamorphoses, book XV (Pattist 1956) Ev'n slime begets the frog's loquacious race: Short of their feet at first, in little space With arms, and legs endu'd, long leaps they take Rais'd o n their hinder part, and swim the lake, And waves repel: for Nature gives their kind,

To that intent, a length of legs behind.

Translation by John Dryden in Garth (1961)

Anuran Phylqeny and the F r t Tadpole is What did the first tadpole look like? Given a phylogeny, some inferences are possible. L. S. Ford and Cannatella (1993) reviewed the status of anuran phylogeny and presented trees based on mostly adult and a few larval morphological characters (fig. 4.1; alternative phylogenies are d s cussed at the end of the chapter). The four most basal lineages of living Anura are Ascaphus trztei, Leiopelma, Bombinatoridae, and Discoglossidae. These taxa have the Type 3 tadpole of Orton (1957). Pipanura (fig. 4.1) is composed of Mesobatrachia and Neobatrachia. Mesobatrachia includes Pipoidea, with Type 1tadpoles, and Pelobatoidea,with Type 4 tadpoles. The tadpoles of Neobatrachia are Type 4 except for those of Microhylidae, which are Type 2. If Orton's (1953, 1957) four tadpole types are mapped as unordered character-states onto the phylogeny, it can be inferred that the ancestor of Anura had a Type 3 tadpole (jaw sheaths, labial teeth, medial spiracle), the ancestor of Pipanura had a Type 4 tadpole (jaw sheaths, labial teeth, sinistral spiracle), and the Type 1 tadpoles (no jaw sheaths or labial teeth, paired spiracles) and Type 2 tadpoles (no jaw sheaths or labial teeth, medial spiracle) were independently derived from Type 4 (fig. 4.2). This is similar to a proposal by 0.M. Sokol

ARCHITECTURE
I

53

Caudata Ascaphus truei Leiopelma Bombinatoridae Discoglossidae Megophryidae Pelobatidae Pelodytes Rhinophrynusdorsalis Allophryne Brachycephalidae Bufonidae Heleophryne "Leptodactylidae" Limnodynastinae Myobatrachinae Sooglossidae ~hiniderma Hylidae Pseudidae Centrolenidae Scaphiophryninae' Scoptanura Dendrobatidae Hemisus Arlhroleptidae' 'Ranidae"

species. Instead, it is hoped that h s chapter will stimulate descriptive and functional studies of other taxa, so that a more complete phylogenetic synthesis of tadpole structure and function may emerge. GeneralAnatomy This chapter provides an overview of the skeleton and muscles of the chondrocranium, jaws, hyobranchium, and trunk. The general anatomy of each region is described, and intertaxic variation is discussed in a ~hvlogenetic context when u possible. Following the general comparative anatomy are sections treating changes during metamorphosis, the functional anatomyand evolution OFgill irrigation mechanisms, and phylogenetic aspects of tadpole morphology. Only recently (e.g., Haas 1993, 1996b, 1997a) have phylogenetic analyses using primarily larval musculoskeletal characters apA ,

Discoglossanura

LA Neobatrachia

Micrhylidae

Fig. 4.1. Phylogenetic relationships among the major taxa of Anura. The dots represknt node-based names. Symbols: asterisk = metataxon, names in quotation marks = nonmonophyletic m a , and dots = nodebased names (L. S. Ford and Cannatella 1993).

(1975) and in stark contrast to that of Starrett (1973), which is discussed in the section Evolution of Orton's Tadpole Types. Although the ancestor of Anura can be inferred to have a Type 3 tadpole, in itself this is not very informative. Orton's tadpole types are defined primarily by the presence/absence of jaw sheaths and labial teeth, and the number and position of the spiracles. The taxa possessing the plesiomorphic Type 3 larva (fig. 4.2) are &verse.Ascaphus truei has a stream tadpole, and Lewpelma has larvae with "direct" development (chaps. 2 and 7). More typical pond tadpoles are present in Bumbina and Disw~lossus in many other groups, but their and nidespread phylogenetic dstribution does not guarantee that the pond tadpole morphology is plesiomorphic for Anura. It is generally assumed that the speciahzed morphologies of the Ascaphus tadpoles andhwpelma larvae are derived from some pond tadpole ancestor (I. Griffiths 1963), but phylogenetic analysis of larval characters does not demand this interpretation. Certain features of the Ascapbus tadpole (e.g., reduced muscular process of the palatoquadrate) might be either plesiomorphic for Anura or apomorphic for Asmphus. What can be said with confidence is that the ancestor of Bombinanura (figs. 4.14.2) had a pond tadpole, but h s ancestral larva is quite removed phylogenetically from the typical pond larvae of Rana catesbeiana and R. temporarta on which much of this chapter is based. It is doubtful that the first anuran tadpole wasstructurally or functionally similar to the pond tadpoles ofRana, and the reader is cautioned not to extrapolate structure or function from Rana to other

An ideal general description of tadpole anatomy would be based on a plesiomorphic pond tadpole such as Bombina or Discog-lossus.Unfortunately, the literature on these species is not rich enough to permit this, although that situation is changing with papers such as those by Haas (1997) and Schlosser and Roth (1995). The descriptive cranial anatomy herein is based primarily on Rana temporaria (De ~ o n ~ h 1968; Edgeworth 1935; Pusey 1938), and the functional aspects are based on Rana catesbeiana (Gradwell 1968, 1969a, b, 1970, 1972b, c; Gradwell and Pasztor 1968). Most of the terminology for the cranium derives from Gaupp (1893, 1894, 1896), and that for muscles comes from Edgewotth (1920,1935), with modifications by Haas (1997). Although Edgeworth's terminology has been almost standard for many years, recent work has altered the proposed homologies of muscles, especially in the branchial region. The description of metamorphosis is generally based on De Jongh (1968) and Ridewood (1897b, 1898a). The information for the vertebral column and associated muscles is derived from Xineus laevis because the literature on that s~ecies much more com~letethan for Rana. The most is I comprehensive summary of development of all organ systems is that of Nieuwkoop and Faber (1967) for Ximpw h s . Unfortunately, there are no anatomical illustrations, but there is an extensive literature section. Staging systems used in the text include Gosner (1960), Kopsch (1952), and Nieuwkoop and Faber (1967). In general, English anatomical terms are used if both English and Latin are available. If no English term is available, or if the English term is ambiguous in meaning, the Latin is used; synonyms are listed in parentheses. Edgeworth (1935) provided a table of muscle synonyms for the early literature.

The Chondrocranium
The first study of the frog chondrocranium apparently was by Dug& (1834), who described and figured the larval chondrocranium of Pelobatesfiscus; other literature on tadpole crania is summarized in table 4.1. Intertaxic variation in the larval chondrocranium is &scussed below and in Haas

54

DAVID CANNATELLA

(1993, 1995, 1996b, 1997a) and O. M. Sokol (1975, 1981b) The reader is referred to Cannatella (1985), Dueiirnan and Tmeb (1986), and Tmeb (1973, 1993) for variation of bones among taxa. The chondrocranium of tadpoles is a cartilaginous case that protects the brain and supports the sense organs and jaw apparatus. The chondrocranium initidy appears as two pairs of cartilaginous elernents, the parachordals and trabeculae (fig. 4.3). The parachordals appear as longinidinal rods flanking the anterior end of the notochord. The trabeculae form in the same plane but anterior to the parachordais; the notochord does not intemene between the trabeculae. The trabeculae se anteromedialiy to forrn a trabecdar

plate (ethmoid plate, planum trabecdae anterior). The space where the trabeciilae have not yet hsed is the basicranial (hypophyseai) fenestra, which disappears as the trabeculae coalesce to forrn the cranial floor (basii cranii). Fusion of the parachordais forrns the basal plate (planum basale; fig. 4.3) as the floor of the neurocranium; the planum rnay be perforated by the notochord. Before the basicranial fenestra disappears, the trabeculae and basai plate fuse. The junction between the basai plate and craniai floor is indicated by the prirnary carotid foramina (foramina carotica primaria) through which the internai carotid arteries pass. The parachordals project posteriorly to the otic capsules and support the occipital arch, whch is seridy hornologous with the ver-

Fig. 4.2. Evolution of Orton's (1953) tadpole types under the assumption of unordered states, based on the phylogenies of L. S. Ford and Cannateiia (1993), Haas (1997a), and Hay et al. (1995).

ARCHITECTURE

55

Table 4.1 Summary of iitenture concerning tadpole chondrocrania, muscles, and gi apparatus. Since t h s table was first compiled, similar compendia have appeared (Haas 1993, 1995, 1996b) that were of great help in updating the present table. Aii taxa treated in Haas (1996b) are also iisted in Haas (1993). Some species names may be incorrect according to current taxonomy. A&caphustmei LEIOPELMA Leiupelma archeyi Lewpelma hochstetteri BOMINATORIDAE Bombina bombina Bombina urientalis Bombina variegata DISCOGLOSSIDAE A&es obstetncans Gradwell1971a, 1973; Haas 1995,1996b, 1997a, b; Pusey 1943; Reiss 1997; Van Eeden 1951; Wassersug and Hoff 1982 N. G. Stephenson 1951b N. G. Stephenson 1955 Goette 1875; Ridewood 1898b (as B. &teus); Severtsov 1969b Haas 1997a; 0 . M. Sokol 1975; Wassersug and Hoff 1982 Haas 1995,1996b, 1997a De Beer 1985; Haas 1995,1996b, 1997a; Kmjitzer 1931; Magnin 1959; Pusey 1938; Ridewood 1898a, b; O. M. Sokol 1981b; Van Seters 1922; Wassersug and Hoff 1979, 1982 Kraemer 1974; Haas 1995, 1996b, 1997a; Pusey 1943; Ridewood 1898b; Schiosser and Roth 1995, 1997a, 199% Pgener and Maglia 1997; 0 . M. Sokol1981b Ridewood 1898b Haas 1996b; h j i t z e r 1931; Ramaswami 1943 Haas 1995 Ramaswami 1943 Ramaswami 1943 Haas 1995 Dugs 1834; Haas 1997a; Luther 1914; Plasota 1974; Ridewood 1898b; Rotek 1981; Schultze 1888, 1892; Severtsov 1969a O. M. Sokol 1975,1981b Haas 1995 O. M. Sokol 1981b; Wassersug and Hoff 1979, 1982 (as Scaphwpus);J. J. Wiens 1989 Ridewood 189%, 1898b; 0 . M. Sokol1981a, b O. M. Sokol 1959,1962,1977a O. M. Sokol 1977a Haas 1995; 0 . M. Sokol 1997a W K. Parker 1876 (as l? monstrosa);Ridewood 1897a, 1898b (as l? americana);Rozek 1989, 1990; . Rozek and Veself 1989 O. M. Sokol1977a, b (asXenupw) Dreyer 1915; Edgeworth 1930; Gradwell 1971b; Haas 1995, 1997a; Kotthaus 1933; W K. Parker 1876 . (as Dadyiethra cnpensis);Paterson 1939a; Ramaswami 1941; Ridewood 1897a, 1898b; Sedra and Michael1956a, 1957; Tmeb and Hanken 1993; Wassersug and Hoff 1979,1982; Weisz 1945a, 1945b Haas 1995,1996b; Orton 1943; 0.M. Sokol1975,1977a, 1981b; Swatt and De S 1998; Wassersug and Hoff 1982 Wassersug and Hoff 1979, 1982 Ramaswami 1944; Van Der Westhuizen 1961 C. M. Jacobson 1968 C. M. Jacobson 1968 Davies 1989 Haas 1996b, 1997a W. K. Parker 1881 Kesteven 1944 W. K. Parker 1881; Reinbach 1939; Ridewood 1898b (as Calyptocephalmgayi);O. M. Sokol 1981b Wild 1997 Fabrezi and Garca 1994; Lavilla and Fabrezi 1992 Lavilla 1991 Hanken et ai. 1992; Schiosser and Roth 199% Lynn 1942 Haas 1995,1996b; Ruibal and Thomas 1988 Lavdla and Fabrezi 1992 O. M. Sokol l 9 8 l b W. K. Parker 1881 Haas 1995,1996b

Discoglossus sardas MEGOPHRYIDAE Leptobrachium hasseltii Megophtys montana Megophtys nasuta Megophtyspania Megophtys robusta Megophtys stejnegmi PELOBATIDAE PelobatesfiLscus Pelobates syn6cus Scaphiopus holbrookii Spea bombtfions PELODYTIDAE Pelodytespunctatus PIPIDAE Hpnochims boettgeri P i p canialhoi P9apart.a P@apipa

RHINOPHRYNIDAE Rhinuphtynus donalis ATOBATRACHIA Heleophtyne natalensis Heleophyepurcelli .\IYOBATRACHINAE Pseudophtyne australis Pseudophtyne bibronii Uperoleia lithomodu LU'NODYNASITNAE Limvwdynmtesperonii Limvwdynbes tasmaniensis Mixophyesfasciolatus -LEMODACIYLIDAEn Alsodes bawioi Caudiverberamudiverbera Ceratophtys comuta Cwatophtys cranwelli Cychamphusstejnegmi Eiatherodactylus coqui Eieutherodac~lus nubicoiu Lepidobatrachus h k Lepidobatrachw liunensis I~ptodmtylus chaqwnsis LLptodmtylus ocellatus Leptodmtyluspentadadylus

DAVID CANNATELLA Table 4.1 umtinued Odontophvynusachalemis Physaiaemwpumlosus Pleurodema bibronii Pleurodema borellii Telmatobius bolavianus Telmatobiw ceimum Telmatobiw mleus Telmatobiusjelskii Telmatobius latkps Telmatobius cf. m a m a t u s Telmatobiuspisanoi PSEUDIDAE Pseudisparadoxa RHINODERMA Rhinoderma ahtwt'nii BUFONIDAE Bufo cf. americanus BufD amwr'canus BufD angusti'ceps Buf bufo Buf %ntiginosusn Bufo melanostictus Bufo regulanS Bufo cf. spinulosw BufD p i d h HYLTDAE Agabchnh callidym Anotheca spinosa Cycloranaplapphala F~ctonotusgoeidii Gastrotheca espeletia Gastrothecagracilh Gastrotheca cf. manupiata Gastrotheca manupiata Gastrotheca orophylax Gastmthecaperuana Gnrtmtbecapse21stes Gastrotheca riobambae Hyla arbma Hyla cinerea Hyla laruzfomtis Hyla microcephala Hyla nana Hyla pukcheh Hyla rosenbergi Hyla sguirella Phnrmabyhguttata Phyllomedusa boliviana Phylhedusa sauvagii Phylhedusa trinitatir Pseudacris cru& PseuMs regila Scinat:acuminata S c i m mbra ''WAE" Amolops af8hana Euphlyctis hexadaqlw Hoploba&chus tigerinus Occidazyga iuevis Rana catesbeiana Rana clamitans Rana mrtipes Rana ahlmatina Rana esculenta Rana cf.pipiens Rana pipiem Rana temporaria Rana "whitehendi" Haas 1997a Schlosser and Roth 199% O. M. Sokol 1981b Fabrezi and Garca 1994; Wassersug and Hoff 1982 Laviiia and De Ia Riva 1993 Fabrezi and Lavilla 1993 W. K. Parker 1881 W. K. Parker 1881 (as Cyclmhamphw culeus, fide Ridewood 1898b) Fabrezi and Lavilla 1993 Ridewood 1898b Fabrezi and Lavilla 1993 W. K. Parker 1881; Ridewood 1898b

W. K. Parker 1881 Wassersug and Hoff 1982 Barry 1956 Haas 1995, 1996b; W. K. Parker 1876; Ridewood 1898b (as B. rmbaris) Sedra 1951 Ramaswami 1940 Sedra 1950; Sedra and Michael1956b, 1958,1959 W. K. Parker 1881 O. M. Sokol1975 Haas 1995,1996b Haas 1996b; Wassersug and Hoff 1982 Ridewood 1898b (as Chimleptes) Haas 1996a Haas 1996b Fabrezi 1985; Fabrezi and Lavilla 1992 W. K. Parker 1881 Haas 1996b Haas 1996b Haas 1996b Haas 1996b Haas 1995,1996b, 1997a Haas 1996b; Ridewood 1898b; Severtsov 1969a Haas 1996b De S 1988 Haas 1996b Fabrezi and L a d a 1992 Fabrezi and Lavdia 1992; Lavilla and Fabrezi 1987 Haas 1996b O. M. Sokol1981b Fabrezi and Lavilla 1992 Fabrezi and Lavilla 1992 Fabrezi and Lavilla 1992 Haas 1996b; Kemy 1969a W. K. Parker 1881 (asAcvispicken'ngi) O. M. Sokol1981b (as Hyla) Fabrezi and L a d a 1992 Haas 1996b Ramaswami 1940,1943 (as Rana) Ramaswami 1940 (as Ram) Chacko 1965,1976; Ramaswami 1940 (as Rana) Ridewood 1898b (as Oyglossw) Gradwell 1968, 1970, 1972a, b; Starrett 1968 W. K. Parker 1881 Ramaswami 1940 Kratochwiil 1933 (as R agilh) Ridewood 1898b W K. Parker 1881 . Starrett 1968; Wassersug and Hoff 1979 De Beer 1985; De Jongh 1968; Gaupp 1893; Haas 1995, 1996b, 1997a; W. K. Parker 1871; Plasota 1974; Pusey 1938; Ridewood 1898b; 0 . M. Sokol1975; Spemann 1898; Stohr 1882 Ridewood 1898b

ARCHITECTURE Table 4.1 wntinued RHACOPHORIDAE Buw~eria buegeri Philautus variabilk Rhaeaphms ieucomystm: ,WCROJ5YLDAE Brevrkps aakpevsus Dennatonotus muelieri Grrrtmphyne mrolinensis Grrrtmphyne wta Hamptophlyne boliviana Hp~pdw bavberi Hpqachus vario[osw M k h y l a omata M k h y l a pukhva Otophynepybumi Phrynonauntk annectens Upmdon ystoma DENDROBATIDAE Caiostethusnubicola Caiostethw subpunctatus Caiostethrrshinztatis Dendrobates auratus Dendrobates tinctoriw Epipdobates anthonyi Epipdobates buuh~eri Eppedobates Phyhbates bicoior Okutomi 1937 Ramaswami 1938 Ridewood 1898b Swanepoel1970 Laviiia 1992b Wassersug and Hoff 1982 Haas 1995,1996b De S and Tmeb 1991 O. M. Sokol1975,1981b Starrett 1968 (as H. albooenter) Ramaswami 1940; Ridewood 1898b Haas 1995,1996b Wassersug and Pyburn 1987 (as 0. vobwta) Gradwell 1974 (as Phynomerrts) Ramaswami 1940 Haas 1995,1996b Haas 1995,1996b Rtdewood 1898b (as Phyhbates) De S and Hdl1998 Haas 1995,1996b Haas 1995,1996b Haas 1995,1996b Haas 1995,1996b Haas 1993,1996b, 1997a

57

Trabecular horn

,Trabecular

plate

Quadratocranial

,Trabecula

Notochord Parachordal cart.

F g . 4.3. Developing chondrocranium of Rana temporaria (7.5-mm lama) in dorsal view. Redrawn from Stohr (1882).

tebral neural arches. The occipital arch fuses to the rear end of the skull.

Etbmoid Region The anterior end of the primordial chondrocranium is elaborated to support the nasal capsuies and the larval jaws. Two
C

trabecuiar horns (cornua trabecuiarum) extend forward vent r d v and anterolateraiiv from the ethmoid = trabecuiar plati, fig. 4.3). The trabedar horns articulati with, or are fused to, a pair of suprarostral cartiiages (cartilago labialis superior, superior rostral carulage), the skeletal elements of th tadpole kpper jaw. In other gnathostomes, the functional upper jaws are formed either by the palatoquadrate cartilage or by a series of dermal bones (e.g., maxillae and premaxiiiae). In this region two ligaments, the quadratoethmoid ligament (ligamentum quadratoethmoidale, I. cornuquadratum mediale; Van Der Westhuizen 1961) and the lateral circumoral ligament (I. cornu-quadratum laterale, I. cornu-auadratum. I. circumoralis: 0.M. Sokol1981b) ioin the ethmoid regi& to the palatocguadrate cartilage; sie'details below under Circumoral Ligaments. The nasal septum (septum nasi; fig. 4.4) arises from the trabecuiar dati and s e ~ k a t e s naial ca~suies. the The nasal capsuies ofllarvae u s d y are partidy forked spheres. The roof of the nasal capsuie (tectum nasi) is confluent posteriorlv with the lamina orbitonasalis and is united to the seDtum nasi. The confluente of the trabecuiar horns joins the ventral part of the nasal capsuie and forms the floor of the nasal capsuie (solum nasi, trabecuiar plate, ethmoid plate; O. M. Sokol1981b). The anterior w d of the nasal ca~suie is the anterior cupola (cupola anterior). The nasal capsuie is separated from the orbit by a w d , the lamina orbitonasalis (planum antorbitale). The lamina orbitonasalis is confluent &th the tectum nasi anteriorlv and the anterior end of the orbital cartiiage (via the sphenethmoid commissure; see below) medidy. The muscular process of the palatoquadrate is bound to the processus antorbitalis (Haas 1995) of the larnina orbitonasalis by the ligamentum

58

DAVID CANNATELLA

teai, which has two parts, the I. teai superius and 1. tecti inferius. These iigaments may be chondrified in some taxa to form the cartilago tecti ("cartilago tectum" of Sokol 1975) or comrnissura quadrato-orbitalis (Haas 1995). As metamorphosis begins, the cartiiages and olfaaory epithelium of the nasal region become elaborated into a highly folded saclike structure; detaiis of these structures can be found in R. S. Cooper (1943), Denis (1959), Higgins (1920), Jurgens (1971), Rowedder (1937), and Swanepoel(1970). Ossifications associated with the nasal region include several well-known dermal bones and some neomorphic elements. The nasais form part of the roof of the nasal capsules. The vomers are palatal bones that form the floor of the nasal capsule; these are absent prirnitively in frogs (Cannatella 1985) and secondarily lost in several taxa. The septomaxillae are generally small elements that are intimately associated with the nasolacrimal duct but develop independently of the nasal cartilages (De Jongh 1968; Jurgens 1971). Neomorphic elements, such as the internasals and prenasals in various taxa of casque-headed frogs (Tmeb 1973), appear relatively late in ontogeny (Trueb 1985). Hanken and Hall (1984, 1988), N,E. Kemp and Hoyt (1969), Stokely and List (1954), and citations in Trueb (1985) provided ossiication sequences in variou5 species.
Diversity. The ligamentum tecti was figured for Pelodytes (O. M. Sokol1981b) andXenopush 5 (Sedra and Michael

1956a). Gradwell (1972b) dustrated it in Rana catesbeiana as the composite of the 1. supraorbitale cranii and 1. supraorbitalis ethmoidale. It is not mentioned in some taxa for which the chondrocranium is well described: Aseaphus, Heleophlyne, or Rana temporaria (De Jongh 1968; Van Der Westhuizen 1961: Van Eeden 1951). Variation in the lateral circumoral ligament is discussed below under Circumoral Ligaments. The ethmoid region of pipoids is difficult to compare with that of other frogs. For example, there is a single medial cartiiaginous bar that supports the nasal septum. 0 . M. Sokol (1975) argued reasonably that this bar represents the fused trabecular horns, but RoCek and Vesely (1989) considered this planum internasale to be nonhomologous with the trabecular horns. Thev also described several other features in which the ethmoidJregion of pipids differs from those of other frogs. The homology of ethrnoidal structures was also discussed by Haas (1993, 1995).

Orbital Region The lateral wall of the cranial cavity (cavum cranii) consists of an orbital cartilage (cartdago orbitalis; fig. 4.4) that arises from the posterior parts of the trabecular cartilages and the anterior ends of the parachordal cartilages. The orbital cartilage is joined to the floor of the braincase by several cartilaginous pillars; in some vertebrates the orbital cartilage arises as a discrete elernent, but it is not clear if this is so in frogs.
ParEtal fen. Taenia tecti rnedialis Anterior semicircular canal

Taenia tecti rnarginalis Orbital cart., \\

Pila antotica T r o c . f Oz!!om!?!orf.8,

- - - . -

1 r
O

sem. canal Posterior serni. canal

Quadratoethrnoid Trabecular horn Lateral circurnoral Arcus subocularis

'~ugu1ar.f. Operculurn Stapes Fenestra ovalis Otic proc. basitrabecular proc.

/ 1 I

Muscular proc. Ceratohyal

Articular Droc. I Meckel's ca,laie

Infrarostralcartilage

Fig. 4.4. Lateral view of the chondrocranium of a Rana temporaria tadpole (13 mm body length). Abbreviations: cart. men, fen. = fenestra, lig. = ligament, and proc. = process. Redrawn from Pusey (1938).

carrilage, f.

= fora.

ARCHITECTURE

Braincase and Otic Capsules The braincase of frog larvae and adults is open dorsay via the frontoparietai fenestra (fenestra frontoparietalis; fig. 4.5). The laterai rim of the fenestra is formed by strips of cartilage caed taeniae tecti marginaies, which are confluent with the orbital cartiiages ventray and with the tecturn anterius anteriorly. Posteriorly, the taeniae roof the gap between the otic capsules and form the tectum synoticum. The frontoparietai fenestra may be subdivided by additional stmts of cartiiage, as in Rama. The taenia teai transversalis is a transverse stmt that partitions the parietai fenestra posteriorly from the frontoparietai fenestra anteriorly. The taenia tecti medialis, if present, extends from the taenia tecti transversalis to the tectum synoticum and subdivides the parietai fenestra into left and right parietai fenestrae. These tectai cartilages persist in adults. The occipitai arch forms the posterior end of the cranium; it is seriaiiy homologous with the neurai arches of the vertebrae. Frogs have only one occipitai arch; most other vertebrates have more than one (De Beer 1985). There are no pre-occipitai arches in frogs (De Beer 1985), but see Kraemer (1974). The occipitai arch is supported ventraiiy by the parachordais; early in ontogeny it fuses to the basai plate and otic capsules. The tectum posterius joins the halves of the occipitai arch dorsaiiy and fuses anteriorly to the tectum synoticum. The ventral part of the occipitai arch bears the occipitai condyles, which articulate with the cotyles of the atlas. The otic capsules (capsulae auditivae; figs. 4 . 4 4 . 5 ) are subsphericai stmctures that adjoin the posterior end of the braincase. Ontogeneticay, they begin as hemispheres that are open mediay. The hemispheres fuse to the parachordais and attach to the basai plate ventraiiy, the tectum synoticum dorsay, and the taeniae tecti marginaies anterodorsally. Within these capsules the anterior and lateral semicircular canals (canalis semicircularis) appear and slightly later the posterior semicircular canals form. Each otic capsule develops a flattened crest, the crista parotica, that extends lateray from the outer wa of the lateral semicircular canal. The posterior part of the ear capsule forms the processus muscularis capsulae auditivae for muscle attachment. The otic ligament joins the anterior part of the crista parotica to the otic process (processus oticus) of the palatoquadrate (see below unDiversity. The pila antotica is absent in rnicrohyiid larvae der Jaws). O . M. Sokol 1975) and Rhirwphrynus (Tmeb and CannaThe dorsai features and general size of the larvai otic capd a 1982). In the basai clades of frogs (Ascaphus, Leiupelma, sules are determined by the three semicircular canals of Diw~lussus,and Bombina) the prootic foramen is subdivided the inner ear. Chondrification and later ossification around by a strip of cartiiage caed the prefaciai commissure into these, especiaiiy the anterior and posterior canals, are evident the prootic foramen (for craniai nerve V) and palatine fora- as the epiotic eminentes. Metamorphic changes in the inner men (for cranial nerve VII; O. M. Sokol 1975). In taxa with ear are minimal; those of the middle ear are pronounced. the prefacial commissure, the trigeminal and facial gangiia The middle ear of adult frogs includes an opercularis system are unfused, whereas in those lacking the commissure the and usuay a tympanic-columeiiarsystem; the Merences in _pnglia are fused to form a prootic gangiion. As suggested developmental timing of these two systems suggest that they by Sokol(1975),the lack of the commissure is a synapomor- are h a i o n a y independent (Hetherington 1988). Both of phy of Pipanura. Other aspects of variation were discussed these communicate with the inner ear via the fenestra ovalis, by Haas (1993). a prominent opening in the ventrolateral wa of the otic capsule that is covered by a membrane in the tadpole. In the adult, the fenestra ovalis is largely occluded by two stmc-

The anteriormost of these piiiars is the preoptic root forming the anterior margin of the optic foramen and attaching to the trabecula. In frogs and salamanders the preoptic root forms an extensive cartilaginous waii between the optic foramen and the sphenethmoid commissure; in caecilians (M. H. Wake and Hanken 1982) it is more of a stmt. The pila metoptica grows upward from the trabecula and separates the optic foramen from the oculomotor (metoptic) foramen, the latter for the oculomotor nerve (111), ophthalmic artery, and pituitary vein. The pila antotica (pila prootica) arises bom the parachordais and separates the oculomotor and prootic foramina. A fourth, more transitory stmt, the pila ethmoidaiis, forms the mediai was of the nasai capsules and olfaaory canals (canales olfactorii, forarnina olfaaoria evehentia) and merges into the nasai septum (De Beer 1985). Posteriorly, the orbitai cartiiage joins the otic capsules via the raenia teai marginalis. Anteriorly, the sphenethmoid commissure conneas the orbitai cartiiage to the lamina orbitonas a h and the nasai capsule. The orbitonasai foramen (foramen orbitonasaiis; fig. 4.4) demarcates the lamina orbitonasalis from the orbitai cartilage; the p r o h d u s nerve (ophthalmic branch of V) passes out of the orbit into the nasai cavity through this foramen. The optic foramen (for craniai nerve 11)is the largest of the foramina of the orbitai cartiiage and is located posteriorly in the orbitai cartiiage. The oculomotor foramen (for cranial rime 1 1 is smaer and lies just posterior to the optic fora1) men. The tiny trochlear foramen (for craniai nerve IV) is k t e d dorsai to the two aforementioned foramina near the raenia tecti marginaiis. The prootic foramen (foramen prooticum) more or less separates the orbitai cartiiage from the otic capsule. The trigeminai (V) and faciai nerves (VII; from the combined prootic gangiion), the abducens nerve (VI), and the laterai cephalic vein (vena capitis lateralis) pass rhrough the prootic foramen. ~s&catins in this retzion include the valatines and sphenethmoid (orbitospheioid). The palatinis form along the larnina orbitonasaiis and usuay extend from the maxilla mediallv to near the midline. The svhenethmoid. usuav an unpaired bone in adults, ossifies as paired elements along the anterolaterai wa of the braincase anterior to the optic foramen.

62

DAVID CANNATELLA

tive tissue that is associated with the processus posterior dorsalis of the pars alaris. Haas (1995) described the details of its connection to the mandibulosuprarostral iigament and the lateral circumoral ligament (iigamentum cornuq u a d r a m ) in dendrobatids. ~e noted its presence (apoe morphic) only in T j ~ 4 tadpoles. I interpret its absence in microhylid larvae ( T p e 2) as a loss. Its absence in pipoids (Type 1) may aiso be a loss or may indicate that pelobatoids are more closely related to Neobatrachia than to pipoids, resulting in paraphyly of Mesobatrachia. Noble (1929) - .. claimed-that &e absence of the suprarostral cartilage was characteristic of microhylid (his "brevicipitid") larvae. Starrett (1973) reported that the suprarostrai was lacking i11pipoid ( T p e 1) and microhylid (Type 2) larvae; t h s was linked to her view that the absence of jaw sheaths was pleiomorphic for anurans. 0.M. Sokol (1975) pointed out that all frog larvae have suprarostral cartilages; in Types 1 and 2 larvae they are firmly ankylosed to the trabecular horns. The presence of keratinized jaw sheaths may be correlated with mobility of the suprarostral. The posteroventral process of the suprarostral (0. Sokol 1981b) is absent in Pipanura. M. Alytes and DWcoglossus have long, distinctive, U-shaped corpora ( 0 . M. Sokol1981b). The lateral tip of the suprarostral Diate L Z n o ~ u uevis fuses with the ~uadratoethmoid I s cartilage to form the tentacular cartilage, which supports the tentacle (Tmeb and Hanken 1993).
L

Palatoquadrate and Suspeitsorium The larvai palatoquadrate (pterygoquadrate; figs. 4.3 and 4.6) is a long flat strip of cartiiage anchored to the neurocranium by four cartilaginous processes: ascending process (processus ascendens), basal process (processus basalis, processus pseudobasalis, hyobasai process), otic process (processus oticus), and the quadratocranial commissure (commissura quadratocranialis anterior). The term suspensorium was used in its present sense by T. H . H d e y (1858) for the common attachnient of the mandibular and hyoid arches to the skuil (Pyles 1987). According to O. M. Sokol (1981b), the suspensorium is the posterior end of the palatoquadrate that iies at the anterior end of the otic capsule. It articulates with the otic capsules via the ascending and otic processes. The ascendmg process is a round bar that curves mediaiiy and fuses with the region of the pila antotica (see Diversity below). The otic process is the lateral part of the suspensorium; it is a thick plate that is curved almost verticaiiv. The otic Drocess is attached to the wall of the lateral seniicircular canal near the crista parotica by the otic ligament; in some taxa this process chondrifies and is then caiied the larvai otic process, which is considered homologous with the adult otic process (Van Der Westhuizen 1961; Van Eeden 1951). The larval otic process is destroyed during metamorphosis, but the ceiis released in the process rechondre to form the adult process. Ln most vertebrates the basal process of the palatoquadrate articulates with the basitrabecular process (fig. 4.4; a lateral projection of the trabecula) anterior to the palatine nerve. However, in most frogs the basitrabecular process is lost during ontogeny, and the basai process articulates or fses with the neurocranium posterior to the paiatine nerve

(De Beer 1926). Because of the position of the process relative to the palatine nerve, Pusey (1938) rejected the homology of the two conditions; the newer process and its postpalatine connection have been termed the pseudobasal process (De Beer 1926) and pseudobasal connection (Pusey 1938). Van Der Westhui7~n(1961) offered compehng evidence that the basal process simply has moved from the basitrabecular articulation to a more lateral position on the otic ledge to produce a postpalatine connection (commissure). Therefore, the basal and pseudobasal processes are homologous (Pyles 1987; Reiss 1997). As one proceeds anteriorly, the paiatoquadrate underlies the orbit and in this region is named the arcus subocularis (fig. 4.4). The wide gai that separates the paiatoquadrate from the neurocranium is the subocular fenestra. The region anterior to the a r a s subocularis is termed the quadrate (O. M. Sokol 1981b). The most prominent part of the quadrate is the muscular process (processus muscularis), a flattened triangular process that curves dorsally and medially and is the site of attachnient for the orbitohyoideus and suspensoriohyoideus muscles. The muscular process is bound to the processus antorbitalis (Haas 1995) of the neurocranium by the ligamentum tecti (cartilago te& or commissura quadrato-orbitalis when chondrified), which may be composed of two ligaments (see Ethmoid Region above), the ligamenti tecti superius et inferius. Anteromedially, the quadratocraniai commissure (figs. 4.3 and 4.5) rises dorsally as a thick cartilaginous strip to join the quadrate to the floor of the braincase near the ethmoid plate. On the anterior margin of the commissure is the s m d triangular quadratoethmoid process (processus quadratoethmoidahs) to which the quadratoethmoid ligament (ligamentum quadratoethmoidale) is attached. A smail processus pseudopterygoideus (Haas 1995) projects from the posterior part of the cornrnissure into the fenestra subocularis. The hyoquadrate process (processus hyoquadrati) is a condyle for the synoviai articulation with the ceratohyal (figs. 4.4 and 4.7) and is located ventrally at the posterior part of the quadrate region. The hyoquadrate ligament attaches to a ridge posterior to the hyoquadrate process and binds the ceratohyai to the quadrate. The most anterior part of the quadrate is the stout articular process (pars articularis quadrati), which forms a synovial articulation, the adult jaw joint, with the posterior end of Meckel's cartiiage. During metamorphosis, the arcus subocularis, ascending process, hyoquadrate process, and muscular process are resorbed as the paiatoquadrate rotates posteriorly (fig. 4.6). The quadratocranial cornniissure is partly destroyed, but its surviving part, with the quadratoethmoid process and processus m d a r i s posterior of the lamina orbitonasalis. forms the pterygoid process (processus pterygoideus) of the adult. Most of the larval paiatoquadrate disappears at metamorphosis, and some of its supportive function is assisted or taken over by severai ossifications: m d a , p r e m d a , quadrate, quadratojugai, pterygoid, and squamosai. The auadrate is an endochondral ossification that forms in the similarly named region of the palatoquadrate at the articular process, the point of articulation with the lower jaw. An arc of the p r e m d a e , m d a e , and quadratojugals takes

ARCHITECTURE

63

Palatoquadrate

Infrarostral caltilage

Meckel's cartilage

Fig. 4.6. Four arbitrary stages i the metamorphosis of the paiatoquan drate of a generalized tadpole; lateral view. Arrow indicates direction of metamorphic shift of the jaw elements. After Sedra (1950).

over the role of the palatoquadrate in forming the upper jaw of the addt frog. The squamosal forms in association with the otic process and quadrate of the palatoquadrate cartilage. The pterygoid invests the addt pterygoid process, which forms in part from the quadratoethmoid process and the basal process.
Diversity. Several features of the palatoquadrate are highly m d e d in pipoids ( 0 . M. Sokol 1977a; Swart and De S 1998). A nove1 structure, the ventrolateral process, is a synapomorphy of Pipoidea; it is present in Pipa, Rhinq&ynus, Silurana, and Xenopus but not Hymenochirus. The muscular process and arcus subocdaris are absent in H. boetcgeri 1 0 .M. Sokol 1962). The arcus subocdaris is tenuously complete in Xenupus laaXr (Trueb and Hanken 1993). In pipids, the otic process does not reach the anterior end of the otic capsde ( 0 . M. Sokol1975), and the ascending and otic processes are h e d into a broad plate ( 0 . M. Sokol 1977a). The connection of the ascending process to the neuroaanium is more dorsaliy placed (a "high" attachment) in basal taxa such as Ascaphus and Bombina and more ventraliy placed ("low") in neobatrachians ( 0 . M. Sokol 1975). In microhylids, the ascending process has a low attachment directly to the trabecula because the pila antotica is absent i O. M. Sokol1975). The ascending process is absent in larvae of Heleophlyne, Otqhlyne, and Philautus (see Haas 1995). The ascending process is lost during metamorphosis in ali frogs except Ascaphus tmei (Van Eeden 1951); Reiss i 1997) disagreed with this interpretation, claiming that the persistent ascending process in Ascaphw is oniy part of &e neurocranium. Ascaphus i is also unique among -inura in retaining the "true" basal process in which the connection to the neurocranium is anterior to the palatine nerve i De Beer 1985; Pusey 1938). The condition in Lewpelma is somewhat intermediate (Pusey 1938; E. M. T. Stephenson 1951), and ali other frogs surveyed have the pseudobasal process. Pyles (1987) summarized the issues regarding the homology of the pseudobasal/basal process. Reiss's (1997) intensive study of the larva of Ascaphus reached somewhat

different conclusions about the phylogenetic significance of certain suspensorial characters than did L. S. Ford and Cannateiia (1993), but because these characters are germane primariiy to addts, a U e r critique of Reiss (1997) wiii not be undertaken here. In Leupelma archeyi, which has nidicolous development (Altig and Johnston 1989), the larval otic process remains unbroken through metamorphosis (N. G. Stephenson 1951b). Heleophlyne purcelli has an apparently unique quadrato-oticligament that attaches the rear end of the palatoquadrate to the floor of the otic capsde (Ramaswami 1944; Van Der Westhuizen 1961); this is distinct from the connection of the larvai otic process to the otic capsdes. Orton (1943) figured a cartilGe of unknown hokology in a RhinUphlynus donalis tadpole, which was identified as a free hyoquadrate process by O. M. Sokol(1975: fig. 12). Ossification of the quadrate is known in onlv a few taxa of frogs (e.g., Ascaphus truei), and quadratojugais are absent in many taxa. The maxillae, premaxillae, pterygoids, and squamosals are present in ahost ali taxa (Tnieb 1993). Four principal ligaments are associated with the jaws (fig. 4.4; see aiso fig. 4.16): (1) the quadratoethmoid ligament (ligamentum quadratoethmoidaie,I. cornu-quadraturn mediale; Schulze 1892), (2) lateral circumoral ligament ( 0 . M. Sokol 1981b) (I. cornu-quadraturn; I. cornu-quadram laterale; 1. circumorale; trabecular quadrate li&ment; Pusey 1943), (3) mandibdosuprarostral ligament (I. rostraie superius cartilago Meckeli of Gradweii 1968), and (4) the suirarostrai-quadrate ligament (I. rostrale superius quadrati; Gradweii 1968). The quadratoethmoid ligament extends from a process (processus lateralis trabecdae; Haas 1995) on the posterolateral aspect of the trabecdar horn to the quadratoethmoid process of the quadratocranial commissure. This ligament lies ventral and lateral to the internal naris. The lateral circumoral ligament passes from the tip of the trabecular horn to the anterolateral margin of the pars articdaris quadrati. The 1. mandibdosuprarostraiearises from the insertion of the m. levator mandibdae ~osterior su~ericialis the dorsal suron face of Meckel's cartilage and passes to the posterior dorsal process of the suprarostral or adrostral. The 1. rostrale superius quadrati connects the posterior dorsal process of the suprarostral to the anterolateral margin of the pars articdaris quadrati near the attachment of the lateral circumoral ligament.
Diversity. The quadratoethmoid ligament seems to be universaliy present. The mandibdosuprarostral ligament is absent in Ascaphus, microhylids, and pipoids (Haas 1995); in Type 4 tadpoles it attaches to the adrostral tissue mass rather than to the suprarostral cartilage as in discoglossids (Haas 1995). The lateral circumoral ligament attaches to the suprarostral rather than the trabecdar horn in Alytes, Ascaphw, Bombina, and Discoglossus (0.M. Sokol 1981b). Pipids have lost the lateral circumoral ligament but Rhinophlynus apparently retains it in the discoglossoid condition (0. Sokol M. 1981b).

64

DAVID CANNATELLA

h e rJaws The lower jaws consist of paired Meckel's cartilages and paired infrarostral cartilages (infralabial cartilages, mentomeckelian cartilages) joined at the midline by connective tissue (fig. 4.5). The infrarostrals are placed posterior and slightly lateral to the suprarostrals. The two infrarostrals are connected by a slender cartilage, the commissura intrarnandibularis (fig. 4.5). In Rana, a second, condylar articulation is present ventral to the cornrnissura, but it is not synovid. Although the infrarostrals have been termed mentomeckelian cartilages (0. M. Sokol 1975), the extent of these is much greater than the mentomeckelian bones (J. J. Wiens 1989). Meckel's cartilages are generally stout and shallowly V-shaped; they are not associated with the lower jaw sheath. The medial end of Meckel's cartilage generally bears a cotyle for articulation with the infrarostral cartilages. The articulation with the pars articularis quadrati of the palatoquadrate is saddle-shaped and subterminal; thus, there is a posterolaterally directed retroarticular process on which the suspensorio-, hyo-, and quadratoangularis muscles insert (0. M. Sokol1981b). The articulation with the palatoquadrate has a well-developed synovial capsule. The bones associated with the lower jaw are the paired angulars (angulosplenials), articulars, dentaries, and mentomeckelians. The endochondral mentomeckelian bone ossifies in the medial ends of the larval infrarostral cartilage. The dentary and angular are elongate dermal bones that invest Meckel's cartilage. The angular is posterior and medial and the dentary anterior and lateral to Meckel's cartilage. The ar-

ticular ossifies endochondrally in the posterior end of Meckel's cartilage at the jaw articulation.
Diversity. The infrarostrals of R r c a p h are splintlike elements (Van Eeden 1951). In pipoids, the shape of the lower jaw resembles the adult co~guration-slender and gently curved. The angular ossifies earlier than in nonpipoid frogs (0. M. Sokol 1977a; Trueb 1985). The infrarostrals are fused medially in some microhylids (De S i and Trueb 1991; 0.M. Sokol 1981b) and apparently in Huplobatrachus tigerinus (Ramaswarni 1940). Fused infrarostrals (0. M. Sokol 1975) are a synapomorphy for Pipoidea. A syrnphysial joint later develops between the infi-arostrals during metamorphosis, but mentomeckelian bones do not develop in Pipoidea. Separate, so-called sub-meckelian cartilages are known in Heleophryne (Van Der Westhuizen 1961), and Alytes (called pararnandibularia; Van Seters 1922); these cartilages are not in the same position in Alytes o b s t (per-~ ~ sonal observation) as in Heleuphryne.

B d Flow Muscles Most of the cranial muscles of premetamorphic larvae are found in the adult frog, albeit with different names at times. However, all muscles are transformed during metamorphosis as the larval fibers die and adult-type fibers develop from myoblasts (Takisawa et al. 1952a, b; Takisawa and Sunaga 1951). Gradwell and Walcott (1971) discussed the dual functional and structural aspects of these fiber types in the interhyoideus muscle. These muscles are summarized in table 4.2. In a premetamorphic tadpole of Raw tempmaria at Kopsch stage 25 (Kopsch 1952), the muscles of the buccal

Ceratobranchial IV

Terminal commissure

Fig. 4.7. Ventral view of the gill arch skeleton of a Rana catesbeiana tadpole. Symbols: dashed l n s = lower ie jaw. Afier Gradwell (1972a).

ARCHITECTURE
Table 4.2 Origins and insertions of selected cranial muscles in a Gosner stagc 35 tadpc)te of Ranacatefbeiam, modified from Gradwell 1972b. Terminology is based on Ram t m p u r i a (Edgeworth 1935) as modified by Haas (1997a); CB = ceratobranchial. Muscle Origin Mandibular Group
-

65

Insertion

Levator mandib~dae posterior superficiaiis L. m. p. profundus L. m. anterior

Dorsal part of palatoquadrate Dorsal part of palatoquadrate Dorsal part of ascending process of palatoquadrate; otic capsule Anterior part of muscular process of palatoquadrate Dorsal, lateral part of palatoquadrate Anterior, medial surface of muscular process of palatquacirate (appears at stage 40 in Rana catesbeianu) Anterior, medial surface of muscular process of palatoquadrate (appears at stage 39 in Ram caesbeiana) Interconnectsinfkarostrals; no median raphe Median raphe Anterior part of Meckel's cartilage Hyoid Group Lateral part of ceratohyd Inferior, lateral part of muscular process of palatoquadate Superior, lateral part of muscular process of palatoquadate Lateral part of muscular process of palatoquadate Posterior, lateral part of muscular process of palatoquadate Median raphe Median raphe
-

Dorsal, medial part of Meckel's cartdage Lateral part of suprarostral cardage Dorsal, lared part of Meckel's cartilage Lateral part of suprarostral cartilage Dorsal, lateral part of Meckel's cartilage Lateral pr of suprarostral cartilage at Dorsal, lateral part of Meckel's cartilage Medial part of Meckel's cartilage Lower lip

I-. m. externus L. m. a. articularis L. m. a. subexernus L. m. a. lateralis

Intermandibularis anterior Intermandibularis posterior Mandibulolabialis

Orbitohyoideus Suspensoriohyoideus Interhyoideus Interhyoideusposterior

Retroarticular process of Meckel's cartilage Retroarticular process of ~Meckel's adage RetroarticuIar process of Meckel's cartilage Lateral part of ceratohyal Lateral part of ceratohyal Lateral part of ceratohyal Palatoquadrate + otic process + diaphragm

Branchial Group Levator arcus branchiatis I

I a. b. I1 .
L. a. b. I 1 1 L. a. b. IV
Constrictor branchidis I C. b. I1 C. b. 1 1 1 C. b. N Subarcualis rectus I, dorsal head S. r. 11, ventral head S. r. 11-IV Subarcualis obliquus Transversus ventralis IV T>mpanopharyngeus Diaphragmato-branchialisIV LTnnarnedmuscle (Cucullaris?)

Lateral part of palatoquadrate Otic process Auditory capsule Auditory capsule Not present in Ram; see text medial part of CB I1 Medial part of CB I1 medial part of CB 1 1 1 Posterior, medial part of ceratohyal Posterior, medial part of ceratohyal Medial part of CB I1 Crista hyoidea Not present in Ram; see text Auditory capsale Medial part of diaphragm Lateral part of CB N Spinal Group

Lateral part of CB I Lateral part of CB I Anterior, lateral part of CB I1 Posterior, lateral part of CB II Tenninal commissure of CB I and I1 Terminal Commissure of CB I1 and 1 1 1 Terminal Cornmissure of CB I1 and 111 Medial part of CB I Medial part of CB II Medial part of CB IV Medial part of CB I1 Lateral part of CB IV Terminal commissure of CB I1 and I11 Scapula

Geniohyoideus Rectus c e ~ c i s

Hypobranchid plate Medial part of diaphragm; continuation of rectus abdorninis

Infrarostral cartilage Medal part of CB 1 1

ARCHITECTURE

67

coili) lies posterior to the interhyoideus; its haives join at a mediai raphe and extend laterdy to attach to the posterior paiatoquadrate, the otic process, and the lateral part of the diaphragm. The larvai diaphragm is the posterior w d of the sinus cervicaiis, or pericardial space, surrounding the heart. In the adult frog these muscles are expanded to form a more or less continuous muscular floor for the buccai cavity. The geniohyoideus is a paired, thm, elongate muscle (figs. 4.84.9) that originates on the hypobranchal plate and inserts on the infrarostrai cartdage near the symphysis. In Rana temporaria it consists of a pars lateralis and pars medialis innervated by spinal nerves that comprise the hypoglossai nerve. The rectus cervicis muscle (sternohyoideus, diaphragmato-branchalis medialis), whch extends from the branchial process of ceratobranchial I1 to attach to the larval diaphragm, is also suppiied by the hypoglossai nerve. The tongue muscles proper, the genioglossus and hyoglossus, have not yet formed, but in Rana temporaria these diierentiate as the foreiimbs are emerging (De Jongh 1968).
Diversity. Differentiated portions of the intermandibularis form supplementary elements (Emerson 1976; Tyler 1971)

in various taxa, but these are not apparent until near metamorphosis. Starrett (1973) reported that in microhylid tadpoles the intermandibularis posterior inserts on the basihyai (copula 11) rather than a mediai raphe. She also reported an infraiabial retractor muscle that joins the infrarostrai cartilage to Meckel's cartilage in microhylids. Presumably it retracts the lower jaw and closes the mouth. Starrett (1973) speculated that t h s muscle was homologous with the mandibulolabialis of Types 3 and 4 tadpoles. In several species a mandibulolabialis superior inserts onto the upper iip and a mandibulolabialis inferior inserts onto the lower iip (Carr and Altig 1991). The mandibulolabiaiis is apparently absent in some Type 1 larvae; it is present in Gastrqpblyne carolinensis and Microbyla pulchra and absent in Hoplophlyne rogersi and Pblynomantis (= Pblynomerus) annectens (Gradweii 1974; Noble 1929). Gradweii (1972b, c) pointed out that the interhyoideus posterior muscle (H3b in his terminology) occurs in Pelobates, Pseudis, Rana catesbeiana (contrary to Edgeworth 1935), and R. fiscigztla but not in Ascapbus, Pblynomantis, pipoids, or Scapbzopus. Gradweii (1974) later diustrated and described an extensive interhyoideus posterior in Pblyno-

ideus

~iaphragmato-branchiali; ~ympano~hafyn~euS ~ e c t u cervicis

Fig. 4.9. Ventraiview of the musdes of the jaws and branchiai apparatus of an idealized tadpole, based on Gradwell (1972a), with additions.

68

DAVID CANNATELLA

mantis annectens. Noble (1929) mentioned that an extensive subbranchialis (interhyoideus posterior) covering the ventral w d of the branchiai chamber in Gactrophlynecarolinensis and Micvohyh pulchra is absent in two species of HoplUplnyne. O. M. Sokol (1975) reported that the interhyoideus posterior is absent in bufonids and leptodactylids and present in some hylids, microhylids, and some pelobatids. Gradweii (1972b) ais0 claimed that the supericial and deep layers of the interhyoideus posterior of Rana catesbeiana were identical to the subbranchialis and diaphragmatopraecordialis muscles (respectively) described by Schulze (1892) for Pelobates. Haas (1997a) pointed out that the diaphragmatopraecordialis muscle of Pelobates is a specialized part of the interhyoideus posterior. 0 . M. Sokol (1975; see fig. 4.9) iilustrated the superficial and deep layers of the interhyoideus posterior of Pelobates syriacus. My comparison of the illustrations in Gradwell (1972b), Schulze (1892), and O. M. Sokol (1975) supports Gradwell's contention. Muscles with similar names were described by Noble (1929) in the suctoriai tadpoles ofAmolUpsricketti. Haas's (1997a) phylogenetic anaiysis indicates that the presence of an interhyoideus posterior is a synapomorphy of Pelobates + Neobatracha, a clade that would render Mesobatrachia (fig. 4.1) paraphyletic. However, i the interhyf oideus posterior is absent in Scaphzopus as claimed by Gradweii (1972b, c), and assurning that Pelobatidae (Scaphzupus + Pelobates) is monophyletic, then homoplasy in this character makes the putative synapomorphy for Pelobates + Neobatracha ambigous. Noble (1929) described a pars locomotorius of the subhyoideus (interhyoideus) that inserts bilaterdy into paired cutaneous flaps flanking the mediai spiracle in HoplUphlyne. Noble speculated that these flaps served as locomotor organs in the damp leaf crevices in which these nonaquatic larvae hve. Edgeworth's (1935) table of muscle synonyms lists the constrictor coiii muscle in frogs as a synonym of the subbranchaiis, but his text does not mention the constrictor colli. Jaw Levator Muscles There are two morphologicai units of levator (adductor) muscles. One group consists of three levator muscles that

originate from the posterior portion of the paiatoquadrate and adjacent otic capsule and extend fonvard through the floor of the orbit mediai to the muscular process of the paiatoquadrate to insert onto the suprarostral and Meckel's cartilage. The second group originates from the anterior mediai region of the muscular process and extends ventromedidy to insert on Meckel's cartiiage and the suprarostral. A of these muscles are innervated by the trigeminai nerve (V). The terminology of these muscles is particularly confused (table 4.3), and Starrett's use of the path of the mandibular brandi (V,) of the trigeminai nerve for distinguishmg anterior, posterior, and externus groups of adductor muscles is not used here. The most supericial muscle (fig. 4.10) in the first group is the levator mandibulae posterior superficialis. It originates on the dorsai surface of the palatoquadrate laterai to the ascending process and extends fonvard to insert on a rounded process on Meckel's c a d a g e lateral to its articulation with the drarostrai cadage. The levator mandibulae posterior profimdus lies just below the I. m. p. superficialis and originates on the dorsai surface of the paiatoquadrate just below and slightly laterai to the origin of the I. m. p. supericialis. In Rana temporaria the I. m. p. supericialis and 1. m. p. profundus are fused through the posterior haifof their lengths (De Jongh 1968). The I. m. p. profundus inserts via a compound tendon on the ventrolateral edge of the suprarostral. This compound tendon aiso receives the insertion of the I. m. externus complex in &na temporaria but apparently not in R. catesbeiana (Gradwell 1972b). The deepest and most mediai of the three muscles traversing the orbit is the levator mandibulae anterior (fig. 4.10). It originates from the anteroventrai surface of the auditory capsule and the ventral aspect of the ascending process of the paiatoquadrate; the area of origin is mediai to that of the I. m. posterior superficialis and I. m. posterior profimdus. Its fibers converge into a distinctive long tendon that inserts on the dorsai side of Meckel's cartilage just mediai-to the jaw joint. The I. m. anterior lateralis differentiates from the I. m. anterior during metamorphosis in &na temporavia and at Gosner stage 39 in R. catesbeiana (Gradwell 1972b). The second group of muscles arises from the muscular process of the paiatoquadrate (fig. 4.10): the levator mandi-

Table 4.3 Comparisons of the terminology of jaw muscles in tadpoles (adapted from Starrett 1968) of Edgeworth 1935, Starrett 1968, Luther 1914, Sedra 1950, and the terminology for adults from Edgeworth 1935 Edgeworth Starrett Luther Sedra L. rn. anterior Edgavorth (Adult) Levator rnandibulae anterior

A. m. pterygoideus Levator mandibulae anterior Adductor rnandibulae anterior internus A. m. anterior longus A. m. posterior longus L. m. posterior superficialis A. m. posterior superficialis superficialis A. rn. posterior profundus A. m. posterior longus L. m. posterior profundus profundus A. rn. posterior subexternus L. m. anterior subexternus A. m. posterior subexternus A. m. posterior articularis A. rn. posterior articularis L. m. anterior artidaris A. m. externus superficialis A. rn. externus L. m. externus
L. m. anterior lateralis

L. rn. posterior superficialis L. m. posterior

L. rn. posterior profundus

L. m. posterior L. m. anterior subexternus

L. m. anterior artidaris L. m. externus L. m. anterior lateralis

A. m. externus lateralis

A. rn. posterior lateralis

L. m. externus pars anterior L. m. externus pars posterior L. m. anterior lateralis

70

DAVID CANNATELLA

ified drawing in which the pars posterior was no longer pres- to the insertion of the suspensorioanguiaris. The quadratoent, and he remarked that h s pars anterior was homologous anguiaris originates on the ventrolaterai aspect of the paiatowith the 1. m. externus and that the I. m. subexternus (pre- quadrate, is bounded laterdy by the bodies of the hyoangusumably his pars posterior) did not appear u n d stage 40 in laris and suspensorioanguiaris, and inserts on a common Rana uztesbeiana. The distinction between the I. m. externus aponeurosis with the hyoanguiaris. These three muscles and I. m. subexternus is particularly important because Star- form part of the depressor mandibulae of the adult. rett (1968) defined character states for the levator muscles based on the relationship of the mandibular branch of the Diversity. Starrett (1968) reported the occurrence of these muscles in many taxa, but these data are inconsistent with trigeminal nerve (V,) to these muscles. The fates of the larvai mandibular and hvoid muscles of some other reports, including Starrett (1973). For example, Bufi regularis, Rana temporaria, and Xenopus lmis were dis- pipoids have a single mandibular depressor muscle, the cussed by Sedra (1950) and Sedra and Michael (1957). quadratohyoangularis, which is thought to represent fusion Metamorphosis in Xenopus appears to be very different from of the three muscles ( 0 . M. Sokol1962,1977a); ontogenetthat in the other two taxa, but this may be caused by rnis- icdy, it appears as a single muscle in Xenopus laevis (Nieuwidentifications of certain muscles. Sedra and Michael (1957) koop and Faber 1967). However, Starrett (1968) reported claimed to foliow Edgeworth's (1930) muscle terminology both the hyoangularis and quadratoanguiaris present in for Xenqus lami, but their treatment is problematic. For ex- Xenopusgilli and X. l m i . The hyoanguiaris is said to be abi ample, they named the largest and most superficial of the sent inhcaphus ~ e (Pusey 1943; Starrett 1968) and Dismandibular muscles the 1. m. anterior and divided it into the coglosuspictus (Starrett 1968), but Starrett (1973) reported I. m. anterior pars intermedius, pars lateraiis, and pars medi- it present in Type 3 tadpoles. aiis. The 1. m. anterior pars medialis, on the basis of its toHaas (1997a) noted that the origin of the suspensorioanpography and its long, slender tendon, corresponds to the guiaris from the paiatoquadrate is behind the orbit in Alytes, I. m. anterior ofRana temporaria. The 1. m. anterior pars in- Ascuphus, Bombina, and Discoglosus (state O of his character termedius corresponds to the I. m. posterior pars super- 23) but in front of the orbit in Pelobates and Neobatrachia ficialis, and the I. m. anterior pars lateraiis corresponds to (state 1). In pipids, he described the origin of the single anthe I. m. posterior pars profunda based on relative positions gularis muscle as from both the quadrate and the ceratohyai and points of insertion; compare Sedra and Michael(1957) and coded the state as missing. Sokol (1977a) provided to De Jongh (1968), Edgeworth (1935), and Gradweli greater detail on the origin of this muscle, but even so, the (1971b, 1972b). Gradweii (1971b) d e d attention to this extreme distortion of the pipid paiatoquadrate precludes a discrepancy, but unfortunately followed the terminology of clear assignment to either state. In Haas's tree (1997a: fig. Sedra and Michael (1956a, 1957). 15) with charaaers optimized using accelerated transformaIn Rana, the 1. m. posterior profundus inserts on the su- tion (Swofford and Olsen 1990), state 1 is a synapomorphy prarosuai cartilage. In pipids the suprarostrai is fused to the of Pipanura. However, the delayed transformation optimizatrabecular horns, and the 1. m. posterior profundus (mis- tion would suggest that state 1is a synapomorphy of Pelobanamed the I. m. anterior pars lateralis by Sedra and Michael toidea + Neobatracha. 1957) inserts on the cartilaginous base of the tentacle of Silurana tropicalis and Xenopus lmis (Gradweii 1971b; O. M. Sokol 1977a); t h s muscle was aiso called the m. tentaculi Hyobranchial Apparatus (Nieuwkoop and Faber 1967). In Hymenochirus boettgri, The ceratohyais and branchiai baskets (fig. 4.7) are the two whch lacks tentacles, this siip inserts on the supralabial pad major components of the hyobranchial skeleton. The cera(O. M. Sokol 1977a), a transversely oval connective tissue tohyais and basibranchiai elements (copulae and hypopad that lies at the anterior edge of the ethmoid plate ( 0 . M. branchiai plates) are the pistons of the buccal pump that conSokol1962). Starrett (1973) tabulated differences in adduc- veys currents of food-laden water to the gdi filters and tor muscles among Orton's four classes of tadpoles but more branchiai food traps and irrigates the respiratory surfaces of the giiis. The branchiai baskets support the giii filters and variation exists ( 0 . M. Sokol1975) than she reported. gills. Water pumped posteriorly passes through the pharynJaw Depressor Muscles geal slits, opercular c a v i ~ findy, the spiracle(s). and The hyoanguiaris, quadratoanguiaris, and suspensorioangularis, which are hyoid arch muscles innervated by the faciai Ceratohyals and Branchial Bashets nerve (VII), insert on the retroarticular process of Meckel's The ceratohyals are the main levers of the pump upon which cartiiage. The suspensorioangularis is a fan-shaped muscle muscle contraction aas. The ceratohyals lie anterior to the that extends from the lateral aspea of the base of the muscu- branchiai baskets and are generdy oriented transversely. The lar process and inserts via a short aponeurosis on the retroar- ceratohyais articulate with the paiatoquadrates by means of ticular process of the lower jaw (figs. 4.94.10). The hyoan- the synoviai hyoquadrate articulations, which are reinforced guiaris is cone-shaped and originates from the laterai aspect by a stout hyoquadrate iigament. The ceratohyals articulate of the ceratohyal slightly ventral to the articulation of the with each other medidy via copula I1 (posterior copula, ceratohyal and palatoquadrate cartilage. It inserts via an apo- copula posterior, basibranchial) and the interhyoid iigament neurosis in comrnon with the quadratomgularis just laterai (iigamentum interhyoideum, iigamentum interhyale), of

ARCHITECTURE

71

which the posterior part is a mass of fibrous cartilage called copula I (anterior copula, copula anterior, basihyale). The copulae represent the medial basihyobranchial structures. The pars reuniens (fig. 4.7) is a medial mass of white fibrous tissue that binds the ceratohyals together medially and also separates copula I and copula 11. The subtriangular space bemeen copula 1 ,the hypobranchid plate, and the ceratohyal 1 is occupied by the intrahyoid ligament (ligamentum intrahyoideum; Gradwell 1972b). The crista hyoidea (urobranchial process) projects posteriorly Erom copula I1 and is the origin of the transversus ventralis 1 muscle. Within Am~hibia. 1 copula I is unique to tadpoles; the pars reuniens is believed to be homologous with the hypohyals of salamander larvae and copula I1 is believed to be homologous with the basibranchials and urobranchials (0.M. Sokol1975). Together, the ceratohyals, copulae, ligaments, and parts of the hypobranchial plates form the major skeletal support of the buccal floor. The branchial baskets are posterior to the ceratohyals. Each branchial basket (fig. 4.7) is composed of a subtriangular hyobranchiai plate and four ceratobranchials. The hypobranchial plate is traditionally considered to be formed From m-o fused hypobranchial elements. However, Haas (1997a) argued that the plate is composed only of hypobranchial I, homologous with the slrmlarly named urodelan structure. The four curved ceratobranchids are fused at their ~roximal and distal ends to form a bowl with three eloniate slits; ceratobranchial I is most anterolateral. The proximal fusions are the proximal commissures (comn~issuraeproxlmales) and the distal firsions are the terminal cornrnissures (cornmissurae terminales). Near the anterior end of ceratobran&ids I1 and I11 there is a prominence, the branchial process (processusbranchialis), to which a number of bralchial musca attach (table 4.2; Gradwell 1972b). In Rana catesbeiana i mtobrandhial I is ksed to the hypob;ancl~ial plate by hyaline cartilage; the other three ceratobranchials are joined by connective tissue [Gradwell 1972b). The spicules (spicula) are four elongate cartilaginous processes projecting dorsally and posteriorly from the ceratobranchials to support the ventral velum (chap. 5). Together tfie hwobranchials and ceratobranchials form the skeletal supGr?cfor the gills and filter apparatus. Diversity. Haas (1997a) provided detailed comparative accounts of the hyobranchial skeleton and muscles of Alytes, Ascapbus, Bmsbina, and DiscoJhssus. Ridewood (1898b) first pointed out that in Alytes, Bombina, and Discoglosus, copula II separates the two l~ypobranchial plates. This state is also present in Ascapbus, and the derived condition (hypobranchial plates in contact) is a synapomorphy of Pipanura. The fusion of hypobranchial I with ceratobranchial I is also a nnapomorphy of Pipanura (Haas 1997a). Pipaparpa andxinupuslack copula I and Pipa lacks copula I (Haas 1995; 0. M. Sokol1975). Copula I of microhylids I is r&e (Haas 1995). Ridewood (1897b) stated that there is no copula I in Pehbates or Pehdpes, but Rotek (1981) reported copula I in Pelobates &stus. Ridewood (1897a) termed copula I1 the basihyal but acknowledged the homol-

ogy of copula I1 with the basibranchial and of copula I with the basihyal. In certain taxa the ceratobranchials are fi~sed to the hypobranchial plates, and in other taxa the attachment is perichondrial or collagenous; the branchial baskets are bound to the copula by ligaments or perichondrium or may be fused in sorne taxa, such as suspension-feeding microhylids (Haas 199613; 0.M. Soko11975).Hymenochirines have only two of the four ceratobranchials (Memies 1967; 0. M. Sokol 1962). Certain ceratobranchials are also lost in the bromeliad-dwelling larvae of Hylapicadui and H. zeteki (personal observation). Noble (1929) reported only one ceratobranchial, with a single branchial slit posterior to it, in two species of Hqkrphlyne. In most pipoid larvae and in some microhylid larvae, the filter plates, which are supported by the ceratobranchials, are chondrified. In many species, both ceratobranchials I1 and I11 bear a branchial process for muscle attachment at their proximal ends (Haas 1997a). Dendrobatids are unique in having free proximal ends of ceratobranchids I1 and I11 (Haas 1995). Haas (1993, 1995, 1997a) described several character states for the ceratobranchials it1 a wide variety of taxa.
Muscles Associated with the Ceratohyal The orbitohyoideus and suspensoriohyoideus (figs. 4.94.10), which are hyoid muscles innervated by the facial nerve (VII), originate on the muscular process and insert on the ceratohyal. The orbitohyoideus is a very large, tapering muscle that originates along the curved upper margin of the muscular process. Most of the concave lateral surface of the process lacks muscle fiber attachments. The orbitohyoideus extends ventrally and slightly posteriorly to a fleshy insertion on the ventrolateral edge of the ceratohyal and conceals the origins of the hyoangularis, quadratoangularis, and suspensorioangularis. The suspensoriohyoideus muscle originates from a broad surface on the posterior edge of the muscular process and is partially concealed by the orbitohyoideus; it extends ventrally to insert just dorsal and slightly posterior to the orbitohyoideus. In the adult these muscles form part of the depressor mandbulae.
Diversity. Haas (1997a) compared the hyobranchial musdes of Alytes, Ascapbus, Bombilza, and Dis~o~hssus. work His indicates significant discrepancies in the literature treating these muscles. For example, the suspensoriohyoideus is reported absent in Ascaphus (Pusey 1943), but Haas (1997a) reported it as present. 0. M. Sokol (1977a) reported the suspensoriohyoideus as absent in Pipa mrpalboi and present in Iiym~nochirws boettflmem Silurana mpicaLis. Gradwell and (1971b) said it was present in Xenopus laeve?. Starrett (1968) indicated that the suspensoriohyoideus was absent in Rhinophpus and X. hems but later (1973) reported it present in pipoids. The muscles associated with the ceratobranchials are cornplex; difficulties in assessing their homology within Anura and with other vertebrates have resulted in various sets of terminology. Clarification of the homologies would require

72

DAVID CANNATELLA

explanation, placed the branchiohyoideus correctiy in his branchiai rather than hyoid group. Schiosser and Roth (1995) did not discuss the significance of their findings for issues of muscle homology. Haas (1997a) interpreted their implicit homologization to mean that the constrictor branchiahs muscles had traditionaiiy been misnumbered, and he argued strongly for adoption of the new labels. Some observations support this new scheme: (1) the seriai innervation of the four putative constrictor muscles by Levator Arcus Branchiaiis. Four h, muscles com- craniai nerve M and three rami of craniai nerve X is consiswide prise the levator arcus branchialis series or levatores arcuum tent with this scheme, (2) the branchiohyoideus of saiamanbranchialium (branchiai levators; fig. 4.9). In Rana tempora- ders is reportedly innervated by craniai nerve VI1 and thus ria these are four discrete muscles that extend from the lat- is unlikely to be homologous with the ccbranchiohyoideus eral and posterolaterai margin of the ceratobranchiai basket externus" of frogs (constrictor branchialis I, sensu novo), dorsaiiy and attach to the posterior part of the paiatoqua- and (3) the dista1 end of the constrictor branchialis I1 and &ate and adjacent otic capsde. This series was confsingiy I11 (sensu novo) is closely placed to the proximai ends of caiied the constrictor branchialis (for 1-111) and levator arcus levators I1 and 111. The distai end of constrictor branchiahs branchiahs (for IV) by Sedra and Michael (1957) and the I (sensu novo) is aiso closely placed to levator I. This is in constrictor arcus branchialis by O. M. Sokol(1977b). accord with Edgeworth's (1935) contention that the anuran Levators 11 1 are fused in Xenopus l d s ; Sedra and Mi- constrictors and levators form fiom the dorsai portion of the -1 chael (1957) incorrectiy reported fusion of I-IV accordiig four branchai muscle plates. to O. M. Sokol (197%). In Pipa carnalboi muscles 11 are Other observations do not support the proposed homol-1 fused, and Hymenochirzrs boetijeri lacks levators I1 and 111, ogy but are not directiy contradictory. (1) The constrictor apparentiy related to the fusion (or lack) of ceratobranchais branchialis I (sensu novo) extends between arches (i.e., 11-IV (O. M. Sokol 1977a). These musdes apparentiy have ceratobranchai I to the ceratohyal) whereas each other condisappeared in addt Rana or possibly form the petrohyoid strictor lies aiong one ceratobranchiai, and (2) the urodelan branchial depressors are believed to be homologous with the series. anuran constrictors. The urodelan depressor branchalis I lies Constrictor Branchiaiis. There are three constrictor only aiong ceratobranchai I and has no attachment to the branchialis muscles (1-111) in Rana catesbeiana; each musde ceratohyal (Haas 1997a: fig. 13), so ifthis muscle is homoloextends from the mediai end of its own ceratobranchiai g o u with the anuran constrictor branchialis I (sensu novo) aiong the anterior margin of the gi slit to the terminal com- then the attachments of the latter have shifted considerably. missure. These muscles are intimately associated with the (3) The general placement of the urodelan depressors I, 11, afTerent and efferent branchial arteries and gdis. The con- and I11 closely matches that of the three traditional anuran strictors apparentiy regulate the flow of water through the constrictors I, 11, and 111. (4) Each of the three classical angiii slits by changing the position of the interbranchiai septa uran constrictors lies in intimate contact with an afTerent (O. M. Sokol 1977a) but do not function during regular branchiai artery aiong the ceratobranchiai; the constrictor phasic pumping (Gradweii 1971b, 1972~). tadpole has branchiahs I (sensu novo) has no such association. (5) AlNo a muscle extendiig aiong the length of ceratobranchiai IV though the innervation of constrictor branchaiis I (sensu (O. M. Sokol1975); Pusey (1943) claimed that constrictor novo) by nerve M is consistent with that muscle belonging to the first branchial arch, it does not necessarily foiiow that IV was fused with levator n ! The traditional seriation of the constrictors was chal- the musde is a constrictor branchahs per se. These differences do not reject the proposed homology lenged recentiy. Using whole-mount imrnunohistochemistry, Schiosser and Roth (1995) identified the innervation of with constrictor branchiahs I (sensu novo) but simply indifbur branchai arch muscles by the principal ramus of craniai cate that the anatomy of that muscle is somewhat modified. nerve M and three branchiai trunks of craniai nerve X, re- For example, Schiosser and Roth (1995) reported that the spectively. They named these muscles constrictor branchialis first branchiai trunk of craniai nerve X innervates structures I, 11, and 111. What was surprisingwas that their constrictors of the second branchial arch, including levator arcus I1 and I11 corresponded in anatomical position to the tradi- branchialis 11, constrictor branchalis 11, subarcualis rectus tionai constrictors I and I1 as described for Rana by others. 11-n! and transversus ventrahs I1 (= subarcualis obliquus of Their (new) constrictor branchalis I had been caiied the Haas 1997a). However, their constrictor branchialis I1 actubranchiohyoideus externus (= ceratohyoideus externus) be- aiiy lies aiong the first branchiai arch in intimate contact with cause of its presumed homology to the urodelan branchio- the first afTerent branchial artery. The same applies to the sechyoideus, a muscle of the hyoid arch. The branchohyoideus ond branchai trunk of craniai nerve X, which innervates the third branchial arch. One can infer that the constrictor musexternus was reported inrlscapbus,Bombina, Dismjlossus (hsey 1943), and Scaphiopus (O. M. Sokol 1975) but not in cles must have significantiy shfted their points of attachPelobates or any neobatrachian. Gradweii (1971a), without ment, assuming that the new homologies are correct. Accep-

extensive comparative work (see Fox 1965; Haas 1997a; Joiiie 1982; Wiley 1979), but compared to the jaw muscles and chondrocranium, there are few comparative data on the muscles of the branchial baskets. Because of the general confusion regardiig muscle identities, taxic diversity is included with the description of each muscle. These muscles are supplied by the glossopharyngeai ( E ) and vagus (X) nerves. The functions of these musdes are discussed below under Gdi Irrigation in Rana catesbeiana.

ARCHITECTURE 'Table 4.4 Homologies of muscles of the branchial baskets, as described by various authors for specified taxa
. -

Edgeworth 1935 Ram Levator arcus branchiah I

Pusey 1943 Ascupbus Levator arcus brachialis I

Sedra and Michael 1957

Ws

Sokol 1975,1977a Pipa Constrictores m u m branchialium (hsed) (fused) (fused) Ceratohyoideus externus Interbranchialis I Interbranchialis I1 S. r. IT-IV S. r. I Ceratohyoideus internus Absent Subarcualis obliq uus Diaphragmatobranchi alis Tympanopharyngeus Absent Laryngcobranchialis

Haas 1996b Gamtheca Levator arcus branchialium I

Haas 1997a Bombina Levator arcuum branchialium I

Constrictores arcuum branchialium (fused) (fused) Levator arcuum IV Absent Subarcualis rectus I1 Subarcualis re~tus 1 11 Subarcualis recms I V Subarn~alis rectus 1 Absent Absent Transversus ventralis I1 Not described Transversus ventralis IV Absent Absent

L a. b. Il L a. b. LII L a. b. IV
.knt Consmctor branchialis I C . b. II C . b. I11 Subardis r mI 5. r. II
S. r.

L. a. b. 1 1 L. a. b. I11 L. a. b. N
Branchiohyoideus externus Constrictor branchialis I C. b. I1 C. b. I11 Subarcualis rectu.9 I Absent Subarcuales recti IV and ?V Subarcualis obliquu% 11-IV and ?V Diaphrapatobranchiah Absent Transversus ventralis N Absent

L. a. b. TI Laa. b. 1 1 1 L. a. b. N
Absent Constrictor branchidis I C. b. I1 C b. 111 Branchiohyoideus internus Subarcualis rectus

L. a. b. I1 L, a. b. 1 1 1 L, a. b. IV
Constrictor branchidis I C. b. I1 C. b. 1 1 1 C. b. IV S. r. I (dorsal head) S. r. I (ventral head) S. r. 11-IV Subarcualis obliquus Diaphragmatobmchialis Absent Absent Absent

m +N

Tiam-ersus venrralis II Dqhragrnatobranchialis IV Sot described -bsent -Absent

Subarcualis rectus IV Transversus ventralis I1 Diapluagmatobranchialis Tympanopharyngeus Absent Absent

m c e of this shift in attachment of constrictores branchiales but were named subarcualis rectus 11-TV. In reporting tlwee I and I Ileads to the conclusion that the attachment of con- muscles in Xinopus, Gradwell (1971b) followed Sedra and I I rtrictor branchialis I is part of a common pattern. Given the Michael's terminology but named them B2a, 2b, and 2c, as rrature of their whole-mount preparations, it is not clear he did with the constrictor branchialis muscles of other nonakxher Schlosser and Roth (1995) were able to see the pipid taxa that he studied. banchial arch or the artery, or whether they recognized the Haas (1997a: fig. 13) also illustrated, but did not discuss, significance of their muscle identifications. three branchial constrictors in Oduntibvynus and Pelabates. I have followed the new arrangement of Schlosser and The first of these in Odontophyus is labeled constrictor RMh (1995) as promoted by Haas (1997a). The traditional branchialis 11, but the last avo are both labeled constrictor amsnictors I, 1 , and I11 are simply renumbered 11,111, and branchialis 111, an error in labeling (Haas, personal commu1 n-:and the branchiohyoideus externus muscle (found in nication). Assuming the last muscle is in fact constrictor ant)- a few frog species) is renamed constrictor branchialis I branchialis IV, its origin is correctly shown on ceratobranI @. 4.9; table 4.4). chid I11 rather than TV as Haas (199%) illustrated forhcaHaas (1997a: fig. 13b) reported a short constrictor phus; this does not tnean that Haas's illustration ofhmphus banchialis IV (sensu novo) in two specimens of leutphus, is incorrect. The attachments of the three constrictor muscles similar in appearance to that described by Pusey (1943). in Oduntophrynus and Pelabates are almost identical to those Hon-ever, Haas described and illustrated it attaching to the in Rana (fig. 4.9). tnse of ceratobranchial IV rather than ceratobranchial I11 as Haas's (1997a) phylogenetic analysis indicated that the in Pusep (1943). Gradwell (1973) reported and illustrated absence of the constrictor branchialis I (sensu novo) was a the same muscle (as R2c) on ceratobranchialII1 ofAscapbus. synapomorphy (character 19) for the ciade Pipanura. HowBixh Gradwell and Pusey reported that the muscle is ever, Haas did not include data from Smphiqpus or Spea, and amached to the incipient branchial process at the base of Sokol (1975) reported that these larvae have the muscle (I oaarobranchial 1 1and inserted into soft tissue rather than have verified &is for Speu bombzfions). If one assumes that 1 mto the ceratobranchial. Haas (1997a) reported this muscle Speu is the closest relative of Pelabates (among the taxa Haas &mt in Alytes, Bombina, and Disc~lassus.Pusey (1943) examined), then under parsimony optimization the muscle ared its presence on ceratobranchial I11 in Dis~~lassus.must have reappeared in Spea. Clearly, a greater range of tax0.AI. Sokol(1975) erred in claiming that there was no con- onomic sampling is needed. striaor (he termed these interbranchialis muscles) on cerato'bc;Lochlal n I in Types 1 and 3 tadpoles. Three constrictors Subarcuaiis Rectus, The homology of the four subarcualis me illustrated in Xmpus hem (Sedra and Michael 1957) rectus muscles with those of other amphibians and fish is

74

DAVID CANNATELLA

particularly troublesome (Jollie 1982; Wiley 1979). In anurans these muscles extend more or less longitudinally, connecting each gill arch to the one in iiont. Subarcualis rectus I connects the ceratohyal to the ceratobranchials, and subarcualis rectus 11,111, and IV span the ceratobranchlals. Haas's (1997a) interpretation of subarcualis rectus I is followed here. This muscle has several conformations. InAscapbus it has but one head. In some species (e.g., Bombina) there are two closely placed slips that connect the base of ceratobranchial to the ceratohyal. In other species the two slips are more dstinct, and the deeper (viewed ventrally; fig. 4.9) dorsal head extends from the base of ceratobranchial I to the processus posterior hyalis on the ceratohyal. The ventral head stretches &om the branchial process of ceratobranchid I1 to the ceratohval. These heads are so distinct as to have been given different names (e,g., ceratohyoideus internus; table 4.4). Subarcualis rectus I is innervated by the principal ramus of cranial nerve LX (Schlosser and Roth 1975). The remaining part of the muscle series, the subarcualis rectus 11-IV, is a band of muscle extending across the proximal parts of ceratobranchials I-IN, as in Ascapbus. The compound nature of this muscle is suggested by the observation that in Dism~lassus subarcualis rectus 11-IV is innervated by the first, second, and (probably) the third branchial rami of cranial nerve X (Schlosser and Roth 1995). In other species (e.g., R a m temporaria) the muscle spans ceratobranchials 11-IV (as subarcualis reams 111-IV) or only ceratobranchials III-IV (as subarcualis rectus IV, Haas 1997a). The literature indicates general confusion about these muscles. The subarcualis rectus 11,111, and IV of Sedra and Michael (1957) are actually constrictor branchiatis muscies. 0 . M. Sokol illustrated a muscle in Pipa he termed the fused subarcualis reaus 11-IS( but this appears to be the third branchial constrictor (1977a: fig. 6). Subarcualis rectus 11-XV is absent inH-~emchiws0 . M.Sokol1977a). Thus, ( it seems that there is no verified subarcualis rectus 11-IV muscle in pipids. In Rana the subarcualis rectus muscles disappear at metamorphosis.

arcualis rectus I might actually represent subarcualis obliquus I, whch is missing in frogs and salamanders. This muscle does not persist past metamorphosis.

Transversus Ventralis I . Transversus ventralis l is presV V ent in Ascaphus truei and salamanders (Gradwell 1973; Haas 1997a; Pusey 1943) but is not present in Rana catesbeiana (Gradwell 1972b). Haas (1997a) reported that in Ascapbus the origin is the proximal part of ceratobranchial IV and the insertion is a median coimective tissue raphe anterior to the glottis. Edgeworth (1920, 1935) reported that a poorly developed transversus ventralis IV (hyopharyngeus in his terminology) disappeared during ontogeny in young larvae of Ram. tempmaria; he also (1935) reported it absent in P$a pipa and Xenopus lamis. Sedra and Michael (1957) reported the muscle present in Xenopus t m ? , and Gradwell (1971b) fbllowed their usage. 0. M. Sokol (1977a) considered the tranmersus ventralis nT Silurana tropicalis and Xenclpus of Lamis to be synonymous with the tympanopharyngeus of Pipa. Haas (1997a) believed that Sokol was correct in his interpretation, and I follow his opinion here; in his data matrix only Ascapbus has transversus ventralis rV:Noble (1929) reported a "hyopharyngeus" in Hophphyte, but its topography does not correspond to the transversus ventralis IV of other taxa. The transversus ventralis IV does not survive metamorphosis. Tympanopharyngeus. The tympanopharyngeus muscle, as described in Rana catesbeiana (Gradwell 1972b) and Pelabates@scus (Schulze 1892), extends from the lateral aspect of ceratobranchial IV laterally to the tympanic region of the otic capsule. According to Sedra and Michael (1957) and Gradwell (1971b), in Xenupus the muscle extends from the medial aspect of ceratobranchial IV dorsal to the dilator laryngis muscle to meet the fibers of its companion in a medial raphe. In Pipa the tyrnpanopharyngeus originates medial to the dilatator laryngis and inserts on the ventral surface of the posterior pharynx (0. M. Sokol 1977a). Gradwell (1972b: fig. 1) described and figured the tympanopharyngeus of Rana catesbeiana as extendirlg from the otic capsule to ceratobranchial essentiaUy following Schulze (1892). Haas (1996b) gave the origin of the tympanopharyngeus in Gmtrotheca as the pharyngeal wall and the insertion as the pericardium and indicated that it was closely associated with the levator arcus branchialis IV and the dilatator laryngis. The descriptions presented by 0.M. Sokol, Sedra and Michael, Gradwell, and Haas are somewhat different. However, Haas (1997a) examined the muscle in a wide variety of frogs, and I accept his conclusions about the distribution of this muscle; the tympanopharyngeus is absent in Alytes, Ascaphus, Bornbina, Dism~lossus, Ximpus and present in G ~ o t b e c a , and Pelabates, and other neobatrachians but not Rana tmporamk. In his analysis, the presence of the muscle (character 16) is a synapomorphy of Pelabates + neobamchians but is reversed in Rana.

Subarcualis Obliquus. The subarcualis obliquus I1 (transversus ventralis 11) is a stout muscle that extends from the crista hyoidea of copula I1 to the branchial process in Rana (Edgeworth 1935). Pusey (1943) argued, probably correctly, that Edgeworth's (1935) terminoloby was in error and that this muscle should be called subarcualis obliquus 11. However, he pointed out that Edgeworth's transversus ventralis IV was correctly labeled. Edgeworth (1920) had earlier stated that transversus ventralis I1 was not present in any amphibian, so his (1935) listing of transversus ventralis I1 is curious. Pusey (1943) also pointed out that only Ascapbus among frogs has subarcualis obliquus 11-V. Muscles 111-V are a fused series of slips that attach to the crista hyoidea and fan out to insert on the medial portions of the ceratobrar~chials and spicdum IV (Gradwell 1973; k e y 1943). Other frogs, indudng Alytes, Bornbina, and ~ s c ~ l o s s uhave only subars, cualis obliquus 11, which Haas (1997a) referred to as the Diaphragmato-branchialis. The diaphragmato-branchialis subarcualis obliquus. Haas (1997a) also speculated that sub- (IV) (diaphragmato-branchialis lateralis) extends antero-

ARCHITECTURE

75

laterally from the larval diaphragm to the process at the 111; distal end of ceratobranch~al identification of this muscle in various taxa seems to be unproblematic. It is present in A[ytes, Ascaphus, Bombina, Discogloms, Gas~votheca, Pehbates, and Rana catesbeiana (Gradwell 1972b, 1973; Haas 1996b, 1997a; Pusey 1943; Schulze 1892). It is also present in Hymenochirus and Pipa ( 0 . M. Sokol 1977a) but not reported in ;Yenopus. Although the muscle name is often followed by a 'W," it does not attach to the fourth arch. However, the muscle is innervated by the third branchial ramus of cranial nenre X, which supplies the fourth branchial arch. Pusey (1943) hypothesized this muscle to be homologous to the onloarmalis muscle of salamanders. Schulze (1892) described a diaphragmato-branchialismedialis in Pelobatesfisas, but Edgeworth (1935) and Noble (1929) considered this part of the rectus cervicis/rectus abdominis series. The diaphragmato-branchialis does not persist in adults.

chael 1957). 1nDbco~lo.r~~~ the ramus Iaryngeus brevis innervates the dilatator laryngis and the r. laryngeus longus supplies the constrictor laryngis (Schlosser and Roth 1995).The dilatator and constrictor survive metamorphosis and bear the same name in the adult. The laryngeobranchialis muscle may be unique to Pipa (0. M. Sokol 1977a); it originates from the medial end of ceratobranchial III and inserts on the developing larynx.

Cucullaris. Gradwell (1971b) and Sedra and Michael (1957) described the cucdaris in Ahupus W . F ;it extends from the lateral part of ceratobranchial IV to the scapula. In Discglossus the muscle appears in a Gosner 33 tadpole (Schlosser and Roth 1997a). The cucullaris is a classical branchial arch derivative in tetrapods, but it is not described from any other frogs. Gradwell (1972b: fig. 1 and table 2) figured an unnamed branchial muscle with its origin on the lateral part of ceratobranchial and insertion on the scapula. This may be the cucullaris.

The Larynx and Its Muschs In adult frogs (except pipids) the larynx consists of paired, crescent-shaped, movable arytenoid cartilages that are nestled within the cricoid ring (annulus cricoideus); in pipids the larynx is highly modified (Ridewood 18976). In Rana tempovaria the laryngeal muscles form before the arytenoid and cricoid ring chondrifjr during metamorphosis (Edgeworth 1920). Ridewood (1897a) noted that the larynx begins to appear in ;Yenupus at stages in which the hind legs are just appearing, and that in Pipapipa the larynx is present in embryos of 12 rnm SVL. Sedra and Michael (1957) noted that the primordia of the arytenoids are feebly developed in a Nieuwkoop/Faber 55 cadpole of %nupus laevis. J. J.Wens (1989) noted that the larynx of Spea bombzj+ons chondrifies rapidly at the end of metamorphosis. Trewavas (1933) discussed the laryngeal development of several raxa. It appears that the larynx develops earlier in P+a and Xenqus than in other frogs. The muscles of the larynx are derived from esophageal constrictor muscles rather than from the muscle of the gill arches (Edgeworth 1920); they are innervated by the la*geal branch (ramus laryngeus) of the vagus nerve (X). Most of the laryngeal muscles develop after metamorphosis and are not discussed here. A few arise just prior to metamorphosis -the dilatator laryngis muscle arises from the edge of levator arms branchialis IV and inserts on the arytenoid cartilage and the cricoid ring. The constrictor laryngis muscle consists of dorsd and ventral uarts that arise from medial raphe and insert on the arytedid cartilages (Sedra and Mi-

M e t a ~ h o s i o the Hyot?ranchium s f The hyobranchial apparatus does not undergo any appreciable change in shape until the beginning of metamorphosis (fig. 4.11). Metamorphosis of the apparatus generally begins at the time when the tadpole loses its jaw sheaths and the forelimbs break out. Metamorphosis will be described generally in three stages-the beginning of metamorphosis (Gosner 41-43), metamorphosis (stages 44-45.), and newly metamorphosed (stage 46). At the beginning of metmorphosis, the transversely oriented ceratohyals assume a more longitudiial orientation. The ceratohyals become larger and are in close medial contact when they reach their maximum size. The ceratobranchials are smaller. The spicules of the ceratobranchials disappear before the ceratobranchials. The cartilage of the ceratobranchids shrinks, and the clefts decrease in size. Copula I disappears and the pars reuniens gradually dlsitltegrates. A thyroid fenestra or foramen initially forms as a depression near the medial margin of the hypobranchial plate and then perforates. The medial margin of this fenestra becomes rodlike and forms the incipient posteromedial process (thyrohyal). The hyoglossal and laryngeal sinuses enlarge. Meckel's cartilages begin to straighten and lose their sharp angulation; they lie in a gentle curve with the infrarostds. As metamorphosis proceeds, the ceratohyals orient yet more posteriorly and become slender (fig. 4.11D). They are united mediauy and lose their articulation with the palatoquadrate. In some taxa, new pieces of cartilage form along the anteromedial margin of the ceratohyal; these will form the anterior process of the ceratohyal. (This process of the metamorphosing ceratohyal is not homologous to the larval "anterior process" on the ceratohyal, which is resorbed; see De Jongh 1968.) The hypobranchial plates reduce, and the thyroid foramina enlarge. The distinction of the hypobranchial plates from copula I1 is less obvious but persists as a clear zone in the cartilage plate demarcating the posterior limits of copula 11. The posteromedial processes form from the hypobranchial plate and do not represent the remnants of the fourth ceratobranchials (Ridewood 1897b). Lateral margins of the hypobranchial plate begin to form the anterolateral and posterolareral processes. The ceratobranchials are much smaller and break up, and the laryngeal and hyoglossal sinuses enlarge. The lower jaw is much larger and more U-shaped. The lnfrarostrals fuse to Meckel's cartilage giving the appearance of one element. The angulospienial and later the dentary ossifies. In the final and third stage of metamorphosis (stage 46) the ceratohyals are very slender and their posterior ends are fused to the otic capsule. They are no longer united medially,

DAVID CANNATELLA

Thyroid f.

Posterornedial proc. (incipient)

Ceratobranchials Laryngeal sinus

Posterornedial proc

Mentorneckelian bone

osterolateral proc.

Parahyoid bone

Posterornedial proc. (bony)

Fig. 4.11. Changes in the hyobranchiai apparatus during metamorphosis i Peiudpespunctatus. (A) Dorsal view, tadpole with intaa mouthn parts, h d limbs 1mm long; (B) dorsal view, rnetamorphosing tadpole with rnouthparts absent and empted forelimbs, tail23 mm,

hind lunb 28 mm; (C) dorsai view, rnetamorphosing tadpole, tail2.5 mm; (D) dorsai view, froglet with resorbed tail, 21 mm SVL, (E) ventral view, froglet with resorbed tail, 17 mm SVL, and (F) ventral view, adult frog, 37 mm SVL. Redrawn frorn Ridewood (1897%).

and resorption of cartilage results in copula I1 forming part of the margin of the large hyoglossal sinus. The body (corpus) of the hyoid generaliy is a flat plate (fig. 4.11F). The anterior processes of the ceratohyal, formed from autochthonous cartilage, become indistinguishable from the ceratohyal proper. The ceratobranchials have disappeared completely, and the thyroid foramen is a sinus in the margin of the hyoid plate. The posteromedial processes begin to ossi$, and the anterolateral and posterolateral processes are weli forrned. The lower iaw at&ins the twicd U shave of the adult, and Meckei's caklage is investedlby the anghosplenial and'dentary bones; the mentomeckelian bones appear. Overaii the appearance of the hyoid apparatus in a newly metamorphosed froglet is very similar to that of the adult. The general development of fate of the craniai muscles is summarized from Edgeworth (1935) in figure 4.12.
Diversity. W. K. Parker's (1871) account of development of the hyobranchial apparatus is riddled with inconsistencies (fide Ridewood 189%). Copula I persists as an ossified ele-

ment in adult Hymemchims and Pseudhymenochims (Cannatelia and Tmeb 1988a, b). In some pelobatoids (e.g., Pelodytes) fusion of the anterolateral process to the ceratohyal forms a "lateral foramen" (Ridewood 189%) that is traversed by the glossopharyngeainerve and the lingual branch of the carotid arterv. The anterolateral Drocess of Alvtes is formed by adhsion of an autochthonous cartilage rather than from the lateral margins of the hypobranchial plate (Ridewood 1898a). In ali pelobatoids and in Rhinupbrynus, the middle Dart of the ceratohval is resorbed in late metamorphosis (fig. 4.11) to produce a disjunct ceratohyal (Cannatelia 1985; Ridewood 189%). In contrast to Ridewood (1898a), Pusey (1943) claimed that the fourth spiculum becomes the thyrohyal of the adult. In pipoid larvae the anterior processes of the ceratohyals are greatly enlarged and underlie aimost ali of the buccal cavity (0. Sokol 1975). M. The vosteromedial Drocesses of the hvoid are ossified in aii frogs. A parahyoid bone forms late in metamorphosis; it is present prirnitively in Anura but lost independently in different lineages (Cannatelia 1985). According to RideI
i

78

DAVID CANNATELLA
Spinal cord

Fig. 4.13. Transverse sections of body wall muscles ofXenopw laepis. Consecutive stages in a premetamorphic tadpole: (A) presence of the myotome, (B) appearance of the Unvirbelfbrsatz (= primordial vertebral process), (C) detachment of the ventral muscles, and (D) appearance of secondary muscles. (E) Sections from the anterior, middle, and posterior body regions of a metamorphosing tadpole. (F) Sections from the anterior, middle, and posterior body regions of a young postmetamorphic animal. Symbols: pale gray = notochord and spinal cord, medium gray = primary musdes, and black = secondary muscles. Redrawn from Ryke (1953).

cles, but because it forms late in ontogeny, Ryke (1953) did not regard it as a homolog of the horizontal septum. Ti-unk myotome HI spans the atlas and second vertebra. The first parts of the vertebrae to appear are the cartilaginous neural arches, which do so in an anterior-posterior sequence beginning at Nieuwkoop/Faber 48 (Gosner 27-28; Trueb and Hanken 1993). Although there are only 8 presacral vertebrae in Xinopus h s , there are 12 neural arch rudiments. Vertebra IX forms the sacrum, and X-XI1 (postsacral) contribute to formation of the urostyle (coccyx) by fusion with the hypochord. The cartilaginous rudiments of the centrum begin to appear after neural arches X-XI. Neural arches I-VIII and centra I-IV ossitjr at the same time as the first cranial elements, the fiontoparietal and parasphenoid, at Nieuwkoop/Faber 55 (Gosner 35) in Xinopus l u k (Trueb and Hanken 1993). Vertebral ossification begins at Gosner 33 in Hyh hncfmis (De Sa' 1988), stage 36 in Spea bombzjvns (J. J. Wiens 1989), and just prior to stage 36 in Hamptqhyw boliaiana (De Sa' and Trueb 1991). Ossification of the urostyle was investigated by Stokely &d List (1955). In pipids the ribs appear as distinct cartilaginous elements on presacral vertebrae II-IV and later fuse to the transverse processes of the vertebrae. The rib anlagen osslfy at Nieuwkoop/Faber 57-58 (Gosner 40-41) at the same stage that all eight presacral centra and the sacrum ossltjr. By Nieuwkoop/ Faber 64-65 (Gosner 44-45) the ribs fuse to the transverse processes, which themselves osslfy at Nieuwkoop/Faber 64 (Gosner 44). The sacral diapophysis does not assume its expanded form until just after metamorphosis. The hypochord appears in cartilage at Nieuwkoop/Faber 60 (Gosner 41), ossifies at about Nieuwkoop/Faber 63 (Gosner 43), and fuses with the postsacral vertebrae to form the urostyle at Nieuwkoop/Faber 66 (Gosner 46). By this stage, the tadpole tail has largely been resorbed. The myotomal trunk muscles can be grouped based on the time of development. The primary muscles develop directly from the myotomes, and secondary muscles develop from mesenchyme that has proliferated from the primary muscles closer to the time of metamorphosis. The primary somatic muscles of the adult fiog trunk (fig. 4.13) are the coccygeo-iliacus, ileolumbalis pars medialis, longissimus dorsi, and reaus abdominis (also see fig. 3.3A-C). The secondary muscles (fig. 4.13) are the interarcuales, intertransversarii, obliquus abdominis extemus superficialis, sternohyoideus, transversus abdominis, and perhaps the reaus abdominis superficialis.

eob
Rectus abdominis atulomln tis Obliquus aWor exfernus superfciafis

Interarcualis

- , lnterarcualis

Anterior

Anterior

Dorsalls trunci

Middle

Middle

Coccygewacralis

Posterior

Posterior

The mass of primary myotomal muscle, called the dorsalis trunci (fig. 4.13A), is covered dorsally by a sheet of connective tissue, the fascia dorsalis. The first secondary muscles to differentiate from the dorsalis trunci are the interarcuales (fig. 4.13D), located between successive neural arches. Near the time of metamorphosis the intertransversarii (secondary)

80

DAVID CANNATELLA

Rana catesbeiana

Ascaphus truei

Fig. 4.15. Diagrammatic view of water flow relative to the visceral skeleton (dorsal view) of Rnna catwbcianu. Upon entering the mouth, the water current is divided into lefi and right streams that come together just before the flow exits the spiracle. Redrawn from Gradwe11 (1972b).

functions at some time as the reservoir of a pump that moves water into the mouth, over the gills, and out through the spiracle.

Xenopus laevis
~ i 4 .~~ 4. . composite ofsagirtaland parasagidsections diagrams through the heah of hna catM&cianu, Amphw - and; X IatPis. Abbreviations and symbols: arrows = passage ofwater into the buccal cavity, BC = buccal cavity, GC = @ and PC = pharyncavity, I geal cavity. Redrawn from Gradwell (1971a, 1971b, 1972b).

and Wassersug 1988, 1989). However, too few taxa have been studied kinematically to permit broad generalizations about the relationship of neuroanatomy and locomotion in tadpoles (chaps. 6 and 9).

Gill Irrigation in Rana catesbeiana

The first important accounts of gill irrigation by Schulze remained the definitive (1888, 1892) for Pehbates works on the subject for many years. In a series of papers, Gradwell (l968,1970,1972b, c) and Gradwell and Pasztor (1968) described the anatomy of the gill irrigation mechanism in R a m catesbeiana as consisting of a buccal cavity (the principal one), pharyngeal cavity, and gill cavity. Each cavity

Buccal, Pharyn~eal, d Gill Cavities a The buccal cavity is demarcated from the pharyngeal cavity (fig. 4.14) by the ventral velum (anterior filter valve), a nonmuscular epithelial fold supported by the cartilaginous spicules of the ceratobranchials (Wassersug 1976a). The v e n d velum extends laterally and dorsally to form the paired dorsal vela, which have no muscular or skeletal support. In most tadpoles the ventral velum produces strings of foodentrapping mucus from secretory columnar cells on its underside. Kenny (1969b, c) pointed out that rows of secretory co~umnar cells on the ventral velum were incorrectly identified as muscle fibers by R M. Savage (1952). Together, the dorsal and ventral vela form a valve preventing backflow of water into the buccal cavity (Gradwell 1970). In Rana catesbeiana the floor of the pharyngeal cavity has three gill clefts or slits. Additionally, endodermal pouch I1 is perforated to form an additional gill cleft, numbered I (figs. 4.7 and 4.15), which lies between ceratobranchial I and the ceratohyal. This gdl cleft opens directly from the buccal cavity into the gill cavity and is not present in any discoglossoids or XenopuJ; it is open in many neobatrachians, but Gradwell (1972b) and Haas (1997a) disagree on its condition in Ranatempmaria. The pharyngeal cavity of R. catesbeiana is separated from the gill cavity by three gdl clefts corresponding to endodermal pouches 111, n and I?The gill cavity is enclosed by the ! operculum, a fold of integument that grows back from the hyoid arch and is fused to the skin of the abdomen (Brock 1929), except at a sinistral opening called the spiracle (=

ARCHITECTURE

81

spout of Gradweii). The spiracle has no muscular or membranous valves in Rana. The lefi and right sides of the gi cavity are conneaed across the rnidline by the opercular canal through which water passes fiom the right side, joining the stream from the lefi, to exit the spiracle (fig. 4.15). Within the gi cavity, there are four pairs of gis, each consisting of a series of about 7-15 highly vascularized tufts attached along a ceratobranchial in close association with branchial arteries and constriaor branchialis muscles 1 - V 1T (sensu novo); no muscle is associated with ceratobranchial N; Additiondy, the opercular cavity has a vascular iining, the membrana vasculosa operculans (Gradweii 1969b). Gradweli (1972b) discussed the association of the first gdi clefi with a weii-vascularized operculum and the absence of the clefi in taxa with a poorly vascularized opercular lining. Haas (1997a) found the first gill clefi to be open (charaaer 30) in Pelobates and d but one of the neobatrachians he examined. Gradweli (1972a, c) reviewed the literature on gi irrigation and pointed out that De Jongh (1968), Kenny (1969c), and R. M. Savage (1952) recognized only a buccal pumping mechanism, whereas Kratochwill (1933) and Schulze (1892) recognized branchial and pharyngeal pumps in addition to the buccal pump. The movements associated with gi irrigation wi be considered in several sections that overlap temporally: movements of the jaws, the buccal pump, the

pharyngeal pump, and the branchial pump. Gradwell's (1968,1972~) work was done on Rana catesbeiana tadpoles under light anesthesia in a supine (beiiy-up) position; it is unclear whether his results can be appiied to tadpoles under more natural conditions. However, his extensive analysis employed muscle stimulation, denervation, electromyography, and pressure recordings. Gradwell's ( 1 9 7 2 ~ ) account differs somewhat fiom his eariier report, and the latter one is used here.

Opening and Closing tbeJaws In general, jaw opening is effeaed by the jaw depressors:
hyoanguiaris, suspensorioangularis, and quadratoangularis. However, opening of the upper jaw is accomplished not by muscles but by a coupling of the mandibulo-suprarostral ligament ("iigamentum rostrale superior cartilago Meckeli"; Gradweii 1968, 1972b). Jaw opening during irrigation occurs in three sequential phases: narrow opening, wide opening, and protrusion. During narrow opening of the mouth (fig. 4.16B), contraction of the quadratoanguiaris muscle causes Meckel's cartilage and the infrarostral to swing forward and outward fiom the buccal cavity, p d i n g open the lower jaw. Wider opening of the lower jaw is effected by additional recruitrnent of the hyoangularis musde (fig. 4.16C). However, the suprarostral cartiiage has not yet moved. During protrusion, additional contraction of the suspensorio-

\ & C -

-D Hyoangulans

Fig. 4.16. Lateral views of the jaws of Rana ~ b e i a m during various phases of opening: (A) closed, (B) narrow opening, (C) wide opening, and (D) protrusion. Symbols: dashed arrows = direction of muscle action. Redrawn from Gradweii (1972b).

82

DAVID CANNATELLA

angularis causes yet fiuther forward excursion of the lower jaw, and this takes up slack in the mandibulosuprarostral ligament, which relays the force to the suprarostral cartilage (fig. 4.16D). The lateral circumoral ligament (1. cornuquadratwn laterale) and ligamentum rostrale superius quadrati prevent hyperabduction of the jaws. The coupling protrudes upper and lower jaws at the same time, and the angle formed by the jaws becomes more obtuse. Although the geniohyoid muscle appears to be in a position to open the lower jaw, its role during rhythmic jaw opening is doubtll (see Starrett 1973). It may function durGg hyperinspiration. In Rana catesbeiana, closing of the mouth occurs faster than the opening. Adduction of the upper and lower jaws takes place simultaneously Closing during gill irrigation does not produce the close shearing of the upper and lower jaw sheaths as does feeding. In addition to elastic recoil, Gradwell (1968) suggested that closing may be effected by three muscles attached to Meckel's cartilage (i.e.. levator mandibulae anterior, 1. m. anterior articularis, and 1. m. posterior superficialis) and two attached to the suprarostral cartilage (i.e., 1. m. posterior profundus and 1. m. externus, which-is considered-to have a pars anterior and pars postelater retreated rior by Gradwell 1968). Gradwell (1972~) from this opinion because he was unable to detect electrical activity in the I. rn. posterior superficialis, 1. m. posterior profundus, 1. m. anterior, and I. m. externus. The first three of these muscles lack pink, phasically contracting fibers, and it may be that these muscles function only during feeding and not during gill irrigation. According to Gradwell (1972c), movements between the two jaws are tightly coupled by the ligaments. If the upper jaw is held closed, the lower jaw does not open; if the upper iaw is held ooen.- the lower iaw cannot clos;. Similar rest&tion of movement applies to the upper jaw if the lower jaw is stabilized. If all of the levator muscles attached to the suprarostral cartilage are bilaterally cut, the upper jaw is still pulled closed by the ligamentum rostrale superius quadrati during closing of the lower jaw. Likewise, the lower jaw was closed by the ligament after the levators to Meckel's cartilage were cut bilaterally. De Jongh (1968) considered musculoskeletal movements during feeding and irrigation to be essentially the same, but muscular contractions during feeding apparently have not been studied explicitly (Altig and Johnston 1989). Gradwell (1968, 1972a) suggested that the mandibulolabialis and intermandibularis posterior muscles function during feeding, perhaps to move the labial papillae and to bring the upper and lower jaws into a shearing position, respectively. Lesion of both muscles did not affect jaw movements during gill irrigation, and Gradwell (1972a) doubted De Jongh's (1968) contention that the intermandibularis wsterior eleiated ;he floor of the buccal cavity.
" \ '

orbnohyoldeus end Suspensorbhyddeus

, - - ~ - ~

Inspiration

Expiration

Fig. 4.17. The visceral skeleton of a R m catubciana tadpole. (A) Dora

sal view. (B) Inspiration occurs by depression of the buccal floor by

contractions of the orbitohyoideusand suspensoriohyoideus. (C) Expiration occurs by contraction of the interhyoideus, which elevates the buccal floor. Symbols: dashed arrows = direction of musde contracdon, dashed line = position of the transverse sections in B and C, and gray elements = skeletal elements of the buccal floor. Redrawn fiorn Gradwell (1968).

B m l Pump
The buccal oumo is the mechanism that draws water into the buccal cavity of the tadpole. Its piston is formed by the paired ceratohyal cartilages that underlie the floor of the bucI L

cal cavity. The ceratohyals are joined medially by the pars reuniens, copulae, interhyoid, and intrahyoid ligaments discussed earlier. These joints, plus the juxtaposition of the hypobranchial plates along the midline, permit bending along the longitudinal axis. The articulations between the anterior part of each hypobranchial plate and the posterior process of the ceratohyal, and between the hypobranchials and copula 11, permit some bending in the transverse plane. Most of the excursion about the elliptical hyoquadrate joint occurs such that the medial part of the ceratohyal moves dorsad and ventrad. Some movement also occurs such that the ceratohyal tilts anteroposteriorly The lateral part of the ceratohyal bears a flat surface for the attachment of the hyoangularis, orbitohyoideus, and suspensoriohyoideus muscles. Contraction of the stout interhyoideus muscle causes the ceratohyals to rotate upward around their amculations with the palatoquadrate and elevates the floor of the buccal cavity (expiration; fig. 4.17C). The water stream is divided into two parts, one to the right and one to the left of the pharynx, as verified by observing the flow of dye particles. The interhyoideus (and its principal antagonist, the orbitohyoideus) has both larval and "adult" fiber types. Larval fibers are his-

ARCHITECTURE

83

tologicaily distinct in that they are thicker and have peripheral nuclei; these pink fibers, with "slow" contrade propertia, many mitochondria, and phasic activity, are used primarily in the regular contractions of giii irrigation. The "adult" fibers are thinner, with centrally placed nuclei; these white fibers, of the "fast" type with fewer mitochondria and intermittent activity, are used primariiy in hyperexpiration. The two types of fibers have different functions in the tadpole of Rana catesbeiana (Gradweii and Walcott 1971). It is the more anteriorly located pink fibers of the interhyoideus
Levator arcus branchialis I-IV Quadratoangularis Rectus cervicis (Tyrnpanopharyngeus)

muscle that are active during gdi irrigation; correspondingly, the anterior part of the buccal cavity, supported by the ceratohyals, is raised more than the posterior part. This configuration directs the water stream posteriorly into the pharynx. The white fibers of the interhyoideus do not function during pumping or hyperinspiration. The interhyoideus posterior consists only of white fibers; its contraction is sporadic and constricts the giii cavity, forcing water out through the spiracle. Electromyographs and mechanograms (fig. 4.18; Grad-

Orbitohyoideus Hyoangularis Suspensoriohyoideus (Suspensorioangularis, Geniohyoideus)

=
0 0

Geniohyoideus (Constrictor branchialis 11-IV, Subarcualis rectus 11-IV) Hyoangularis Suspensorioangularis Orbitohyoideus

Interhyoideus Subarcualis obliquus

0.5 sec

Interhyoideus Subarcualis obliquus Interhyoideus posterior (Levator mandibulae posterior L. rn. anterior, L. m. externus) Subarcualis rectus I Diaphragrnato-branchialis (L. m. anterior articularis)

Fig. 4.18. Sumrnary of muscle activity and hydrostatic pressures during gill irrigation in Rana catesbeiana. Muscle names in parentheses indicate that activity is postuiated, not observed. Symbols: dashed line = typicai cyde of pharyngeai pressure, dotted line = branchiai cav-

ity pressure, open rectangles = data frorn anatorny and direa observation, solid line = pressure tracing of buccai pressure for severai cycles, and solid rectangles = data from electromyography.Redrawn from Gradweli (1972b).

84

DAVID CANNATELLA

well 1972c) indicate that both elastic recoil and contraction ment joining the crista hyoidea of copula I1 to the branchial of the hyoidean depressors (orbitohyoideus and suspensori- process of ceratobranchiaiI1. Thus, during buccal floor eleohyoideus) depress the buccai floor for inspiration. The neg- vation, the ceratobranchiais are puiied fonvard and mediaiiy. ative pressure draws water into the buccai cavity through the The movements are assisted by subarcualis rectus I, which mouth. Sedra (1950) incorrectly judged that the hyoidean aiso assist in opening gill clefi I. depressors lifi the buccai floor. The suspensoriohyoideus aiso may help to stabilize the jaw joint during hyperinspira- Branchzal Pump tory contraction othe hyoanguiaris muscle and to assist in Substantiai changes in branchiai cavity volume are only occaclosing gdi clefi I (Gradweii 1968, 1972a). During inspira- sionai and occur during hyperexpiration (figs. 4.184.19). tion in Rana catesbeiana and R. Jsc@da (Gradweii 1972a) Contraction of the interhyoideus posterior muscle comwater flows in through the nares as weii, and a nariai vaive presses the opercular charnber, and contraction othe subarprevents egress of water during expiration (Gradweii cualis obiiquus (= transversus ventraiis I1 of Gradweii) and 1969a). Experiments using dye particles suggest that the lefi the normaiiy inaaive fibrdiar parts of the interhyoideus musand right currents do not mix in the pharyngeai and buccai d e elevates the buccai floor. Water is thus forcehlly expeiied through the spiracle. cavities (Gradweii 1972a) of these two species. Gradwell (1968) found that denervation of the interhyoideus, orbitohyoideus, and suspensoriohyoideus muscles Pressure Dynamzc~ greatiy reduced ceratohyai osciiiations, but a siight rhythmtc Inspiration begins with elastic recoil of the buccai floor from pumping was maintained. Bilateral denervation of the hy- an elevated position. The buccai pressure drops (fig. 4.18), oidean depressors did not completely eliminate osciiiatory and as it approaches its minimum, the orbitohyoideus and buccai floor depression caused by the continued activity of hyoangularis contract, actively depressing the buccai floor the r e m s cervicis and elastic recoii of the buccai floor. De- and opening the mouth, respectively. Probably the suspennervation of the interhvoideus showed that buccai floor ele- soriohyoideus and suspensorioanguiaris aiso contract at this vation persisted because of contraction of the subarcuaiis time. Just after buccal pressure reaches its minimum, the inobiiquus (transversus ventralis 11) and subarcuaiis rectus I terhyoideus and subarcuaiis obiiquus contract to begin the . (Gradweii 1 9 7 2 ~ ) Severing the iigamentum te& (liga- expiration phase. The pharyngeai cavity is enlarged as the mentum supraorbitale cranii and I. supraorbitale ethmoi- buccai floor is elevated and the ceratobranchiais are displaced daie) caused the muscular process of the quadrate to bend anteromedidy. Probably the subarcualis rectus I and diaoutward, suggesting that these iigarnents stabilize the ori- phragmato-branchiaiis assist as weii. The mouth closes from gins of the hyoidean depressors (Gradweii 1972~). elastic recoil, and buccai pressure increases greatiy. The osciiiations of the buccai pump are mechanicaiiy At the peak of buccai pressure, the branchiai levators and iinked with opening and closing of the mouth. When the reaus cervicis contract, constricting the pharyngeai cavity, ceratohyai was held against the buccai roof, contraction of closing the velar vaive, and elevating the pharyngeai presthe jaw depressors opened the mouth only siightiy, and the sure. The resultant hydrostatic pressure is added to that just jaws were not protruded as in normal mouth opening. When transrnitted to the pharym from the buccai cavity. Thus, wathe mouth was held open the amplitude of ceratohyai excur- ter is pushed further posteriorly across the giii clefis. Immesion was reduced. diately after the peak of buccal pressure, the inspiration phase begins again as the ceratohyals begin to recoii pasPharyngeal Pump sively, the mouth opens, and buccai pressure drops. The phaMovements of the ceratobranchiais during irrigation cause ryngeai pressure drops as weii but lags behind the buccal changes in the volume of the pharyngeai cavity; t h s consti- pressure curve. At the onset of inspiration the pressure in tutes the pharyngeai pump. Simultaneous with buccai floor the branchiai cavity rises because of d o w generated by elevation (expiration), anteromediai displacement of the compression of the pharyngeai cavity; this is the time of ceratobranchids enlarges the pharyngeai cavity, and the gds greatest efflux from the spiracle. Most other workers have are swept through the water in the gill cavity (fig. 4.19). This suggested incorrectiy that spiracular outflow is greatest durmovement is effected by contraction of the subarcualis obli- ing expiration (Gradweii 1 9 7 2 ~ ) . quus (transversus ventraiis 11), which is synchronized with that of the interhvoideus. When the ceratobranchiais reach their fonvard limit, the buccai floor recoils passively, and the Evolutionary and Comparative Aspects grlls are swept posterolaterally, reducing the pharyngeal cav- of Gil1 Irrigation and Feeding ity. The spicules support the ventral velum against the buccal The mechanics of giii irrigation were studied in hmphus roof and prevent backflow into the buccal cavity. Movement w e i (Gradweii 1971a, 1973), Bufo regularis (Sedra 1950), of the ceratobranchiais occurs both by elastic recoii and ap- Hymenochiws boett~eri(0.M . Sokol 1962), PelobatesJscus parentiy by contraction of the rectus cervicis and the (Schulze 1888, 1892), Phlynomantis annectens (Gradweii branchiai levators I-R the apparent antagonists of the sub- 1974), Phyllomedusa trinitatzs (Kenny 1969c), various species ofRana (De Jongh 1968; Gradwell1968,1970,1972b, arcualis obliquus. Movements of the ceratobranchiais and ceratohyals are c; Kratochwdi 1933; Severtsov 1969a), and Xenopus laeas synchronized to some degree by the subarcualis obiiquus, (Gradweii 1971b). Gradwell's studies of Rrcaphus, Phlynowhich in addition to contraction apparently acts as a liga- mantzs, Rana, and XencIpus provide consistent anatomicai

ARCHITECTURE

85

and functional comparisons of a species from each of the tur tadpole types. The reader is cautioned against generalizing single-speciesaccounts to aii members of a particular tadpole type. Moreover, it is worth investigating whether the differences described herein exist at a more inclusive leve1 for any tadpole type. The tadpole ofhcaphus truei (Type 3) has a large oral disc c see Noble 1929 for comparison with Amolops ricketti, another species with a large oral disc). The buccal purnping mechanism, powered by the orbitohyoideus, sucks water trapped between the substrate and the disc into the buccal cavity and generates negative pressure by which the sucker &era to the substrate (fig. 4.14). A passive oral valve seals

the mouth when sucker pressure becomes less than buccal pressure, thus maintaining the suction (fig. 4.20). Experimental lesions to the periphery of the oral disc or to the oral valve greatly decrease the effectiveness of the sucker mechanism. Rrmphus tadpoles can "hitch" forward (but not lateraiiy or backward) while adherent by altemating movements of the upper and lower iip (Altig and Brodie 1972; GradweIi 1971a; M. H. Wake 1993a). Backward movement of the lower jaw and iip by the quadratoangularis muscle opens the mouth and oral valve. Suction is thus reduced, and the friction of the lower labium against the substrate aiiows the snout to be pushed forward. During this phase the trunk and tail are abducted from the substrate, probably by contraction of the dorsalis trunci muscles. Closing of the lower

Levator arcus branchialis I-IV

Inspiration
nares

Tyrnpanopharyngeus

/'i7

Rectus cewicis

Suspensoriohyoideus

Expiration

Levator rnandibulae anterior articularis

i
I

'-F> Diaphragrnatoobliquus branchialis

Interhyoideus (Interhyoideus posterior)

Fig. 4.19. Gradweii's model of the mechanics of gi inigation in Rana catesbeiana. Symbols: dashed white arrows = direction of muscle action, gray area = pharyngeal cavity, and solid black arrows = water flow. The musdes in this model clearly have conuacted or have elecuical activity during gi irngation (see fig. 4.14). The interhyoideus posterior muscle is enclosed with parentheses to designate its activity only dunng hyperexpiration. After Gradweii (1972b).

DAVID CANNATELLA

Disengaged sucker

Peripheral contact

Engaged sucker

Fig. 4.20. Cornposite drawings of cross sections of the rnouth and nares during oral disc engagernent in Ascaphur tncei. (A) Narial and oral valves closed during expiration. (B) During inspiration, water enters rnostly through the mouth but some also enters through the nares; the periphery of the oral disc becomes sealed to the substrate

and evacuation of water frorn the sucker cavity into the buccal cavity causes the oral disc to becorne engaged to the substrate. (C) During inspiration, reduction in buccal pressure draws in water through the nares, but the oral valve rernains closed and adherence to the substrate is maintained. Redrawn from Gradweii (1971a).

iaw bv adduaor muscle contraction moves it fonvard. de- ryngeai force pumps that operate siightly out of phase. Unpression of the buccal floor increases the suction, and the like Rana, the pharyngeal pump is smaii relative to the bucbody is puiied close to the substrate, possibly by the rectus cai pump, and the outfow of water from the branchial abdominis muscles. The combination of suction pressure chamber is intermittent rather than continuous. Although a and friction of the lower lip may rasp aigae from &e sub- ventral velum is present, it is reduced and not completely strate. Thus, "hitching" may aiso function sirnultaneously in free, and the dorsai velum assists substantiaiiyin maintaining feeding. However, the labial teeth do not serve in anchoring unidirectional flow, &eAscaphus and Rana. A large interthe tad~ole the substrate (Gradwell 1971a). to hyoideus posterior muscle can simultaneously constria the Tracings of buccai pressure and suction pressure are iden- buccal, pharyngeal, and gili cavities and may function in extical, and the frequency and amplitude of irrigation is similar peiiing detritus from the irrigation system. The flow of water i tadpoles whether or not suction is engaged. The flow of from the midventral spiracle is intermittent, but because this n demonstrates that when the sucker is en- tadpole is nektonic, there is no interferente from the subindicator ~articles gaged, water enters the buccal cavity only through the nos- strate. trils; when not engaged, water enters through both nostrils and mouth. Nariai valves prevent reflux of water through Xenupus 1amG the nostrils. As in Rana, there are both buccal and pharyn- The tadpole ofXenopuslua% (Type 1)has a simpler jaw sysgeal force pumps, siightly out of phase, with a velar valve tem than those described for Ascaphus, Phynomantis, or that ensures the posterior flow of water over the gilis. Pres- Rana (fig. 4.14). Keratinized jaw sheaths and labial teeth are sure in the gili cavity fluauates but is continuously positive, lacking, and the larva is obligately microphagous. Meckel's indicating a constant flow of water from the branchial cavity. cartilage is elongate, as in the adult frog, and h e d to the Unlike Rana, there is no interhyoideus posterior muscle in infrarostrals. The suprarostrals are fused to the trabecular the opercular waii, so there is n branchial pump. Gradwei horns. This results in a wide transverse slit for the mouth. (1971a) speculated that the single, midventral spiracular Many of the jaw muscles appear to be fused. &nupus larvae opening maintains a constant oudow and prevents detritus aiso lack gilis and are obiigate air breathers, so it is inappropriate to refer to buccal pumping as giii irrigation. Gradwei or parasites from entering the giii cavity. (1971b) stated that the buccai pump is not active until the Phrynomantis annectens larva finds suspended food particles. Phasic pumping then In contrast to Ascaphus and Rana, Phlynomantzs annectens begins, and the larva suspends itself in the water column. (Type 2) is a midwater filter feeder and lacks keratinized jaw InXenopus the mouth opens as inspiration begins, but the sheaths and labial teeth. In the larvai stage, the external nares lower jaw is not protnided because of the nature of the jaw have not yet opened (as in most microhylid taxa), so the only joint. Also, Meckel's cartilage is restricted to movement in inlet is the mouth. As described for Rana, there is a jaw- the dorsoventral plane, like Phyrwmantis but not as in Ascaligament coupling such that both upper and lower jaws are phus or Rana. Correspondingly, the three depressors of the protnided simultaneously. The intermandibularis anterior lower jaw are h e d into a single quadratohyoangularis. Unand mandibulolabialis muscles are absent. Unlike Rana, the like the ligament-dependent linkage of Rana, the quadratolower jaw rotates in only one plane and the suspensorioan- hyoangularis muscle provides the only force to open the gularis muscle is absent (Gradwell1974). lower jaw during buccai depression. The upper jaw does not Like Ascaphus and Rana, there are cyclic buccal and pha- move during mouth opening and closing.
I

ARCHITECTURE

87

There is no ventral velum in Xknrtpus, and the combined mouth and throat cavity is called the buccopharyngealcavity. Simultaneous recordings of pressure in the anterior and posmior parts of the cavity are perfectly in phase as expected considering the absence of a ventral velum; therefore, there i one functional pump. The posterior pressures are less than s thase recorded anteriorly, indicating a posterior flow of auer.The interhyoideus muscle is the principal elevator of dx buccal floor, but the subarcualis obliquus and the four bxamhal ievators assist as well. -4s buccopharyngeal compression begins, the mouth and u r a valves are closed as in Rana and water passes posteriail through the branchial clefts (there are no gills) and into rhe operdar cavity. The spiracles are paired and lateral with @like covers; water that enters the opercular chambers does not m x between left and right sides and passes quickly i o m of the spiracles. Passive depression of the buccal floor contributes more rn water intake than in other taxa. but contraction of the otbitohyoideus is important as well. The inspiration phase is much shorter than expiration. Because there is no ventral d u n , the entire buccopharyngeal cavity fills with water, md reflux of water through the gill clefts is a possibility. However, depression of the buccopharyngeal floor causes rhe spiracular flaps to close, preventing reflux (Gradwell 1975~). Thus, water egress from the opercular cavities is ind t t e n t , as i11 microhylids, but for a different reason. The constrictor branchialis muscles close the gill clefts, but they do not fire during normal phasic pumping. They do contract simultaneouslv with the interhvoideus muscle as shown &en India ir;k is introduced iito the buccopharynx. The mouth remains open, and the ink is regurgitated from the buccopharynx. There is no interhyoideus posterior to power a branchial Dump. Overall, k e ;urnping apparatus of Xiruytzls is simplified compared to other tadpoles. Because of the fusion of suprarwnal cartilage and shape of Meckel's cartilage, the number of movable skeletal joints and planes of movement is reduced. Several of the muscles are absent or fused to others. The ventral velum and a separate pharyngeal pump are absent. GiUs and active muscular control of the gill cleft closing are absent, as is the interhyoideus posterior muscle and the branchial pump.

+-

attach) to the total width of the ceratohyal and the amount of rotation of the ceratohyal in the transverse plane about the hyoquadrate joint (assuming a contraction to 90% of resting length of the fibers of the orbitohyoideus) are two other morphological variables that correlate with feeding ecology. Buccal volume is strongly correlated with snoutvent length, but neither lever arm ratio nor ceratohyal rotation show significant correlation with size (Wassersug and Hoff 1979). Wassersug and Hoff (1979) demonstrated that species they considered c'microphagous suspension feeders" had small lever arm ratios (0.14-0.17) and largest angles of rotation (33.5"45.0); these included three species of microhylids, Pachymdusa ducniwlor, and Xenopus M s . Species they classhed as "macrophagous carnivores" are at the other end of the range of lever arm ratio (0.40-0.50) and rotation angles (14O-23'): Anotheca spinosa (oophagous), Hymemchiws boettoeri, Scaphiopts holbrookii, and Spea bombzfions (see Altig and Johnston 1989 for a different classification). Species termed "benthic, thigmotactic," such as Heleophyne natalensis, Hyalimbahzl~hiumfiisch~nni, noropa miliaris, and were less tightly clumped but distributed in the same half of the range for either lever arm ratio or angle of rotation. The middle part of the range consisted mostly of species with "typical" pond tadpoles. A regression of predicted buccal volume against snout-vent length (although not accounting for phylogenetic structure; Pelsenstein 1985) was highly sigt&cant and indicated that both microphagous suspension feeders and carnivores had predicted buccal volumes that were larger than those expected from the regression model. In contrast, benthic larvae with suctorial discs had smaller buccal volumes than expected from the regression (Wassersug and Hoff 1979). These oatterns were exolained bv a considerationof feeding mechanics and prey characteristics (fig. 4.21). Carnivo-

GmtparatipeEcowholo~iGal F~nctianal and Aspects @Buccal P m p i n ~ ,ilthough the diverse ecologies of tadpoles have been appreciated, relatively little attention has been given to the relationship of ecology and the mechanics of buccal pumping. Wassersug and Hoff (1979) devised a model by which they estimated buccal volumes in a broad spectrum of tadpole p &Ranacatesbeiana, 50 p1; Ranasylvutica, 10.55 p1; and i: Xmopus kvis, 14.7 p1. Negative allometry of body length mith buccal volume among and within species suggests that f h g structures impose a size limit on how big a tadpole can get before it must metamorphose. The lever arm ratio, the ratio of the projected width of the lateral part of the ceratohyal (that part to which the buccal floor depressors

Buccal Floor Area

---

Fig. 4.21. Patterns U buccal pumping design of tadpoles related to I the feeding ecology oftadpoles (see text). The size of the clouds is arbit r r . Redrawn from Wassersug and Hoff (1979).

88

DAVID CANNATELLA

rous tadpoles were predicted t o require a large buccal vol- t o rocks. Slow-water stream forms such as Hyla lgleri and ume to produce the single burst of suaion needed t o capture Rana boylii had a lower mean vaiue (0.30) than did pond moving prey. In contrast, the feeding efficiency of micropha- generalists (0.44), but the three species of obiigate, suctorial gous forms depends on the quantity of water passed through stream forms (Rrcaphus, Heleophryne, and Litoria) had a mean the buccai cavity, and a larger buccal cavity wi increase food value of 0.44, not different from generalists. The small samintake. In bo& cases, b&cal volume i; increased, but in ple sizes may be inadequate to detect significant Merences. different ways. Macrophagous tadpoles had a short angle of These comparative studies were done before appropriate ceratohyai rotation (stroke of the piston) and long lever arm methods were avadable for anaiyzing correlations among combined with a large buccal floor area (bore of the cylin- continuous characters in an expiicitly phylogenetic context der). Conversely, midwater suspension feeders had a large (Harvey and Pagel 1991). Apparent correlations among degree of rotation and short lever arm, but a s m d buccal traits surveyed in a range of taxa may be artificidy high because a certain component of the covariation is attributable floor area (Wassersug and Hoff 1979). Like the carnivores, the benthic larvae examined had long to phylogeny. It wouid be instruaive to readdress these lever arms and iittle rotation of the ceratohyal, but the buccal questions in an explicitly phylogenetic context with the apfloor, and thus buccal volume, was smaii. The large mechani- propriate methods. Reiiable information on tadpole diet and cal advantage of long lever arms generates lar& negative ecology is also needed. Although macrophagous tadpoles pressures at the suctoriai disc for adhering to the substrate have certain traits in conimon, the actual diet of these may (e.g., hmphus; Gradwell 1971a, 1973). Because benthic vary greatly. Some are oophagous bromeiiad dweiiers (e.g., larvae presumably feed on periphyton on the rocks, large Anothecaspinosa; Lannoo et ai. 1987), while others are openbuccal volumes needed for either elusive prey or to move water predators (Hymenochimsboe~~eri;. M. Sokol1962). 0 large volumes of water with tiny food particles were not ob- Other faaors, such as physiological constraints (Feder et ai. 1984a, b), merit consideration in generalizations concerning served. Tadpoles can effea a constant ingestion rate when food associations of diet, morphology, and function. concentration is experimentaiiy altercci (Seale and Wassersug 1979). For suspension-feeding tadpoles, the rate of water Phylogenetic Aspects of Tadpole Morphology pumping must be adjusted such that the g d iiters do not become obstruaed; this can be accompiished by either am- Caenogenesis and Pipoid Tadpoh plitude moduiation (AM) or frequency moduiation (FM) Frog tadpoles are caenogenetic (De Beer 1951; Gouid of the buccal pump (Wassersug and Hoff 1979). In frequency 1977); they have deviated from the ancestral ontogeny in moduiation the rate of pumping is adjusted, but each pump having larval specializations (nonterminal additions, such as cycle transports the same volume of water (e.g., Xenupw keratinized jaw sheaths and labial teeth) and less obvious but laevis). In am~litude modulation the volume of the buccd e q u d y important skeletal modifications (e.g., orientation of cavity is altered so that more or less water is pumped in one the palatoquadrate and presence of suprarostral and infraroscycle (e.g., Rana sylaatica). The type of moduiation appears trai cartilages). These larvai feeding specializations provide a to be related to the feeding morphology. Forms with short way of prolonging the larval Me cycle (Slade and Wassersug lever arms are prediaed t o have more difficuity in moduiat- 1975), and it may be that a reduaion in food is a cue for ing the amplitude of buccal floor displacement than do those metamorphosis (Wassersug 1986).At metamorphosis, these with longer lever arms. Xenopus h s , with a short lever arm, larval specializations are lost as the cranium of most tadpoles is a FM tadpole, and R. sylvatica, with a longer lever arm, is is extensively remodeled (Orton 1957). Metamorphosis in saiamanders, the outgroup, is relatively gradual, but phyloan AM tadpole (Sede and Wassersug 1979). The muscles involved in moduiating either amplitude or genetic analysis indicates that the ancestor of Anura had frequency of the buccal pump are primarily the orbitohy- drastic metamorphosis. oideus (buccal floor depressor) and interhyoideus (elevator). In Pipoidea, nearly ali of the skuii bones appear prior to Satel and Wassersug (1981) prediaed that because macro- the completion of metamorphosis whereas in other frogs phagous carnivores must generate a large suction pressure, several of the skull bones appear after metamorphosis (Trueb they would have orbitohyoideus (OH) muscles that are large 1985; Trueb and Hanken 1993). It seems that the appearrelative to the interhyoideus (IH), or a low IH/OH ratio. ance of craniai ossifications is accelerated because metamorMicrophagous larvae have dense gi iiters that strain s m d phosis in these pipoids is not as drastic as in other frogs partic&s &d were predicted to rec$re large buccal pressures (Wassersug and Hoff 1982) and less remodeling of the osand a large IH/OH ratio to force water through these filters. teocranium is required. Pipoids have several peramorphic Quantification of muscle cross-sectional areas supported features (i.e., new morphologies or shapes that are added to these predictions (Satel and Wassersug 1981). The cross- the end of the ontogenetic sequence; Aiberch et ai. 1979). sectional areas were isometric to body length. Obiigate mid- Examples of such features are the fusion of the frontopariewater feeders (i.e., five species of microhyiids, Agalychnis tais in Pipoidea, fusion of the vomers in Xenopus, fusion of callidryas, Rhinuphrynus domalir, and Xenupus hms) had the the parasphenoid to the sphenethmoid, expansion of the mehighest IH/OH values of 0.82), and macrophagous tadpoles dial ramus of the pterygoid, and an elongate columeiia in such as Anotheca spinosa, Hymenochims boett~eri,and Spea Pipidae (Canriateiia and Trueb 1988a, b). In contrast, some bombifions had a mean IH/OH value of 0.26. Suaorial forms features are interpretable as paedomorphic (i.e., developwere expected to have strong buccal depressors for clinging ment has been arrested at an earlier stage in the ontogenetic

ARCHITECTURE

89

trajectory): lateral line organs of the tadpoles retained in adults of Pipidae, the absence of eyelids and quadratojugals i Pipidae, and the absence of mentomeckelians in Pipoidea. n Pipids remain aquatic as adults (they arc the most aquatic frogs), and lack of a transition to a terrestrial life may explain, in part, the less dramatic metamorphosis. Ossification begins earlier, and peramorphic features result from prolonged development. Although the extreme degree of filter feeding in most pipid larvae is apomorphic, pipids have reverted to an ontogeny reminiscent of salamanders by abandoning specializations associated with feeding.

Ewlgtion of Orton's Tdpole T9es

Starrett (1968,1973), following Orton (1953,1957), conceived of an evolutionary sequence (fig. 4.22) in which Type 1 was the most primitive, Type 2 was more derived, and

T p e s 3 and 4 were the most derived, with T p e 4 more specialized than Type 3. These assumptions of the primitiveness of pipoid larvae led to a historical controversy about hgher level frog relationships because of incongruence of trees derived kom larval data with trees based on adult data. Studies of adult morphology had concluded that "discogiossoid" frogs such as Ascapbus, Bombina, and LeiqeImu, were the most primitive anurans (1. Griffiths 1963; Noblc 1922). This "larva versus adult" controversy was comparable to the recent, but now obsolete, c'molecules versus morphologyn debates. Starrett suggested that the plesiomorphic mode of feeding was similar to that of the Type 1 (pipoid) larvae. Water passing through the "beakless" mouth carries food particles, which are filtered out, and a steady stream of water exits the two spiracles. Feeding and respiration are the same actions.

Type 2 tadpole
Feeding and respiration sirnilar actions Some modification of Meckel's cartilage, separate infralabial cartilages MuscJes specialized for protrusible lower jaw *Separate branchial chambers with single ventral external opening Forelimbs develop posterior to branchial chambers

Type 3 tadpole
*Separate branchial chambers with median external opening Forelimbs develop close to branchial chambers

Type 4 tadpole
*Single branchial chamber with sinistral external opening Forelimbs develop within branchial chambers

Type 1 tadpole
*Characters same as Primitive tadpole

\\
-

*Feeding actions more separated from respiration *Long coiled gut for algae feeding Extra jaw cartilages, muscles and accessory mouth structures for feeding *Smaller branchial chambers Forelimbs develop close to branchial chambers

Prirni1:ivetadpole
-

Feeding and respiration same action Jaw cartilages simple Mandibular and hyoid muscles not specialized for feeding Separate branchial chambers with two external openings Forelimbs develop posterior to branchial chambers
in boldface are discussed in the text.

Fig. 4.22. Starren's interpretation of the evoiutionary sequence of Orton's (1953) tadpole types. Characters Redrawn from Starrett (1973).

90

DAVID CANNATELLA

The mouth is opened by the angularis complex and the geniohyoideus and dosed by the adductor muscles. She further argued that in the Type 2 (microhylid) tadpole some separation of feeding from respiration occurs. A strange, spoon-shaped lower lip is extended by the geniohyoideus and intermandibularisposterior. The lower lip and mouth are closed by the infralabial retractor, interhyoideus, and interhyoideus posterior. The evolution of keratinized jaw siieaths in ~~~i and 4 tadpoles contributes to the sep3 aration of feedng and respiration. The adductor muscles move the jaws in a rasping motion. The labial teeth and papillae are controlled by opposing actions of the geniohyoideus and adductor muscles. The addition of a joint between the infrarostral and Meckel's cartilage increases the range of motion of the lower jaw. The intermandibularis nosterior muscle has abandoned its ori~inal function of as" &ing the interhyoideus in buccal floor elevation and functions in feeding only. This evolutionarv seauence suffers &om rwo ~roblems. One is the misinterpretation of larval anatomical and functional data, and the second is the lack of a phylogenetic context. Starrett's statenlent that feeding and respiration are the same action in pipoid larvae is incorrect. AU pipoid larvae lack internal gills and are obligate air breathers ( 0 . M. Sokol 1977a; Wassersug 1996); thus the movement of water through the buccal cavity serves no apparent (or a minor) respiratory fiuiction. LXCmpus laePis tadpoles die if they cannot get air, even in highly oxygenated water (Wassersug 1992). Moreover, feeding in pipoid larvae is more diverse a . than w s represented by ~ k r e k x 1 9 7 3 )Rhinopbrynus larvae are known to be both carnivorous and microphagous (Satel and Wassersug 1981). Silwana anddXbwpus larvae are obligtely microphagous. Hymenuchimcs boett$jni is a predatory zooplanktivore and lacks gdl filter plates completely (0. M. Sokol 1962), free-swimming Pipa larvac are macrophagous filter feeders (0. Sokol 1977a), and the direct-develM. oping Pipa species (Tmeb and Cannatella 1986) cannot feed as larvae. I suggest that the diversification of heterogenous feeding modes of pipoid larvae results from the complete decoupling (sensu Lauder 1981) of feeding from respiration. When the buccal Dumn is freed from the constraint of irrigating the giUs, its mechanical parts are then able to evolve with the exploration of other trophic niches. Some of Starrett's other statements ( i s , purported action of the geniohyoideus in opening the mouth and controlling the mouthparts, the muscles that close the mouth in microhylid tadpoles, the action of the intermandibularis in tadpoles with jaw sheaths, and a steady watcr strem exiting the spiracles in pipoid tadpoles) are not supported by the work of Gradwell (1968, 1971a, b, 1972b, c, 1973, 1974, 1975~). Other discrepancies were addressed by 0. M. Sokol(1975). Starrett's (1973) proposal assumed that the simpler musculoskeletal system (lack of jaw sheaths and labial teeth, fewer muscles and joints) of pipoid larvae is primitive for Anura, following an implicit assumption that structural simplicity is plesiomorphic. It is well established that determination of character polarity must be based on outgroup com1
I 1

parison ( W e y 1981) rather than preconceptions about the direction of momhological evolution. The structural sim" plicity argument is flawed because it assumes as fact the very hypothesis of morphological evolution that one is testing. It should be noted that although 0.M. Sokol(1975) opposed many of Starrett's interpretations, his conclusions as well were often based on a priori notions of morphological transformation, including the assumption that fusiolls of cartilaginous elements are evol&ionarily derived. Regardless of whether or not one accepts the struaural simplicity argument, both outgroup comparison and the ontogenetic criterion (G. Nelson 1978) suggest a priori that certain "simple" characters of pipoid and rnicrohylid larvae are plesiomorphic for Anura. Salamandersand caecilians lack labial teeth, and keratinized jaw sheaths are questionable in salamanders (perhaps only in sirenids), so ou&roup analysis suggests their absence is plesiomorpluc for frogs. These jaw sheaths and labial teeth appear later in ontogeny than do infiarostral and suprarostral cartilages, so by the ontogenetic criterion their absence is plesiomorphic. By either criterion, Starrett's assumption was correct. However, such a priori hypotheses are tested by other data in the context of phylogenetic analysis. That is, the bestfitting phylogeny, based on all of the relevant data (Kluge 1989),will corroborate certain evolutionary transformations but reiect others. An~arent coficts in trees derived from differekt data sets often resolvable if one analyzes the character data on which the trees are based (,Miyamoto 1985). Starrett's hypothesis of the primitiveness of Type 1 larvae was not corroborated by phylogenetic analysis of other data. Analyses of primarily adult characters (Cannatella 1985; L. S. Ford and Cannatella 1993; J. D. Lynch 1973) indicated that the Type 3 larva is plesiomorphic, and that absence of jaw sheaths and labial teeth is derived within frogs " and is an e;olutionary loss. More recently, a phylogeny from larval morphological characters (Haas 1997a; fig. 4.2) also supported the placenlent of the Type 3 tadpole as plesiomorphic for Anura. In contrast, the phylogeny based on mitochondrial 12s and 16s nucleotide sequences (Hay et al. 1995; fig. 4.2) suggested that the Type 4 tadpole is plesiomorphic for Anura. However, a combined analysis by Kjer, Graybeal, and Cannatella (unpublished data) of morphological characters and this sequence data yielded a tree that supwrts that of L. S. Ford and Cannatella (1993). This analvsis also i~ldicatcd much of the inconmence is because of the rooting position of the molecular Free. The continued use of Orton's tadpole types diverts attention from issues of morphological'evol~ion to issues of definitions. L. S. Ford and Camatella (1993) discussed the problem of the Tjpe 2 (microhylid) tadpole, whose definition was strained after the discovery that tadpoles of at least one species of scaphophqnine microhylid have jaw sheaths (Wassersug 1984, 1989b). None of the four tadpole types is homogeneous in structure or function, and the acceptance of Orton's groups as monolithic categories will ordy hinder our understanding of the tadpole evolution. Morphology evolves, and one would expea that definitions of the larval types wiU become fuzzy as more data are available.

sways

ANATOMY
Viscera and Endocrines
Bruno Viertel and Susanne Richter

Introduction
An anuran larva is a nonreproductive, highly specialized stage within the complex, amphtbian life cycle and not simply an extended embryonic phase that concludes with metamorphosis. A tadpole is a mosaic of organs that are either remodeled (e.g., exocrine pancreas and posterior alimentary traa; see Pretty et al. 1995; F. Sasaki and Kinoshita 1994; Yoshizato 1992) or degenerated during metamorphosis (e.g., blood-forming organs, flter apparatus, integumentary ceiis, lymphatic system, and pronephros), remain functionaiiy quiescent (e.g., parts of the excretory and reproduaive systems), or are functional throughout both larval and adult stages of the cycle (e.g., liver, lungs, pancreas, spleen, thymus, and thyroid, and the vascular system). The nonreproductive tadpole feeds and grows to produce a metamorph that has some probability of entering the breeding component of the population. In that context, much of the anatomy of a tadpole involves the gathering and processing of food, and many of the structures used in these aaivities, especiaiiy involving the buccopharyngeal area, are unique among vertebrates. We summarize the anatomical diversity of the circulatory, digestive, respiratory, urogenital, and endocrine systems of tadpoles. Much of the data available on these subjeas is derived from studies of only a few genera (e.g., Bufo, b n a , and Xenopus); when known, information on embryonic development and metamorphic changes is included. Additional information on some subjeas is found in chapters 3, 4, and 6. We used the staging tables of the original authors; translations of selected staging systems to that of Gosner (1960) are shown in table 2.1. The foiiowing staging systems are cited i this chapter and are presented by the author name(s) n and the appropriate arabic or roman nunieral(s): Cambar and Gipouloux (1956a, Bufo bufo), Cambar and Marrot (1954, Rana dalmatina), Cambar and Martin (1959, Alytes obstetricam), Gailien and Houhon ( 1951,Disco~lossuspictzls), Gosner (1960, ali generalized anuran larvae), Nieuwkoop and Faber (1956, Xenopus laePir), Rossi (1959, Bufa bufo), Sedra and Michael (1961, Bufo rgularis), Shumway (1940, Ranapipiem), and A. C. Taylor and Kollros (1946, Ranapipiem). The terminology we used in defining periods of development are as foiiows: embryonic = Gosner 1-24 (zygote to closure of operculum); larval or tadpole = Gosner 2 5 4 6 ; premetamorphic = Gosner 25-35 (growth of trunk and tail); prometamorphic = Gosner 3 6 4 1 (growth of iimbs, thyroxine titers increase), and metamorphic climax = Gos-

The s m d , apparently insignificant, arnphibian tadpole is a wondrous mechanism of organic complexity that is of profound significance in the study of basic biological problems . . .
Fox 1984:x

ANATOMY

93

dium and together with blood flow affects the epithelial folds. Cells migrating between the rnyocardium and endocardiurn stiffen-the flds and forrn a &e. The conal valve changes sequentially (fig. 5.1E-G) from dorsal, to left, to Circulatory System ventral positions, and hally comes to lie near the ventricle The organs of the circulatory system are derived frorn rneso- at the proxirnal end of the conus (Shumway 25). At Shumderrn, and the heart and viteiiine vessels are the first flly wav 24 an endothelial double-lavered inter&ricular sevtum fnctional embryonic organs. The circulatory system trans- divides the atrium. By ~ h u m w a i 2 5 auriculo-ventricular the ports rnaterials between organs with extemal contaas (e.g., valves are connected with the interauricular septum, the sialimentary tract, lungs, giils, and skm) and those with inter- nus venosus is distina. trabeculae have develo~ed the venin nal fnctions (e.g., endocrine, excretory, muscular, nervous, tricle, and the wall of the ventricle is spongy. The heart does and reproductive systems). The lyrnphomyeloid systern is an not develop frther until &er rnetarnorphosis. Bending and immunological buf'ier between the external and cellular envi- twisting of the heart tube is a typical feature in tetrapod heart ronments of the larva, especially for those organs that come develo~rnent. anurans the twisting. of the ventricdar reIn u into contact with an aquatic environment rich in micro- gion and the transitional region of ventricle and conus conorganisrns. stitute essential changes, whereas the sinus venosus retains its position. The conus approaches the atrium (fig. 5.1H-I). Heart In the fllv develoved larva. the heart is dorsal to and The histogenesis of the heart is similar in all vertebrates bounded lterally the psterior branchial arches. In (Hirakow 1989; Hirakow et al. 1987, Cynopspyrrhogaser; adults, the heart tilts more anteriorly than in the lama (fig. Hirakow and Sugi 1990, Gallus;Volkov 1982, R. &bunda; 5.1H-K; Benninghoff 1921, Bombina parigata, Bu$ bufo, also Burggren and Fritsche 1997; Farrell 1997; Icardo Leptodaqlus pentadactylus, R. catesbeiana, R. escubnta com1997). Heart-forming potency is expressed only in the dor- plex, and R. temporaria; Ekrnan 1924, 1925, R. temporaria; sal lip of the blastopore and in deep mesoderm between 30' Erdmann 1921, Bombina and Rana; Goette 1875, Bombina; and 45" lateral to the dorsal midhne of the gastrula. Trans- Ison 1968: Nieuwkoov and Faber 1956. X. h s ; Prakash plantation experiments show that heart rnesoderrn is estab- 1954 and'Thornas 1672, ~oplobatracbistigerinis; Rugh hhed by a dorsalizing signal frorn the dorsal blastopore lip 1951). (Sater and Jacobson 1990,Xenopushis) well before the end of gastrulation. In contrast to the results of A. G. Jacobson Arteries and Ve'eins and Duncan (1968, Caudata), Sater and Jacobson (1989,X. In Xenupus h s , elongated cells appear in the lateral plate h) showed that the pharyngeal endoderm does not have near the inner laver of skin at Nieuwkoo~Raber31-32. , I' an induaive effea and does not promote the onset of heart They migrate frorn the lateral plate and engulf and surround fnction. the blood celis, and by Nieuwkoop/Faber 33-34 they are flat In Rana temporaria (Shumway 1940), cells from the dor- and in desrnosornal contact with each other. At Nieuwkoopl sal edge of the visceropleura forrn a single endocardial tube Faber 35-36 these cells are found throughout the ventral that is continuous with the endothelium of the ventral aorta side of the ernbryo, and by stage 37-38 they have extended by Shumway 18 (fig. 5.1A-G). The rnyocardial anlage origi- toward the endoderm to forrn the blood vessels of the vennates from mesoderrn at the ventral edge of the lateral plates tral intestinal tract (Mangia et al. 1970). (somatopleura and visceropleura; C0penhaver 1955 and The forrnation of ao& arches and vulmonarv arteries at DeHaan 1965). The dorsal part of the ventral edge folds the sixth aortic arch and the anatomical displacement and around the endocardial tube frorn anterior to posterior. At reduction of the aomc arches during metamorphosis have Shumway 18 the heart is a longitudinal tube consisting of been studied extensively (figs. 5.2-5.3; see Delsol and Flatin the conus arteriosus, ventricle, and sino-auricular cornplex 1972 and references therein). In contrast to the arterial sys(i.e., fture sinus venosus and atrium). At this time, the peri- tem of the pharynx, little anatornical change occurs in the cardium develops as a folding of the more lateral soma- arterial svstem of the trunk and limbs. The subclavian (foretopleura around the rnyocardium. Dorsally (Shumway 20) limbs), rnesentrial (intestines and kidneys), and iiiac arteries the myocardium attaches at the dorsal pericardium in the (hind limbs) branch sequentiallyfrom the dorsal aorta posteregion between the conus arteriosus and the paired aortic rior to the pharynx. The dorsal aorta extends into the tail as trunks.The ~aired vitelline veins come into contact with the the caudal arterv (see Rhodin and Lametschwandtner 1993). , , undivided auricular region and form the sinus venosus. Tra- and this vessel degenerates at rnetarnorphosis. Rernodeling of the venous systern occurs before and durbeculae forrn shortly &er the heart begins to beat at Shumn-ay 20. Up to Shumway 22, the ernbryonic heart tube is ing rnetamorphosis prirnarily in the postcranial region (figs. situated ventrornedially and posterior to the pharynx. At 5.4-5.5). The rnedial posterior cardinal veins fse to form Shumway 23 the right, left, and posterior distal conal valves an unpaired interrenal vein; the lateral posterior cardinal and the septum coni differentiate frorn the endotheliallining veins forrn and supply the opisthonephros with blood indeof the conus arteriosus at the bfircation of the aortic tninks. pendently of arteries. As such, they represent the rnain drainThe endothelium grows rapidly cornpared to the rnyocar- age systern for the caudal and iiiac veins (i.e., the lateral posI
i
x

ner 4 2 4 6 (remodeling and degeneration of rnany larval structures, thyroxine titers reduced).

BRUNO VIERTEL AND SUSANNE R I C H T E R

H
C

Q++p
v

@: 4
V

AKm
RA VA

: ; &c

v
Fig. 5.1. Development of the heart of Rana temporaria. Cross sections of (A) plate of ceiis that wiii form the 7, endocardium of the heart (Taylor/Koiiros stage 1 ) (B) dorsai mesocardium (Taylor/Koikos 18), (C) endocardium continuous with the endothelial lining of the aortic arches (Taylor/Koiros 18), and (D) septum coni originating anteriorly as a continuation of the nght distal conai vaive (section through conus septum). Three horizontal sections (E-G) from dorsai to ventral, interauricular septum developing from the anterodorsal w d (Taylor/Koiiros 24). Twisting of heart tube at early (H) and later (I) stages. Right lateral view of the heart at Taylor/Koiiros 25 (J) and in the adult (K). Abbreviations: A = auride, AS = interauricular sepnim, C = conus, DM = dorsai mesocardium, E = endocardium, EP = plate of ceiis that wiii form the endocardium of heart, LA = left auricle, LAT = left aortic arch, LDCV = left distal conai valve, M = myocardium, P = floor of pharynx, RA = nght auricle, RDCV = right dorsai conai vaive, S = sinus venosus, SC = septum coni, SM = splanchnic mesoderm, V = ventnde, and VA = ventral aorta. Orientationai crosses: A = anterior, D = dorsai, L = left, P = posterior, R = nght, and V = ventral. From Ison (1968); reprinted by permission of Taylor & Francis, Ltd.

ANATOMY

95

terior cardinal veins are the afFerent renal veins). One of the flow into the anterior and only persistent ventral commistwo omphalomesenteric veins forms capiiiaries in the liver sural vessel and thereby drain the hind lirnbs and t d . The and, passing caudaily, drains the intestine via the newly head-pharynx region drains via the external (especiaily the formed portal vein; the lefi omphalomesenteric vein drains flter apparatus) and internal jugular veins. the stomach region via the gastric vein. The posterior vena cava arises anteriorly from the hepatic vein and posteriorly Blood from the interrenal vein. A capiliary network connects the Turpen and Knudson (1982, Ranapz$iens) described hemaabdominal and musculo-abdominal veins. Blood flows pos- topoietic precursor ceiis in the late gastrula and early n e d a teriorly via three ventral cornrnissural vessels to the inter- (fig. 5.6). Celi aggregates in these plates of precursor ceiis renal vein, then to the hepatic vein (i.e., the posterior vena differentiate to hematic cords of the blood islands and oval, cava), and from there to the sinus venosus. During meta- primitive blood cells. Precursor celis identical with common morphosis, a blood sinus connects to the larval subintestinal plunpotent hematopoietic stem ceUs (PHSC; Broyles et ai. vein (from the lefi omphalomesenteric vein) and forms the 1981 and Cltne and Golden 1979) migrate from the dorsal adult abdominal vein; blood flows anteriorly, opposite to part of the lateral plates into the larval hematopoietic organs that of the larval abdominal vein. Femoral and iliac veins (Broyles 1981; Turpen et al. 1979; Turpen and Knudson

A
H/

Anterior cardinal vein (jugular)

..W' .. ,

Internal carotid artery

Bulbus arteriosus icle

Anterior calinal vein (jugular) Dorsal aorta Ductus Cuvieri Subclavian vein

\.
A

+C
1 C

= -/I "

:-:.: . ,

\ \

_Ductus arteriosus

Externa1 carotid

artery / Pulmonary artery Posterior vena cava

'

// / tyC
-:.. . ............ ....:. . .. . . . ..: .. -... .... .:.::....:. : ......
L

Ventricle

Hepatic vein Sinus venosus

Right auricle

Fig. 5.2. Blood vascular systems of an early (A) and late (B) frog embryo viewed from the nght side. From Rugh (1951); reprinted by perrnission of McGraw-Hi, Inc.

96

BRUNO VIERTEL AND SUSANNE RICHTER

FA BRANCH III

FA BRANCH IV TO SUBRECT III

SUBRECT 111

1982). Hemoproteins, first detectable in the anterolateral blood islands, diflerentiate in the lateral plates at Nieuwkoop/Faber 33-34. Ln subsequent stages, hematic cor& extend into the posteroventrai trunk region, and by Nieuwkooppaber 39 the anterior cords have disappeared. At Nieuwkoop/Faber 40 the subintestinai veins contain hemo-

proteins (Mangia et al. 1970). The intertubular liver and hepatic ducts, pro- and opisthonephros, and spleen are primary centers of hematopoiesis (Broyles et ai. 1981, R. catesbeiana; G. Frank 1988,1989a, b, R. esculenta complex; Maniatis and Ingram 1971a, b, c, R. catesbeiana (fig. 5.7). In Ranapipiens (Shumway 24-25) and other species, hematopoiesis begins

ANATOMY
Fig. 5.3. Ventral views of the development of the aortic arches ofXeno-

97

p laevis. (A) In a 5-mm embryo, aches 1, 3, and 4 are complete; the

other two have primary and secondary vessels. (B) In a 6-mm embryo, aches 3,4, and 5 are complete; arch 1 is degenerating at point x; aches 2 and 6 are stumps.-(c) In an 8-rnm larva, secnd&y vessels of arches 3 and 4 loop backward into externai gis; nght side (&in d&ram) is unusual in having two externai giiis on branchial arch 1, the second of which is supplied by an extra loop formed by the anastomosis between arch 3 and its secondary vessel. (D) Arrangement of aortic arches and arterial supply of a t e r apparatus (m$pled) in a fourlegged larva before completion of metamorphosis; note relative siw of aortic arches; arch 5 is starting t o degenerate (dushed line). (E) Reconmuction of aortic arches, branchiai skeleton, and associated musdes in a 10-mm larva. Vessels supplying extemal gills are starting t o atrophy, and arch 3 is wider than others; arteries to filter apparanis were omitted. Abbreviations: ANAST = anastomosis, AORTA = aorta, BAS = basihyal, BRANCH I - N = branchial bars I-TV, CAR EXT = arteria carotis externa, CAR INT = arteria carotis interna, CER = ceratohyal, FA BRANCH I-TV = arteries of filter apparatus of branchiai bars I - R GC 11-TV = gi clefts I-IV, GEN = musculus geniohyoideus, ILC = inferior labial cartilage, MECK = Meckel's cartiiage, CITERHY = musculus interhyoideus, INTERMAND = musculus intemandibularis, ORBHY = musculus orbitohyoideus, QUADHYrLUG = musculus quadrato-hyoangularis, RECTCERV = musculus recnis cervicis, SUBRECT I-IV = musculi subarcuales recti I-n TRVENT I1 = musculus transversus ventralis 11, numerals 1-6 = aortic arches 1-6, OCCVERT = arteria occipito-vertebraiis, PAL = artena paiatina, PDA = paired dorsai aorta, PR 3 4 = primary vessel of aomc arches 3-4, PULM = arteria pulmonalis, RMUSC = arteria carotis externa, ramus muscularis, SEC 3 4 = secondary vessel of aortic arches 3 4 , SUBCL = arteria subdavia, and T R = t m c u s arteriosus. From Millard (1945); reprinted by permission of the Royal Society of South Africa.

about 100 days in premetamorphic R. catesbeiana (Forman and Just 1976). Spleen In X. h s , the spleen anlage forms at the dorsal mesentery near the anterior end of the stomach by Nieuwkoop/Faber 43 and is well-defined by stage 4 5 4 6 . Blood corpuscles are visible at Nieuwkoop/Faber 47, lymphocytic diierentiation is detectable at stage 48, and the organ is highly vascularized by stage 49. Large celis in the white pulp (lymphoid foiiicles) do not mature into medium or large lymphocytes before Nieuwkoop/Faber 50 or into s m d lymphocytes before stage 51 (Manning and Horton 1969; Turner and Manning 1972, 1974). At Nieuwkoop/Faber 50 "degenerating macrolymphocytes" are present in the boundary between the red and white splenic pulp. The spleen aniage of Ranapipiens appears at Shumway 24-25 and contains erythrocytes, mesenchymal celis, and yolk platelets at Taylor/Koiiros I. At Taylor/Koliros I1 the aniage consists of s m d masses of mesenchymal and elongated reticular ceiis interspersed with erythrocytes, large and medium lymphocytes, and a connective tissue capsule. At Taylor/Kollros I11 s m d lymphocytes, neutrophils, and granuiocytes occur in the mesentery near the spleen, and the spleen begins to increase in size. The white pulp is formed during Tay1orF;ollros VIII-IX. The red pulp merges smoothly into the white puip as in the aduit spleen (J. D. Horton 1971a, b). The spleen is an important secondary lymphoid organ with significant reticuloendotheiiai functions. Phagocytic c e h of the reticuioendotheliai system of aduits are situated in the red pulp and grouped around the white pulp (E. L. Cooper and Wright 1976, Ascaphus mei; and Diener and Nossai 1966, B u . marinus). Macrophages prevent large particles and other ceiis from entering the white pulp. Turner (1969) ascertained that free macrophages of the body cavities have the same funaion in X. b i s .

in the pronephros (Broyles and Deutsch 1975 and Broyles and Frieden 1973, R. catesbeiana; K. L. Carpenter 1978, K. L. Carpenter and Turpen 1979, Foxon 1964, Hoiiyfield 1966, and Jordan 1933, R. pipiens; Meseguer et ai. 1985, R. ridibunda; and Turpen et ai. 1979, Ranapipiens), but the majority of larvai erythocytes is produced in the liver (Maniatis and Ingram 1971a, b, c, and Turpen et ai. 1979, R. Lymphatu and Reticuloendothelzal Organs catesbeiana). The increased chance of mfection caused bv bucco~hawnSalvatoreh (1970) distinguished four generations of geai contaa with the water that passes through it is counerythrocytesin Bufi b .. u Ceii diameters decrease in size from tered by the lymphatic and reticuloendothelial systems embryo to larva and increase at the beginning of metamor- (Aschoff 1924).The lvm~hatic. reticuloendothelial. and imphosis. In embryos, the ceiis are 250-750 p,m2 (Cambar/ mune systems are closely related to each other and to the Gipouioux I11 10), 150-500 p,m2in larvae (Cambar/Gipou- immune responses. Especidy in amphibians, lymphoid and loux IV 5), and 150-350 p,m2at the beginning of metamor- myeloid tissue appear together with macrophages and other phosis (Cambar/Gipouioux N 12). Rana catesbeiana (Tay- reticuloendotheliai ceiis (i.e., p h a g o q c ceiis) in the same lor/Kollros X-XII; Broyles et ai. 1981) has Type 1 cells organ. Lymph vessels drain body fluid from interstitial (26.920.03 [S.E.] p,m; dorninates in the kidney; contains spaces back to the veins and to lymph glands, where it is Td-4) that are oblong to oval with an acentric nucleus, and filtered ffirr. 5.8). The reticuioendothelial svstem consists of Type 2 (23.5020.04 p,m; dorninates in the liver; contains phagocytic macrophages in most organs and myeloid tisTd-l,2, and 3) cells are eiiiptical to round with the nucleus sues, which diierentiate into eosinophils, erythrocytes, maccentered (Benbassat 1970; Hoiiyfield 1966; McCutcheon rophages, neutrophils, and thrombocytes. During ontogen1936). In contrast to V. I. Ingram (1972), Broyles et ai. esis, stem ceiis popuiate many stmctures (e.g., Ijmph glands, (1981) concluded that there were different erythrocytes lin- thymus, ventral cavity bodies, branchial and aortic arches, eages (Type I) emanating from the kidneys and (Type 2) the intertubular tissue of the kidneys, intestinal w d , liver, lungs, liver. The distribution of erythrocyte classes in the vessels mesenteries, skin, and the w d i of the bile and hepatic duas; varies individually within a species, and these ceii survive for E. L. Cooper 1966 and Witschi 1956). E. L. Cooper (1967,
L ,
i

"

T
OMR ANAST SUB SUB HEP

ABD

OML MPC

MPC DUOD

CAUD GAS

M
RAC SUBC

. MP
PVC

ST

ISCH

'OP

,
I

ABD

COMM

"'

CAUD -

3,& J [ "Pr"
...^ ..
<

...... .... . ..'.. . . . '.... : . :.:... . .:: ......... .... , ..: .. . . . . . . :. . ', ;:
'

ISCH RAP

Fig. 5.4. Development of the venous system ofXenopus laevis. Diagrammatic representation (not to scale) of the development (ventrai view) of the omphalomesenteric veins and hepatic portal system in (A) 5, (B) 6, (C) 6.5, (D) 7, (E) 8 4 0 mm larvae, and (F) metamorph. Diagrammatic representation (not to scaie) of the development (vmtralview)of the postcardinal veins and posterior vena cava in (G) 6, (H) 8, (I) 10, (J) 12, (K) 14, (L) 24-mm larvae, and (M) metamorph. Diagrammatic representation (not to of scale) of the development (ventralh) the abdominal and hid limb veins in (N) 7, ( 0 ) 10, (P) 18-24, (Q) 37-mm larvae, (R) early metamorph, and (S) adult. Reconstmction of the alimentary canal and its blood vessels in (T) 6.5, (U) 8, and (V) 24-mm larvae. Abbreviations: ABD = abdominal vein, AC = anterior cardinal vein, AFF = afferent vein, AGAS = anterior gastric vein, ANAST = dorsai anastomosis around gut, AVC = anterior vena cava, BRAC = brachiai vein, CAP = capiiiary network, CAUD = caudal vein, COMM = commissurai vessels between abdominal and caudal veins, CUT = cutaneous vein, DC = duct of Cuvier, DUOD = duodenum, EFF = efferent renai veins, FEM = femorai vein, GAS = gastric vein, GASD = gastroduodenai vein, HEP = hepatic vein, IJ = internai jugular vein, INT = intestine, IR = interrenai vein, ISCH = ischiadic vein, LATV = lateral vein, LPC = lateral postcardinal vein, MAV = muscularis abdominis vein, MESS = mesonephric sinus, MP = main portal vein, MPC = mediai postcardinal vein, OML = left omphaiomesentericvein, OMR = right omphaiomesenteric vein, OP = opercular vein, PH = phacynx, PItS = pronephric sinus, PVC = posterior vena cava, RAA = ramus abdominalis anterior (anterior anastomosis), RAP = ramus abdominalis posterior (posterior anastomosis), RECT = r e m , RP = renai portal vein, SOM = somatic veins, ST = stomach, SUB = subintestinai vein, SUBC = subdavian vein, and SV = sinus venosus. After Miilard (1942); reprinted by permission of the Royal Society of South Africa.

BRUNO VIERTEL A N D SUSANNE RICHTER

Dorsolateral plate rnesoderm

Colonization of Interstitial ; , pronephros migration ., Post. cardinal Dorsal aortae

'.,

B-Lymphopoiesis (1) Granulopoiesis (2) Monopoiesis

'S.

II
(deterrnination) (4) Gastrula

Thyrnus T-cells (3)

' - . - - r ~ i r c u ' l '4t* n i ~ io Mesonephros S~leen Liver , Liver+ Etyihropoiesis (5)

\
Stage 12 -14-

(determination)

Ventral blood island rnesoderrn

Viteiiine veins

Circulation / ~mbryonic erythropoiesis (6,7)

I

15 - -17 -18 16

-19 -20 - 21,22,23 -24,25-

Fig. 5.6. Postulated relationships and migratory patterns of hematopoietic precursor ceiis during embryonic development in Ranapiphns. Dashed lines indicate events not documented experimentally. Stages 12-25 (embryonic period from gastnila to hatching) foiiow Shumway (1940) and stage I (beginning of larval period) is of Taylor and Koiiros (1946). Numerais in parentheses refer to the foiiowing refer-

ences: 1 = Zettergren et ai. (1980), 2 = K. L. Carpenter and Turpen (1979), 3 = J. D. Horton (1971b), 4 = Fales (1935), 5 = Turpen et al. (1979), 6 = Hoiiyield (1966), and 7 = Turpen et ai. (1981a). From Turpen and Knudson (1982); reprinted by permission of Academic Press.

STAGE
15 18-19 24-25 X-XII

Metamorphic climax XXI-XXI II

o - & - - Area of anterior presumptive mesonephric

Ventral 0 0 islands

Liver pronephros (mesnephros?) , , , ,

Mesonephric

PHSC

Primitive RBCs (Hb type?) $1%-4)

Irnrnature RBCs T Y P2~ (Hbs Td-1 , Td-2, Td-3)

Ranapipiens and R. catesbeiana. Arabic numerals 15-25

Fig. 5.7. Ontogeny oferythropoiesis and hemoglobin synthesis in = Shumway stages and roman numerals X-XXIII = Taylor/Kollros stages. Abbreviations: Hb(s) = hemoglobin(s), PHSC = pluripotent hematopoietic

stem ceiis, RBC = red blood cells, and Td-1 to Td-4 = hemoglobin g l o b u h fractions. From Broyles (1981) and Turpen et al. (1979); reprinted by permission of Plenum Publishing Corporation and Academic Press.

1976), J. D. Horton (1971a, b), and Manning and Horton (1982) provide terminology for the lymphatic and endotheliai systems. The development of the trunci lymphatici lateraies corporis occurs immediately after the cranial lymphatic hearts form at Gosner 18 (Rana esculenta complex). At Gosner 20 the lateral lymphatic sacs (vasa lymphatica caudaie ventraie

and dorsaie) are visible. The hemolymphatic system has separated into blood vessels and lymphatic vessels at Gosner 24; lymphatic heart vaives are present, and lymphatic vessels and lateral lymphatic sacs are present dorsaiiy and around the eyes. By Gosner 25 (fig. 5.8) the vas lymphaticum caudaie laterale (VLCL) and the trunci lymphatici laterales corporis (TLLC)have appeared and paired ductus thoracici are

ANATOMY

101

VLS LLS ALH

,LLS

VLCL VLCD VLCV

ular ceils predominate (fig. 5.9A). Parts of the cortex are surrounded by connective tissue capsules, and blood capiaries begin to penetrate the dorsoventrdy elongated m&s. The increasing tendency of branching blood capillaries to push through the cortex increases the blood supply, and this invadand. In 25-27 dav-old larvae. consion forms lobes in the u centric cytoplasmic striations found in a small nurnber of these cells resemble those in Hassali's unicellular corpuscles in the cvstic soaces of the medda. Uniceilular and multicelI lular Hassall's corpuscles are conspicuous in the m e d d a at Tay1orF;ollros ni) from stages V-XVI, the organ increases in size to a maximum of about 1 x 2.5 mm (J. D. Horton 1971a, b). The persistence of the thymus throughout amphibian development suggests a favorable reaction to thyroxine or synthesis of a hormone that counteracts thyroxine (E. L. Cooper 1976).
i

Lymph Glunds and Ventral CavityBodies The term lymph gland (LM1, lyrnphomyeloid organ 1; dorsal or subthymic tonsils of R. temporaria, corpus lymphatrior lymph heart, D N H = dorsal net of head, DNT = dorsal net of trunk, L U = lateral lymph sacs, TLLC = tmncus lymphaticus lateralis icurn ofR. esculenta, dorsai cavity body ofR. temporaria, dorcorporis, VLCD = vas lymphaticum caudale dorsale, VLCL = vas lym- sai gi rernnant of Bufo and Rana, and glandula interposita phaticum caudale laterale, VLCV = vas Iymphaticum caudale ventrale, or inclusa of ~opluba&a~hus tzaerinus, Raka catesbeiana,li. ocand VLS = ventral lymph sac. After M. H . Hoyer (1905a). cmkntalis, R. ridibunda, and Rana sp.; E. L. Cooper 1967) was first used by Witschi (1956) for Rana catesbeiana and, after J. D. Horton (1971a, b), it was used for R. pipiens and differentiated. Caudal lymphatic hearts appear at Gosner 27. eventudy for all anurans. These glands disappear at metaAt Gosner 4 1 the paired ductus thoracici enlarge as sinusoi- morphosis in R. catesbeiana and probably other closely reda1 spaces, and the lymphatic vessels of the trunk are reorga- lated anurans. Authors who have described lvmoh ~ l a n d s in , L U niz.ed as lymphatic sacs at stage 43. At the same stage one aduits may have looked at a different lymphomyeloid organ pair of craniai lymphatic hearts and two pairs of caudai lym- (e.g., J. D. Horton 1971a, b). phatic hearts develop. At Gosner 45, there are three caudai In 9-12 day-old larvae of Rana pipiens (Taylor/Kollros Iymphatic hearts (M. H . Hoyer 1905a, b). Kampmeier 11), groups of irythrocytes, lymphoq%s and netrophils oc(1915, 1922, 1958, 1969) supplied additional information cur laterai t o the forelimb aniagen. Each lymph gland has on the anatomy of lymphatic vessels and hearts. close contact to similar cells in the pronephric region. Dors d y the lymph gland anlage extends int the anterior lymThymus Gland phatic space (division of the primary maxillary sinus; KampThe thymus, a dorsai, ectodermal (J. D. Horton 1971b) de- meier 1922) and contacts the dorsolateral wail of the rivative of the second visceral pouch (fig. 5.9A-C), is visible opercular chamber ventraily. Structuring of the anlage into at Shumway 22-23 in Ranapipiens. At Shumway 23 the an- ceilular cords enclosing sinusoids is complete in larvae of lage consists of round and columnar epithelial ceils with yoik Taylor/Koilros 11. Lyrnphocytes and reticular fibers occur platelets and scattered pigment. At Shumway 24-25 the thy- in the glandular parenchyma, while granulocytes occur in rnic anlage lies ventrolaterai to the auditory capsule and pos- blood sinuses lined by reticuloendothelial cells (phagocytic terior to the eye and is attached to the epithelium of the first littorai ceils; Bacuh and Cooper 1968, R. catesbeiana). gdi slit (Sterba 1950). At the same level, laterai branches of The lymph glands grow rapidly until Taylor/Kollros I11 &e dorsai aorta course more proxirndy or mediaily, while and now lie anterolateral to the forelimb anlagen. Lymphothe anterior cardinal vein extends dorsally to the thymic cytes, macrophages, and elongated reticular ceils occur in the buds. Posteroventral to the bud, the first efferent branchial parenchyma; eosinophils, erythrocytes, lymphocytes, and arterv extends to the branch of the dorsai aorta. At Shum- neutroohils occur in the sinusoids. A connective tissue capway 25 the aniage, situated dorsolaterally at the level of the sule sukounds the gland so that a lymphatic space appe&s. auditory capsuie, increases in size and fibroblasts condense At their maximum size, lymph glands (Taylor/Kollros Xaround it. XII) appear reddish and lie anterodorsolateral to the base of In Ranapipiens at Tay1orF;oiiros I (J. D. Horton 1971a, the forelimb. They are connected lateraily and medially to b), the thymus enlarges but remains attached to the visceral epithelium that limits and penetrates the anterior lymphatic pouch. At Taylor/Kollros I1 growth and differentiation ac- space (J. D. Horton 1971a, b). Tadpoles ofXenopusl& lack celerate and the thymus begins to detach from the visceral lymph glands (Manning and Horton 1969,1982). pouch. In 14-19 day-old larvae, lymphocytes of d diameThe ventral cavity bodies (LM2, lymphomyeloid organ ters are present in the medda, although paie-staining irreg- 2) originate as invaginations of the epidermis of the opercuFig. 5.8. Lymphatic vessels and spaces in Rana esculenta. (A) Lateral xiexv. (B) Ventral view. (C) Dorsal view. Abbreviations: ALH = ante-

102

BRUNO VIERTEL AND SUSANNE RICHTER

lar chamber waii (= peribranchiai chamber w d of Viertel 1991).The first pair appears between days 1619 in R. pipiens (Taylor/Koiiros 1 ) at the leve1 of the second giii slit and 1 Lies in close contaa with the filter apparatus and gill irrigation current. The bodies consist of nodular outgrowths underlain by connective tissue with nurnerous capillary branches. Lymphocytes and neutrophils occur in the capillaries and extravascularly in the surrounding tissue. Up to four pairs of ventral cavity bodies have formed by

Taylor/Koiiros 111. The most anterior lies mediai to the posterior corner of the eye, and two of the posterior pairs are Large positioned half-way up the opercular w d (fig. 5.9B). round or oval cells that resemble hemocytoblasts (Cowden et ai. 1968)or monocytes (Rouf 1969)are concentrated in the central region of the nodule connective tissue. Up to eight pairs of ventral cavity bodies may develop by Taylorl Koliros N. pairs are in the lateral and ventral opercular Six chamber waii. Bodies positioned lateral to the heart and

ANATOMY

103

Kg 5.9. (A) (Rilular details of the dorsal region of the right thymus
16-day-old Ranapt@m larva (Taylor/Kollros 11). The cortex and d u i i a are beginning to differentiate; smaJi lymphocytes can be recogrzed in the cortex; and paler staining, more irregularly shaped epithefLI cells predominate in the central, meduiiary zone. From J. D. H m o n (1971b). (B) Ventral branchial region of a 68-day-oldRana larva (Taylor/Kollros X) showing (low magnification) the location of a number of ventral cavity bodiis project&g from the ventral md lateral opercular w d s (peribranchial chamber wds) of the right side. The epitheiia of aii these nodules are lymphoid. The ventral cavity bdies shown here are diffusely populated (i.e., central connective tisnie t-egions contain only a scattered array of leucocytes). Densely popuhtcd ventral cavity bodies are also found in leopard frog larvae i t mrious stages of development. From J. D. Horton (1971a). (C) Right chymus of a 97-day-oldRanapip>iem larva (Taylor/KolirosXXIII) showing (low magnhcation) the extent of this organ during metamorphosi. The ratio of meduiia to cortex has increased since earlier larvi life, and rhe densely staining cortical zone with its mass of s m d lymphorrj appears lobuiate because of inwardly projecting connective tissue ~ p t aFrom J. D. Horton (1971a). (D) The extent of the intembuiar . ephomyeloid tissue of the right opisthonephros of a 68-day-old RaMppiew larva (Taylor/Koliros X). The section passes through the more posterior part of the kidney whose ventral portion consists of a mass of hematopoietic tissue. From J. D. Horton (1971a). AbbreviaWns: AC = auditory capsuie, C = capsuie, E = epithelial cells (ineguiady shaped), GA = giii arch-second branchial, HT = hematopoietic -e, MOW = medial w d of opercular chamber, MY = myotomal m u d e , OC = opercular chamber (peribranchialchamber), SK = skin, and SL = s m d lymphocytes. ~ e ~ r & t bydpennission of the Wistar e institute Press and AWmn Zwh&.
&-a

esophagus are not always paired. At Taylor/Kollros VIII-IX, a maximum of four pairs lie lateral to the heart and esophag s and others are at the ventral and lateral opercular wa u (fig. 5.9B). The highest number of pairs of ventral cavity bodies (15) is reached between Taylor/Kohos X-XII, and by stage XIV this number has deched (Ranapipiens). A maximum of five pairs occur in the was of the opercular cavity (E. L. Cooper 1967, Rana catesbeiana; J. D. Horton 1971a, b). Tadpoles ofXenopus hhave three pairs of ventral cavity bodies that resemble those of R. catesbezanu. One pair lies ventral to the posterolaterai sides of the first branchial chamber between the first and second branchial arches and dorsal to the m u s d u s subarcuaiis rectus 11. The second pair lies along the m u s d u s subarcuaiis I11 close to the point where the second branchial chamber opens into the opercular chamber near the spiracular openings. The third pair is simated at the cartilaginous base of the third branchial arch cartilage between the openings of the second and third branchial chambers near the spiracle. Based on their position, the procoracoid bodies of Sterba (1950, Prokmacoidkmpev) appear identical to the first and second pairs of ventral cavity bodies (Manning and Horton 1969). Prontpbros and 0pZrthont.pbros Neutrophils and blast cells occur in the pronephric intertubular tissue in Ranapipkns only afier the onset of erythropoietic activity (Shumway 23). The complete ceii repertoire includes eosinophds, various lymphocytes and neutrophils.

Lymphocytes are present in the intertubular tissue of the opisthonephros. By Taylor/Koilros I11 the percentage of intertubular lymphomyloid tissue ha.5 increased, especially granulocytes in the lateral and anterior regions of the pronephros. The pronephros and opisthonephros grow through Taylor/Kollros N, by stage VI1 the pronephros begins to decline. and At Taylor/Koiiros XVI the lymphomyeloid tissue of the pronephros consist mainly of granulocytes (fig. 5.91; E. L. Cooper 1967, R. catesbeiana, and J. D. Horton 1971a, b). The opisthonephros continues to develop up to Taylor/Kollros XVI. The interstices of the pronephros and opisthonephros and the mesenchymal sheath of the pronephric duct are granulopoietic (G. Frank 1988, Rana escubnta complex). Equal densities of eosinophils and heterophils are found in the pronephros and opisthonephros (F. R. Campbell 1970; Curtis et al. 1979; G. Frank 1988). G. Frank (1988) concluded that the opisthonephros has a lymphomyeloid interstitium. The only ceiis found in the mesenchymal sheath of the pronephric duct are eosinophils and heterophils of equal densities (G. Frank 1988). Tadpoles ofxiinopus W s lack the lymphomyloid tissue in the pronephros (Manning and Horton 1969). At Taylor/Koiiros I11 of R. pipiens, sma lymphocytes and a few neutrocytes occur both intra- and extravascularly near the fibrous lamina propria and esophageal epithelium. M e r Taylor/Kollros 111, lymphocytes accumulate in many areas of tissue with the sma intestine and cloaca. Lvm~homveloid , lymphocytes and granulocytes associate with capiiiaries and lymphatic spaces in the skm near the hind lunb buds, and a pair of dorsal lymphomyloid accumulations near the myotomes are retained until prometamorphosis. Identical tissues are found in Xenopw near the limb anlagen (Nieuwkoop and Faber 1956). Between Taylor/Koiiros V and IX,lymphomyeloid tissues ( g d remnant, LM4, lymphomyeloid organ 4; E. L. Cooper 1967) associated with the oral (lateral wa), pharyngeal (ventromedial wa), and dorsal branchial regions(ventrd to thymus) appear. These typicay paired structures are similar histologicay to the lymphomyloid tissue of the esophagus. In the branchial region, two paired lymphatic accumulations appear at Taylor/Kollros V ventrolateral to the anterior parathyroid (visceral pouch 3) gland near the second gill clefi. A lymphatic space separates each pair from the anterior parathyroid. The posterior pair lies close to the posterior parathyroid glands (visceral pouch 4), is connected to the epithelium of the h r d gill clefi, and appears first in Taylor/Kollros VI. According to Sterba (1950) identical lymphatic accumulations are fond in Xenopw. Epithelial body is the term used by a other authors for lymphomyeloid organ 5 (LM5) of adults near the parathyroid glands. Ofien, the term epithelial body is used to describe the parathyroid glands together with the LM5 l p g on top of them. These lymph tissues increase in size and form paired bodies as the larva develops. Two pairs lie opposite the tissue from which the jugular bodies of adults form during metamorphosis.
L

104

BRUNO VIERTEL AKD SUSANNE RTCHTER

Sma accum~dations lymphocytes and granulocytes ap- larva. During metamorphic clunax, cutaneous arteries and of pear from Taylor/Koliros X i and around the blood capillar- veins increase in diameter, subcutaneous arteries and veins n ies near the g d s and branchial arches and in the posterior become numerous, and a second venous network is develpharyngeal cavity at the leve1 of, and medial to, the lymph oped in the form of anastomoses between the subcutaneous gland. T h s tissue contacts the blood vessels in the region veins (De Saint-Aubain 1982, Bufi bufo and Rana temporaof the branchial arches and continues to grow until Taylor/ ria; also sec Strawinski 1956, Rana esculenta comph). Kollros XVI. Du Pasquier (1968, 1970, Alytes ~ b s t ~ c a n r ) Merkel cells are found in R. temporaria from Cambat-/ found large accumulations of lymphoid tissue in the esopha- Martin 33-36 and in X. lamis during Nieuwkoop/Faber 49gus and smalier lymphatic iniltrations in the giiis, intestine, 50. These s m d cclls occur in the tail and body as well as in lungs, and rncsentery. G. Frank (1988) showed the existente the tooth ridges of R. japonica and the barbels of &nopus. of granulopoietic tissue in the mesentery and in the mesen- Synaptic contacts with nonmyelinated nerves (Whitear chymal coat of the bile and hepatic duas in the larvae of the 1983) suggest that barbels may be mechanoreceptive. R. esculenta complex. The mesenteric artery and arterioles are Membrane-bound Merkel granules originating from the surrounded exclusively by granulopoietic tissue, and smaer Golgi complex and located dose to the synapses may contain accumulations of granulocytes are attached to the bile duct. various components that are released into the cytoplasm (Fox 1984; Fox and Whitear 1978; O v d e 1979; Tachibana 1978). Mitochondria-rich celis (Type A cells of Hourdry Respiratory System 1974) in the tail and trunk (Ranatemporaria, Canibar/MarTadpoles use the general body surface, lungs, and pharyngeal tin 33/34) have a high concentration of microridges on the organs for gas exchange. The low capacitance coefficients of apical pole and many mitochondria and may be stimulated 0, and CO, in water compared to air demand that a water by oxytocin (D. Brown et al. 1981). Type B cells Beaumont breather ventilate 28 times as much as an air breather under (1970) also have a high mitochondrial density and microvilli the same conditions (Prosser 1973). The ontogenetically and may transport substances (Alvarado and Moody 1970; early differentiation of the lungs enables many tadpoles to Maetz 1974) or have an osmoregulatory function. They rebe air breathers. Gas exchange is effected automatically with semble die cldoridc celis of fish (Whitear 1977), but it is the locomotory movcmcnts of at least the tail (Wassersug unclear whether Type B celis are precursors of flask cells of 1989a) if the body surface is used for respiration. Under adult amphibians (Fox 1985, 1986). Stij2chenzeLlen (= pin some hypoxic conditions, some tadpoles use aerial respi- cells) found in the dorsal tail fin cpidermis of tadpoles of Rana (CambarlMartin 28-29) but absent in adults are probration. ably chemosensory (Fox 1985,1986; Whitear 1976). These Integument and P@mentatwn pear-shaped cells with microvilli on the apical poles are The epidermis of premetamorphic Iarvae consists of an outer thought to contact nerve fibers. periderm and an inner sensory layer (B. I. Balinsky 1981; Mucous or goblet cells are known in X. laevzs (Gartz Fox 1985; Gerhardt 1932; E. Marcus 1930). Numerous 1970; Le Quang Trong 1974; Pflugfelder and Schubert populations of teus make up the larval anuran epiderrnis, 1965; Schneider 1957) andR. temporaria (Fox 1972,1988). and considerable changes occur at metanorphosis (e.g., Ka- Leydig mucous cells (Leydig 1853, 1857; "clear cells" of wai et al. 1994). Three epidermal layers occut during pro- Fox 1988) present during prometamorphosis are rich in tometamorphosis and about six cell layers are present after noflaments and sparsely populated with granular ER and metamorphosis (Fox 1977, 1984). Epidermal cells are con- mitochondria. The mucous ceils that Elias (1937), Le nected by tight junctions and desmosomes and have little Quang Trong (1973), Poska-Teiss (1930), and Sato (1924) ultrastructural differentiation other than fiiarnentous and fi- described as Leydig cclls in the Bufonidae were classified as brous structures (Fox 1977; Junqueira et ai. 1984, Pseds Riesenzellen (= alarm cells) by Fox (1988).The Akgelzelle (= ~aradoxa). Dense arravs of tonofilkents are associated with ball cell or unicellular gland; Meyer et al. 1975a, b) occurs abundant a& filaments near the surface membranes of cells in high densities in Ascaphus twei until metamorphosis and of the sensory layer. Collagen in thc basement lamella first uncommonly in X. luevis and Hymerwchims boettgeri (Frohappears in Raha temporaria-at Cambar/Martin 25. An adepi- lich et al. 1977; Meyer et al. 1975a, b). Pfeiffer (1966) called dermal space with lameiiated bodies, the adepidermal mem- these ceiis Riesenzellen in the Bufonidae. The mucus of some brane and collaeen fibrils of the basernent lamella form un- of these cells is acidic and extrusion is merocrine. Riesenzelder the membrane. This system is remforced by fibrils in the len, originaliy descnbed in Bufo bufo and B. calamita (Wenig 1913), that develop deep in the epidermis in young Xenopus tail (Fox 1985). Epidermal vascularization (subepidermal capiiiaries) is are not homologous to Leydig cells (Fox 1988). The rnusparse in premetamorphic larva. Head, back, and tail have cous cuticle is not an evaporation barner in larvae or aquatic the smaliest capillary net mesh size and are therefore consid- adults but probably functions as an ion trap in larvae, aquatic ered to be the most active respiratory parts of the body sur- frogs, and fishes (Friedrnan et al. 1967). Epidermal melanoface (De Saint-Aubain 1982, Bu$ bu$ and Rana temporaria). phores up to 500 pm in diameter form a network between In general, the role of skin in gas exchange is not as signi- epidermal cells and are primarily involved in color change ficant i larvae as one might expect from their large body (Bagnara 1972,1976; Nikcryasova and Golichenkov 1988). n surface. The capiilary net bccomes dense in prometamorphic Various immigrant cells (e.g., mesenchymal macrophages,

ANATOMY

105

granulocytes, lymphocytes, leucocytes, and phagocytes) Dermal gland ceiis proliferate in the epidermis and mimove to the evidermis but are not conneaed to the ceils grate to the-dermis. ~ h are found in th body and at the e ~ by desmosomes. base of the limbs inX. laeYis (Nieuwkoop/Faber 57 onward), Although various ceils are found in the dermis, there is a in the dorsal skin of R. pipkns at the beginning of prometahigh concentration of connective tissue and capdaries. In morphosis, and in R. temporaria at the beginning of metathe t , d muscle tissue penetrates deeply into the dermis. Mel- morphic climax. Two types of secretory ceils surrounded by anocytes, or melanophores, like iridophores and xantho- myoepithelial-and mitochondria-rich ceils on a basement phores, are derivatives of the neural crest (Bagnara et ai. membrane similar to the adepidermal membrane of the epi1979; B. I. Balinsky 1981). Melanophores are not con- derrnis are included. Secretorv ceiis of the mucous ~lands neaed to other ceiis by desmosomes and move amoebalike have a high density of granular ER and electroniucent vesibetween epidermal and dermal ceils. When the ceiis are in a des; secretory ceils of the granular glands have dense grandis~ersedstate. numerous dendritic arms extend between ules and microvilli at the apical pole. The ducts of the granuthe ceiis, and the pigment is brownish. In a contraaed state, lar ~lands establish contact with the surface at Tavlor/Koilros the ceiis are about a f of their dispersed size and are xv?~, and the mucous gland ducts contact thi s&ace in i a black. The average ceii size is 300 pm, and melanosomes, Taylor/Koilros XIX (Deifino 1977; Deifino et ai. 1982). organeiies within the cells that contain melanin, measure 0.5-1.0 pm (J. D. Taylor and Bagnara 1972). Melanin is a complex polymer of tyrosine derivatives and proteins (Bag- Many authors consider the gds, branchial system, or nara et al. 1979). There is a typical distribution of melano- branchial ar& to consist of the dorsal filter plates (= gill cytes for each species. ~isco~iossids form an adepidermal filters), as weii as the branchial food traps (= crescentric ornet of melanophores (G. Andres 1963). gans) and ventral g d tufts. The g d system is supported by Indophores occur in large numbers in the body and less branchial cartdage, which is a visceral arch, and most authors frequently in the tail. They are up to 300 pm in diameter describe it as a branchial arch. The derent and efferent and contain crystais of adenine, guanine, and hypoxanthine branchial arteries and the musculus constrictor branchialis as reflecting stacks of platelets in organeiles evidently derived extend beside the arches. from endoplasmic vesicles (Bagnara et ai. 1979). In transMcIndoe and Smith (1984a) discussed the respiratory mitted lighi, iridophores exhibc the refractive structural col- function of filter plates and branchial food traps. They arors red, blue, and green depending on the position of the gued that the large netlike concentration of blood vessels on platelets relative to the light. In refleaed light they are gold the rows of filter plates, as weil as the placement in the irngaand silver (Bagnara 1976). tion current, indicate a respiratory funaion for these organs, ~anthophoiescontain fat-soluble carotenoids obtained but they recognized that certain features chaiienge this asfrom the diet and pteridines synthesized by the ceii (Bagnara sumption. The epithelia are thick relative to the g d tufts, et al. 1979) and stored in pterinosomes. Their color depends covered by mucus, and, as in the integument, the efferent on the Dattem of carotenoids and ~teridines that often oc- blood stream is drained via venous vessels and has no concur in the same xanthophore (Bagnara 1976; Bagnara et ai. neaion with the heart via the branchial arteries. They sug1979). Xanthophores are rare in premetamorphic larvae. gested that the low PO, blood supply via the afTerent Erythrophores with carotenoid vesicles and perinosomes at branchial arteries probably means that the filter plates extraa th ceii periphery are more numerous in adulis than in larvae their own supply of O, from the surrounding irrigation curor juveniles. rent. McIndoe and Smith (1984a) argued that the g d tufts Larvae generaiiy adapt to light conditions by changing are the more signiicant organ in gaseous exchange. color primarily through dispersion, contraaion, and migraThe terms external and internal ~ d do not reflect the acs tion of melanophores. Dermal melanophores surrounding tual positions in the embryo. Prior to the closing of the operthe iridophores isolate them from the light and cause the cular fold, ail gds and their anlagen are external. The terms larva to darken. The cells contraa by withdrawing the den- "outer" and "inner" gds generaiiy apply oniy to Gosner 23dritic Drocesses and come to lie in &e oroximal Jermis be- 24. A ~ a r from their ~osition t relative to the branchial arch. low the iridophores and xanthophores. Light transmission the fundamental diffkence lies in the fact that the proxima and reflection are particularly influenced by epidermal mela- gds persist, whereas the dista1 (= outer) gds atrophy by the fold nophores. The state of expansion of melanophores and xan- time the o~ercular closes. Viertel 1991 considered the thphores and the posit&n of the plateletS in iridophores terms persistent and transient gds to be more exaa than inis influenced by the melanophore-stimulating hormone ner and outer gills. Persistent gis develop along the ventral (MSH) of the pars intermedia of the pituitary gland. In hy- and transient epidermal giiis along the ventrolateral parts of pophyseaomized larvae, melanosomes are contraaed, plate- branchial arches I-n! A giii tuft (fig. 5.10) consists of an lets of the iridophores are dispersed, and xanthophore pteri- epidermal covering, dermd conneaive tissue, areas occupied dine and carotenoid content is low. Thyroid hormones by individual endodermal-pharyngeal ceiis, and the blood antagonize MSH, and catecholamines, epinephrine, estro- vessels and mesodermal ca~iiiaries.Low temDeratures and gen, norepinephrine, pineai melatonin, progesterone, and high 0, pressures stirnulate short gds while high temperatestosterone aiso influente pigmentation (Fox 1985; also see tures and high CO, pressures stimulate the growth of longer Hayes and Licht 1995). gds (bvtrup and Pign 1968).
V

B R U N O VIERTEL A N D SUSANNE R I C H T E R

cavity and the dorsal cavity of the trachea. At Nieuwkoopl Faber 39 the laryngeal anlage is visible as a horizontal ridge separating the tracheal cavity from the buccopharym. The opening of the tracheai cavity to the esophageal and gastric cavity closes by Nieuwkooppaber 40. Laryngeal cartilages differentiate in Nieuwkooppaber 43, and the glottis perforates after stage 44 (B. I. Balinsky 1981; Nieuwkoop and Faber 1956, X. W). Nieuwkooppaber 51-52 the lung At e~itheliumfolds, A d comective tissue and blood vesseL surround the lung sacs. The lungs expand posterodorsolaterd y and by Nieuwkooppaber 54 extend to the pronephros. In Rana ridibunu (Shumway 21), ventrolateral evaginations of the esophagus are visible anterior to the first nephrostome, and the esophagus closes soon thereafter. By Shumway 23 the lung sacs extend beyond the pronephros, and the lanmgeal chamber is visible. The lateral and ventrai Fig. 5.10. Diagram of a gill tuft showing pattern of connections of aiTerent and efferent branchial arteries with primary and secondary tuft w d s of the Gophagus form the aditus laryngis anlage. The vessels. Abbreviations: numerais 3 and 4 = higher order branches, musculus dilator laryngis opens the aditus laryngis at about ABA = afferent branchiai arteries, CEB = ceratobranchial cartilage, Shumway 26, but laryngeal cartilages are not present at this COB = coracobranchialis musde, EBA = efferent branchial arteries, stage (MUller and Sprurnont 1972). GT = giii tuft, TV1 = primary tuft vessels, and TV2 = secondary tuft The inner w d s of the lung sacs consist of a thin layer of vessels. From McIndoe and Smith (1984); reprinted by permission of squamous epithelium underlaid by thin, smooth muscles. Springer-Verlag. The few blood vessels are not always in close anatomical cont a a with the smooth muscle layer; capiaries are sparse. The Mitochondria-rich cells (Beaumont 1970) in the gi epi- outer surfaces of the lungs and trachea are surrounded by Both these and the blood vessels are rich dermis probably are responsible for the high adenosinetn- the visceral ~leura. phosphatase activity (Boonkoom and Alvarado 1971, Rana in melanoqes. Septa do not occur in premetamorphic larcatesbeiana) linked to the active transport of ions. The epi- vae, but during and after metamorphic clirnax, septa develop dermis also is permeable to O,, CO,, and water (Bentley and as infoldings of the lung epithelium and the smooth muscle Baidwin 1980). Comective tissue is well formed at the base layer. Blood capiaries grow into the septa between the two of the gi tufts, and the presence of endodermal cells from smooth muscle w d s . The smooth muscles thicken at the the visceral pouches shows that the gi tuft is part of the end of each septum (Atiunson and Just 1975; Goniakoweaodermai-endodermal transition zone. In contrast, there ska-Witalinska 1986). Strawinski (1956. Rana escuhnta is very little dermis and connective tissue in the distal areas complex) found that 'the capiary nitwork becomes denser of the g d tuft so that capdaries contact the sensory layer during metamorphosis. Histological examination of the preof the epidermis. Blood is provided by four aortic arches metamorphic lung clearly iustrates its limited role in gas (= branchiai arteries). The afFerent branchial arteries supply exchange. A nearly complete absence of capiaries, a smaii blood from the heart, while three caudal efferent branchiai surface area, and a low capacity for seif-ventilation suggest arteries end blindly. Branches of the primary tuft vessels that the tadpole lung is a precursor of the adult lung. Lungs from the branchiai arteries branch into secondary tuft ves- function as respiratory organs when the 0, concentration sels. Shunts bridge between arterial and venous capiaries in of the water drops or when the buccopharym and gis the g d s and connea afFerent with efferent tuft vessels (fig. are blocked during intensive feeding. The primary function 5.10; De Saint-Aubain 1981, 1985, Rana temporaria and of lungs in premetamorphic larvae may be related to Bufo bufo;McIndoe and Smith 1984a, Litoria ewingi). These buoyancy. The trachea begins posterior to the laryngeai chamber, authors assumed that the shunts maintain the flow of blood during the degeneration of gi tufts and capdaries. It seems which is a bilobed cavity connecting the glottis to the pnlikely that vasoconstriction of the shunts may regulate the mary bronchi or lung buds (Rugh 1951). The laryngeal amount of blood that flows through the capdaries and thus chamber is supported lateray by the laryngotracheai card u e n c e aquatic gaseous exchange. In species of the R. escu- tilages and surrounded ventromedidy by the transversus lenta complex, there is surprisingly little vascularization of and dilator laryngeus muscles (Sedra and Michael 1957). the g d tufts and filter plates. Giii regression begins when the The tracheai epithelium of the developed larva is one cell operculum grows over the persistent gills. thick (Nieuwkooppaber 44). A substance formed from fibroushke bodies extruded bv the e~ithelial cells lines the inLun.s and Trmhea ner margin of the trachea and bronchus (Nieuwkoop/Faber Lung anlagen (Nieuwkooppaber 35-36, Xenopus laevis) ap- 47, thickened at stage 4 8 4 9 ) . Folds in the tracheal epithepear as two ventrolateral pocketlike evaginations of the ante- lium are penetrated by the fibrous substance. Epithelial cells rior foregut. These primordia originate from a transverse are charaerized bv their s m d number of sma. a~icallv , sitridge that separates the buccopharynx from the intestinal uated mitochon&ia. They also contain lipid granules, traa, and a horizontal ridge divides the ventrai primary liver smooth vesicles, rough endoplasmic reticulum, Golgi com, L

ANATOMY

107

plexes, and a high density of nbosomes in the matrix (Fox

et al. 1970, 1972, Xiinupus h Fox et ai. (1970) found ) .

no ciiia in the tracheal system of Xenops, but ciiia occur in R. temporaria at metamorphic climax. It rernains unclear whether the lameilar bodies in the lung sacs of adult Bufo buf and Hyh arbmea (Goniakowska-Witalnska 1984, 1986) are identical to the fibrouslike substance in IarvalXenopus. The outer surface of the trachea is surrounded by connective tissue and blood vessels.

Digestive System
Most tadpoles are suspension feeders (Wassersug 1972), and the flter apparatus enables them to trap a wide range of particulate food. Tadpoles lacking keratinized mouthparts fiiter particles suspended in the water colurnn. Tadpoles with keratinized mouthparts scrape or bite food from the substrate and fiiter the suspension. Many bottom feeders ingest periphyton, detritus and interstitial organisms from the sediment, and large quantities of indigestible material of different particle sizes (Seale et ai. 1982; Seale and Wassersug 1979; Viertel 1990, 1992, 1996). The buccopharynx consists of the buccal cavity (anterior buccopharynx) and pharynx (posterior buccopharynx). The pharynx develops as a filter apparatus by integrating anterior parts of the esophagus and lateral areas of epidermal eaoderm (Viertel1991,1996). Because of their position relative to the visceral skeleton and because the buccal cavity and pharynx cannot easily be distinguished from one another, these components are coilectively referred to as the buccopharynx (fig. 5.1 1). Wassersug (1976a, b) established the first comprehensive anatomical terrninology and made broad comparisons (Wassersug 1980) among members of eight farnihes (also Lannoo et al. 1987, Orteopilus bmnneus; Viertel 1982, 1984b,Alpes muletensis, A. obstetricans, Bombina variegata, Hyh arborea,

Pelobatesfuscus, and central European Bufo and Rana; Wassersug and Heyer 1988, Leptodactylidae). Other investigators concentrated on the ontogenesis of the tongue (Hammerman 1965, 1969a, b, c, R. catesbeiana and R. chmitans; H e H and Meiiicker 1941, R. sylvatim) or described the developmental changes and chemoreceptive characters of buccopharyngeal structures (Fiorito de Lopez and Echeverra 1989; Nomura et al. 1979a, b, R. japonica). Authors from the beginning of the last century did not realize that they were describing a fiiter apparatus. They lirnited their investigations to individual components (Dugs 1834; Huschke 1826; Maurer 1888; W K. Parker 1881; . Rathke 1832), and Goette (1874, 1875) even rejeaed the notion that the "internai giils" functioned as a filter apparatus. Oniy Rusconi (1826) and Boas (1882) realized that they were investigating a fiiter apparatus. Informative descriptions by Kratochwill (1933, R. dalmatina), R. M. Savage (1952, Buf bufo and R. temporaria and several microhylids), and Schulze (1888, Pelobatesfuscus) form a bridge to the view of this organ that has been established in recent investigations. The first descriptions of the surface of the branchial food traps examined by scanning electron microscopy (Wassersug and Rosenberg 1979) included members of nine families. Other papers that addressed the filter apparatus of speciic species include the following: Das (1994, frogs from South India); O. M. Sokol (1981a, Pelodpes punctatus); Viertel (1984b, Alpes muletensis, A. obstetricans, and Bombina variedata); Wassersug (1984, compared Scaphiophyne with larvai Types 2 and 4); Wassersug and Duellman (1984, eggbrooding hylid frogs), and Wassersug and Heyer (1983, leptodactylids living on wet rocky surfaces, Cycloramphusduseni, Thoropa miliaris, and Tpetvopolitana).The first investigations of the fiiter apparatus by transmission electron microscopy were by Viertel (1985, 1987, Bombina variegata, Bufo buf, R temporaria, and X. h s ) . The mouth, buccal cavity, pharynx, glottis, and esophagus are arranged seriaiiy only during the embryonic stages

Table 5.1 Abbreviations used in figures 5.11-5.16 and 5.25 ACI = aorta carotis interna, AFP = anlage of filter plates, AFPC = anlage of e p i d e d fold of peribranchial chamber (anlage of operculum), AGO = anlage of gonad, AL = anterior labium, AO + numerals = aomc arches (3,4, and 6, latter is lung artery), AOC = arteria coronaria, APOP = anlage of postnarial papilla, ATE = aniage of tuba Eustachii, ATEG = anlage of transient epidermal gills, BI-BIV = branchial arches I-IV, BA = bulbus arteriosus, BAR = barbel, BC = buccal cavity, BFA (black dashed outline) = buccal floor arena, BFAP = buccal floor arena papillae or pustulations (shorter than twice the diameter of the base), BP = buccal pocket, BRA = buccal roof arena, BRAP = buccal roof arena pupiae, CC = ciiiary cushion, CE = brain, CG = ciliary groove, CH = choanae, CHY = ceratohyale, C 0 = colon, C = connective tissue, I DU = duodenum, DV = dorsal velum, DVI-DVIII = dorsai velum 1-111, EP = epidemiis, ES = esophagus, E Z = ectodermal-endoded I transition wne, EY = eye, FB = fat body, FL = forelimb, FP = ilter plate, FPC = epidermai fold of peribranchial chamber (operculum), FPIFPIV = filter plates I-n! G = gdi, GL = glottis, GS = gdi slit with numerals, GZ = glandular wne, HL = hind limb, HM = hyoidean and mandibular muscles, H P = hypobranchial plate, HY = hyoid arch, IL = &um, IN = internasai plate, IP = infralabial papiae, LA = left atrium, LFP = lateral filter plate, LI = liver, LJ = lower jaw, LP = luigual papillae, LRP = lateral ridge papdae, LTR = labial tooth row, LU = lung, MA = mandibular arch, MG = manicotto glandulare, MGH = musculus geniohyoideus, MGU = midgut, MLB = muculus levator branchiae, MNG = mediai notch above glottis, MR = mediai ndge, MSH = musculus subhyoideus, MST = musculus stemohyoideus, MT = musculus transversus, NA = external nares, NVP = nariai valve projection, OC = oral cavity, OP = oral papiae, OPI = opisthonephros, OS = mouth, P = pancreas, PAP = prenarial arena pustulations, PCW = peribranchial waii, PD = pericardium, PET = peritoneum, PL = posterior labium, PONA= posuiarial arena, POP = postnarial papillae, PRNA = prenariai arena, PRO = pronephros, PRP = prenarial papiae, RA = right auium, RE = rectum, SC = secretory ceii, SPL = spleen, SPM = spiracle, ST = stomadeum, SU = spiculum of hypobranchial plate, TA = tongue anlage, TEG = transient epidemal gdis, TF = tail h, = tracnis olfactorius, TR = truncus arteriosus, UJ = upper jaw, VA = visceral TO arch, with numeral, VE = venuicle, VJE = vena jugularis externa, W = prevelar papiae, and W = ventral velum.

BRUNO VIERTEL AND SUSANNE RICHTER

FPIV

MLB

EP

I FP I1

~y

Fig. 5.11. Dorsolateral views of the alirnentary tracts of (A) Rana temporaria (Orton's larval Type 4, Gosner stage 36) and (B) Xenopus la&s (Type 1, Nieuwkoop/Faber stage 54). Right, dorsal halves of the heads lifted and rotated and abdominal waii reflected. See table 5.1 for abbreviations. Redrawn frorn =ertel (1985, 1987); reprinted by perrnission of Springer-Verlagand Gustav Fischer Verlag.

of anuran larvae (figs. 5.1 1-5.14). In U l y developed larvae, only the axis from the mouth to the ventral velum remains unaltered. The position of the pharynx in relation to the buccal cavity and longitudinal axis of the tadpole differs considerably among anuran larval types, and the position of the n glottis relative to the buccopharynx is not the sarne i ail larval types. In pipoid tadpoles (larval Type l), the pharynx is situated

lateral to the posterior extension of the buccai floor whose lirnits are defined posteriorly by the glottis. The longitudinal axis of the pharynx is much sliorter in ail other larval types so that the bucal floor is widest at about the rniddle of the buccopharynx (fig. 5.1 1). Each side of the ventral velum forms an angle of 45" to the sagittal plane (Rhinophrynus)or extends posteriorly at an angle of 55"-70" (Types 2, 3, and 4; fig. 5.11A). Between a quarter and a third of the ventral

ANATOMY

pharynx is covered by the ventral velum. The branchiai arches are more ~osteroventraithan lateral to the buccai floor. The glottis 6 f ~ h i n u p h ~ lies at the posterior extent ux of the buccai floor and, as is the case in the Pipidae, forms is posterior limit; it lies on the ventral velum and is cont neaed to it medidy. In microhylids, the glottis lies in the center of the buccai floor arena, while in larvai Types 3 and 4, the glottis is posterior and more or less covered by the ventral velurn between branchiai arches 4. Cdiarv e~itheliun~ surrounds the glottis, and although typical for'th'e esophagus, is found nowhere else in the buccopharynx. This suggests that the glottis has no direa contact with the posterior buccai floor. The morphology of the buccopharynx, its relations to the visceral skeleton and its musculature, and the role of the viscera1 skeleton as the pump for the irrigation of the fiiter apparatus (Gradweil 1970, 1971a, b, 1975b; Kenny 1969a; O. M. Sokol 1977a, 1981b; Wassersug and Hoff 1979) indicate how far the pharynx extends anteriorly. The positions

tadpole (Orton's larval Type 1, NieuwkoopJ Fig. 5.13. Xempw Faber stage 54) i ventral view dissected with peribranchial wall, perin cardium, and abdomen opened. See table 5.1 for abbreviations.

Fig. 5.12. Rana temporaria tadpole (Orton's larval Type 4, Gosner stage 36) in ventral view with ventral peribranchial wa, pericardium, and abdomen opened. See table 5.1 for abbreviations.

of the hypobranchiai plate, the ceratohyai, and the unpaired copula under the buccai floor show that the buccai floor is part of the anterior buccopharynx (figs. 5.11,5.14GD, and 5.15). Besides those cited speciicaily in this chapter, the following references include studies of the buccopharyngeai morphology of various tadpoles (Echeverra 1995, 1996, 1997; Echeverra et ai. 1992; Grillitsch 1992; Inger 1985; Khan and Malik 1987; Pelaz and Rougier 1990). The anterior buccopharynx of larvai Types 3 and 4 has the most advanced structurai organization of anuran larvae (figs. 5.15-5.16). DdFerentiation occurs in Bufo bufo between Gosner 25-28, and, beginning with the papdiae, regression of the anterior buccopharynx starts at Gosner 4142. Various authors have concentrated on the signiicance of the structures of the anterior buccopharynx forjarvai ingestion, systematic classification and understanding larvai evolution. Wassersug (1980) and Wassersug and Heyer (1988) dassified buccai structures according to family and genera and suggested evolutionary trends. Vierte1 (1982) wrote a key for the Central European species based on structures of the anterior buccopharynx. Wassersug (1980) linked the funaionai significance of the buccai papiilae with suspension feeding. Although it is not yet possible to describe the exact function of each structure, a sieving apparatus may emerge via the interplay between anterior buccai structures and

BRUNO VIERTEL AND SUSANNE RICHTER

ST

STAGES 17-18

OS

STAGE 21

STAGE 23

0.3mm

the ventilation movements of the entire buccopharynx (also and the tongue aniage is flanked on both sides by infialabial Viertel 1991). The lingual papdae, buccal roof arena papil- papillae (fig. 5.15). In the premetamorphic stages the tongue lae, and buccal floor arena papiliae may function as chemore- aniage forms a transverse bulge that changes in position and ceptors (Hamrnerman 1969a, b, c; H e H and Meiiicker form. In prometamorphic stages it becomes cushion-shaped 1941; Honda et al. 1992; Nomura et. ai. 1979a, b). and extends along the longitu&nal axis. Lingual papdae are The buccal floor begins posterior to the lower jaw sheath, differentiated on the tongue anlage during premetamorphic

ANATOMY

Fg.5.14. Ontogenesis of the anuran pharynx involving composite reconstructions of Bombinu variepzta,Bufi calamita, and Ram tempmaria (Gosner stages 17-23). (A) Frontal section at the plane of stornod m (generalized scheme of embryonic vertebrate pharynx, i.e., Gosner 17-18). (B) Frond section at the level of transient epidermal & and anlagen of filter plates (Gosner 21). Anlage of buccal floor arena arches upward b m e e n asterisks, visceral pouches 1-3 have broken through to form gll slits 1-3. Anlagen of transient epidermal gills are developing; sensory layer of epidermis overlying the endoderm of me visceral pouches (ectodermal-endodermaltransition zone) forms the anlagen of the filter plates. (C) Left side, frontal section at the level of transient epidermal g d s and anlagen of filter plates (same level as [B] but Gosner 23). The buccal floor arena is more strongly arched upward at the future sites of the mandibular and hyoid arches (= ceratohyal and hypobranchid plate). Epidermal fold of peribranchial chamber (= operculum) overgrowing the elongated transient epidermal &. Right side, frond section (plane higher than on left side) of transient epidermal gills, anlagen of filter plates, and esophagus. Venaal velum, buccal floor arena, and anlage of tongue reconstructed --...... h m SEM micrographs. Ventral velum is the posterior elongation of MNG the buccal floor arena. (D) Lateral view of the anlage of the filter apparatus (Gosner 23). Right half of head lifted and reflected, oral cavity Fig. 5.15. The floor of the buccal cavity of a tadpole ofBuJ% bJ% and penbranchid chamber opened, and epidermal fold of peribran(Orton's larval Type 4, Gosner stage 36). See table 5.1 for abbreviachid chamber removed. Symbols: arrows = anlage of transient @ S From =ertel(1982); reprinted by permission of E. J. Brill. I tions. 1-5, arrow heads = direction of water movement, boxed areas = anlagen of filter plates, dotted lines = sensory layer of epidermis, heavy Mack lines = peridermis of epidermis, numerals = visceral pouches, sipple = visceral skeleton, and vertical lines = endoderm. See table : 5.1 for abbreviations. FromViertel (1991); reprinted by permission of Springer-Verlag.
sages (e.g., two in Amis, Agaycbnis, Amtbeca, O t o d s Pelse p u ,

P ~ ~ d u sPtycbohyla, Rana esmlenta complex, and a ,

Slilirca; four in BuF, Gmtmtheca, central European brown h g s [= R a m spp.], R. catesbeiana, R clamitanr, and R. 9 1
d ; 6-10 more or less fused together in a comblike and

nnacrure in Bombina, Discoglossw pictw, Scapbu?phync, and TbDnrpa).In Alytes cistenmi, A. muhensis, and A. obsteh.icans, i authors report a number of small pustulations similar to rbose in Ascaphus and 0-2 lingual papillae. Wassersug and Hqer (1988) provided information on leptodactylids with 2.3, or 4-1 1lingual papillae; in several genera h g d papilk are absent, and by prometamorphic stages the surface of dx tongue has fungform papillae. The buccal floor arena (fig. 5.15) lies posterior to the umgue anlage and is flanked by papillae. The arrangement, kngth, and number of these papillae differ among genera md species. The number of larger papillae (ca. 300-400 pm maximum length, except about 600 pm in C Y O S S O ~ ~ W ) e e s from 10 (BuF) to > 80 (Ptychohyla leonhardFch~h.ei). Shorter pustulations (about 100 pm long) occur in roughly d o r m numbers and are not confined to the buccal floor arena, and in many species (e.g., Hylaarborea and R. lessom) &ese pustulations occur across the entire buccal floor as far as the buccal pockets. In this' case they are termed prepocket papillae. The buccal pockets, situated lateral to the buccal floor arena, are slits in the buccal floor that form a bypass between the lumen of the buccopharynx and the peribran- Fig. 5.16. The roof of the buccal cavity of a tadpole of Bufi buJ% chid (= atrial or opercular) chamber. These pockets are not (Orton's larval Type 4, Gosner stage 36). See table 5.1 for abbreviagill slits. The buccal pockets have lateral contact with the lat- tions. From Vlertel(1982); reprinted by permission of E. J. Brill.

112

BRUNO VIERTEL AND SUSANNE RICHTER

eral areas of the ventral velurn. The buccal floor terminates posteriorly at the ventral velurn (Goette 1874 and R. M. Savage 1952, 1955; gill cover plate of Schulze 1888, 1892; and anterior filter valve of Kratochwd 1933). The ventral velurn has a rnedial notch above the glottis that is particularly weii developed in some species (e.g., Hyla dendroscarta and H. phlebodes). On the rnargin of the velurn, particularly rnediay, there are several long, t h papiiiae (e.g., Alsodes, Alytes obstetrians,A. muletensis, Atelognathus reverberi;, Pelobates+cus, Ptychohyla leonhardschultzei, and Thoropa), short, broadbased papdae (e.g., European Bufo and Rana, Crinia tasmaniensis, Gdrtrotheca, H. arborea, Leptobrachium hasselti, and M@stolotis lignarius) or none (e.g., H. ebraccata, H. mtfitela, Osteopilus bmtnneus, and 0 . septentrionalis). The buccal roof (fig. 5.16) begins posterior to the upper jaw sheath and is divided into prenarial, postnarial, and buccal roof arenas. The posterior ierminus f the buccal roof is the glandular zone (Kratochwiii 1933; = dorsal food trap of Kenny 1969b) and the dorsal velurn; (Goette 1874; R. M. Savage 1952, 1955; = posterior filter valves of Kratochwiii 1933). A srna nurnber of pre-narial pustulations in the prenarial arena (e.g.,Alytes, Bombina, Bufo, and Pelobates) are arranged in a U-shaped field with a posterior opening and are more or less fused together at the base (e.g., Colostethus whymperi, Cyclorana australis, Hyla spp., Litoria alboguttata, and Osteopilusseptentrionalir).The internal nares (= choanae) lie between th ~renarialk d ~ostnarialarena and are direaed posteriorly in a V-shaped pattern (e.g., Alytes, Bombina, Bufo, Hyh dendroscarta, M~aelosiagoeldii, Osteopiand lus), directed anteriorly in an inverted V-shaped pattern (e.g., Pelobates and Rana), or are parael (e.g., Hyalinobatrachium jleischmanni, Hyh spp., hptodactylus knudreni, L. wgwri, andMamogenwglottus al$wi). The choanae are flanked by the narial papdae and the narial valve projection. One to five postnarial papiiae extend into the postnarial arena. In microhylids, Hyla arborea, H. ebraccata, H. phlebodes, H. sarayacuensis, and Rhimphlynus these papdae are not weli developed and are absent in Pelobates fiscus and several leptodactylids. The lateral ridge papiiiae (to ca. 800 prn long) are situated posteroventral to the postnarial papiiiae. They occur individudy (e.g., Bufo, Pseuabphlyne bibronii, and Rana) or forrn a flat planar structure fused only at the base (e.g., Odontophvynus occidentalis, Paratelmatobius lutzii, and Pleurodema ciwrea) or along their entire length (e.g., Alpes muletensis, A. obstetrians, and Bombina). In some species, only one lateral ridge papiiia occurs on either side (e.g., Pelobatesfiscus, some leptodactylids), but there are usuay 3 4 . The rnedial ridge is situated between the lateral ridge papdae. The base is broad, and the terrninus is shaped like a pointed lancet (e.g., Anotheca spinosa, Bufo spp., and Hyla mixe) or blunt (e.g.,Alytes spp., Discoglossus spp., and R fiscus). In some cases, the ridge is a flat structure with a wide base (e.g.,H. arborea, H. mtfitela, and Thoropa) with lateral or apical pustulations (e.g.,Alytes, Bombina, and Scaphiophlyne). The buccal roof arena is surrounded by papiiiae that rnay be short or long like the papdae of the buccal floor arena. The nurnber of buccal roof arena papdae varies frorn ca. 10 large ones in H. arborea to 3 0 4 0 in Crossodactylus. There are no

papillae in the buccal floor arena in Adenomera mamzorata and Heleophlyne natalensis. About 20 (Bufo)to over 100 (Pelobates and Ptychohyh) sn~a pustulations less than 100 prn high are found on the buccal roof in a species and the postnarial arena in rnost species. In Rhimphlynus dorsalis (larval Type l), lingual papiiiae are absent, the surface anterior to the buccal pockets is large, the buccal floor arena papdae are arranged in a posteriorly direaed arch, and papiiiae are absent on the ventral velurn. The lateral ridge papiiiae, the rnedial ridge, and pre- and postnarial papiiiae of the buccal roof are absent, and the buccal roof is not divided into pre- and postnarial arenas. Buccal roof arena papiiiae and pustulations are present. The buccal floor of rnicrohylids (larval Type 2) is surrounded by papiiae and lateral ndge papiae are present. Particularly weli-developed prenarial papiliae are found in Micvohyh berdmorei, and the narial valve projection is very large in M. heymonsi. The papiiiae of the buccal floor and roof arenas are arranged in an arc in M. heymonsz, and prepocket papiiiae are present. Infralabial papiiiae have been described only for M. heymonsz. Lingual papdae and the rnedial ridge are absent in a rnicrohylids. In pipoid larvae (Type l), the total width of the buccal floor arena is only a little larger than that of the oral orifice. It tapers just posterior to the rnouth and extends posteriorly as a narrow strip between the two ventral vela and terminates at the glottis (fig. 5.1 1B). Its structural organization cannot be cornpared to that of Types 2 through 4. There is no tongue anlage before rnetarnorphosis (see Channing 1984; Harnrnerrnan 1969a, b, c; Helff and Meiiicker 1941; Lannoo et al. 1987; Viertel1982,1984b; Wassersug 1976a, b, 1980, 1984; Wassersug and D u e h a n 1984; Wassersug and Heyer 1983, 1988). The filter plates and ciliary grooves are part of the poste- ) rior buccopharynx (i.e., filter apparatus; Gradweli 1972a, b, c, 1975b; Kenny 1969a, b; Kratochwill1933; R. M. Savage 1952, 1955; Viertel1984b, 1985, 1987; Wassersug 1976a, b, 1980; Wassersug and Rosenberg 1979). In larval Types 3 and 4, branchial food traps (= anterior filter valve of Kratochwdl 1933) and ciliary cushions (Viertel 1985; = pressure cushions of Kratochwill 1933) are developed as parts of the posterior buccopharynx (figs. 5.1lA, 5.17A, and 5.18A). Because of their functional signiicance (Gradweli 1970, 1972a, b, c), the ventral and dorsal vela are assigned to the filter apparatus (figs. 5.17A-E and 5.18). Although the persistent and transient gilis are not involved in suspension feeding, they are important in the ontogenesis of the filter apparatus (fig. 5.17C). The anatorny of the filter apparatus in Zmpus Ia&> (larval Type 1) dfiers considerably frorn those of Bombina variegata (Type 3) and Bufo 4, B. calamita, and Rana temporaria (Type 4). In Xempus, the ventral velurn is paired (= pharyngobranchial traa of Weisz 1945a, b; filter valves of O. M. Sokol1977a), and each haifextends parael to the longitudinal axis of the tadpole (figs. 5.11B and 5.19A-H); in larval Types 3 and 4, the single ventral velurn is positioned transversely (figs. 5.11A, 5.14C, D, 5.15, and 5.17). Branchial food traps projeaing ventray frorn the velurn via a transi-

ANATOMY

113

tional zone (Wassersug and Rosenberg 1979) probably are an integral part of the ventral velum in Xiinupus (figs. 5.18A, G, and K and 5.20F-G). In larval Types 1, 3, and 4, four pairs of branchial arches are attached posteriorly to the ventral velum; the first has exclusively posterior filter plates, the second and third have filter plates on both sides, and the fourth has only anterior plates. In Types 3 and 4 only a rim (fig. 5.18G, K) separates the branchial food traps from the filter plates. The surface of the fdter plates is enlarged by filter rows (fig. 5.18A-D). In Xenopus hmi the filter plates have a particularly large chondrified basal lamina (visceral skeleton; figs. 5.19BYG, and K) from which the jilter row pedicels branch out into the filter rows. In most larvae of ail four types the fdter rows fold into primary and secondary side branches and thereby separate smail filter niches from the fdter canals. The transversely oriented dorsal velum in larval Types 3 and 4 is the dorsal equivalent of the ventral velum. In Xequs, three pairs of dorsal vela (= velar vanes of Gradweil 1975c; pressure pads of O. M. Sokol 1977a) extend into the fdter cavities between the fdter plates (figs. 5.11B and 5.19AY B). It is not clear whether the dorsal vela of larval Type 1are homologous with the single velum of larval Types 3 and 4. In the latter, a glandular zone extends transversely anterior to the dorsal velum (Kratochwill 1933; = dorsal food trap of Kenny 1969a, b) and the posterolateral ciliary cushons. * ~ h e ciliary grooves enclose the ciliary cusiuons (figs. 5 . 1 7 4 E) on both sides posteriorly and continue into the esophagus. Cihary cushions are absent in Xequs, and the ciliary groove extends into the esophagus lateraily (figs. 5.11B and 5.194 D). Tadpoles that are not suspension feeders (e.g., carnivores) have highly modified buccopharyngealmorphology. The visceral skeleton of the macrophagous-carnivorous tadpole of Hymenochirus boetitgeeri is small by comparison with that of other pipids (P@a andXenopus). A chondrified basal lamina of the filter plates and chondrified supporting fdter row pedicels are lacking. The fdter rows are flat and do not branch or enclose the fdter canals. The surface of the a t e r plates is smaii, and ventral and dorsal vela are lacking ( 0 . M. Sokol 1962,1977a).The carnivorous larva oflepidobatrachw hmi
Table 5.2 Abbreviations used i figures 5.17-5.22 and 5.24 n

lacks manv oral structures. There are some vustulations on the buccal floor and roof, flat narial valve projections and a tongue aniage. The ventral velum, branchial food traps and fdter plates are lacking (Ruibal and Thomas 1988; Wassersug and Heyer 1988). Ai structures of the anterior l buccopharynx, apart from the tongue aniage and the free ventral velum, are lacking in the arboreal, oophagous larva of Osteopilus brunneus. Filter plates with branching dter rows (primary and secondary side folds) and branchial food traps of the pitted type are present. In the closely related nonoophagous 0 . septentnonalis, ali structures of the anterior buccopharynx are present, although they are not as M y developed as those of generalized hylid larvae (Lannoo et al. 1987). Wassersug and Dueliman (1984) noted that the seauence of Merentiation of the keratinized mouth~arts. buccal projeaions and papiliae, fdter plates and gilis, and ventral velum and branchal food traps in embryos of egg-broodiig hylids (Cryptobatrmhus, Flectonotw, Gmotheca spp., Hem@hractus, and Stqania) relates to developmental mode, supply of yolk, and respiration. Some buccopharyngeal structures are lacking because the embryos do not pass through the late developmental stages (e.g., Cryptobatrachus and Hem@hractus) in whch the anlagen of these structures are formed (= paedogenesis). Leptodactylids (e.g., Cyclmamphusdusenz, Thmopa miliaris, and T petmpolitana) that inhabit . steep, wet rocky surfaces resemble stream-adapted larvae, but they share buccopharyngeal charaaeristics with arboreal or semiterrestrial larvae. They have fewer buccal papiliae, no secretory ridges on the branchial food traps, and-an- exposed glottis (Wassersug and Heyer 1983).
L

Bucccrpharyng.ea1HNil0g.y and Ultrastructure A basic scheme (squarnous epithelium and bottle-shaped secretory ceils, SC1) applies to the ventral and dorsal vela in Xequs. The degree of Merentiation appears to be advanced compared to larvae of Types 3 and 4 (figs. 5.20A-H). The SC1 are arranged in rows, and their secretions flow into secretory grooves and onto secretory ridges. The secretory grooves are bordered by secretory ridges and random ceils of the squamous epithelium. The secretory grooves and ridges foliow the course of the ventral velurn. In the dorsal velum,

AC = apical cell, ACE = cerebral aniage, AFP = &age of filter plates, APEG = aniage of persistent epidermal gills, A W = aniage of ventral velum, BI-BIV = branchal arches I-ll( BFA = buccal floor arena, BFT = branchial food trap, BL = basal lamina, BLA = basal labyrinth, BP = buccal pocket, BRA = buccal roof arena, c = caudal, C = cilia, CA = carrilage, CC = ciliary cushion, CIC = ciliary cell, CL = capillary vessel, CT = connective tissue, d = dorsal, DV = dorsal velum, E = merocrine extrusions, ED = edge of filter plate 111, ENC = endodermal cell, EP = epidermis, EPC = epidermal cell, ER = endoplasmic reticulurn, ES = esophagus, ET = erythrocyte, EX = extrusion, EZ = zone of extrusion, FC = filter cavity, FN = filter niche, FP = filter plate, FPI-FPIV = filter plates I-R FPC = epidermal fold of peribranchial chamber (operculum), FR = filter row, G = gill, GL = glottis, GO = Golgi apparatus, GR = grana, GS = gill slit, IC = intercellular space, enlarged by fixauon and dehydration, LFP = lateral filter plate, LV = lipid vacuole, M = mitochondrion, MF = middle fold, MRC = squamous epithelial cells with microridges, MV = microvilli, N = nucleus, NENC = nucleus of endodermal cells, NEPC = nucleus of epidermal cell, NSIE = nucleus of sensory layer cell, NC = nucleus of endothelial cell, NCT = nudeus of connective tissue cell, NSC = nucleus of secretory cell, NSQC = nudeus of squamous epithelial cell, OS = mouth, P = peripheral pustulation(s), PC = peribranchial chamber, PD = pericardium, PE = periderm, PET = peritoneum, PG = pigment granule, PI = pigment cell, PP = pusdation at base of filter row, PQ = cadage of palatoquadrate, PS = primary side fold, R = rim, SC = secretory cell, SC1 = goblet cell Med with mucous vacuoles, SC2 = goblet cell after extrusion, SL = sensory iayer, SP = secretory pit, SPC = supporting cell, SQ = squamous epitheliurn, SQC = squamous epithelial cell, SQE = squamous epithelium, SR = secretory ridge, SS = secondary side fold, TEG = transient epidermal gds, v = ventral, V = ventride, VC = vacuole, W = ventral velurn, and W = yolk vacuoles.

BRUNO VIERTEL AND SUSANNE RICHTER

Fig. 5.17. Filter apparam of Ranu temponzria (Gosner stage 32). (A) Oral cavity opened showing buccal floor arena, buccal roof arena with glandular zone, and glottis; fiiter plates free; ciiiary area of ciliary cushion ( @ ) a &reaching the opening of the esophagus. Symbol: asterisks = planes of cross xctions B and C. (B) Cross section of the right side of the buccal floor arena and filter apparatus. (C) Cross section of ventral velum, velar edge, and posterior Mter apparatus on the right side. (D) Peribranchial chamber opened, posteroventral view; dose

connection of ciliary cushion and fiiter plates is distinct; waii of peribranchial chamber united with pericardium and peritoneum. (E) Proximai parts of Mter apparam removed, posteroventral view; position of ventral velum and dorsai velurn forming a slit; ciliary cushion and ciiiary groove (stipph)posterior to the dorsai velum. (F) Glandular zone of the buccal floor arena of Bufi buf0 with mucus secreting seuetory pits. Scale = micrometers. See table 5.2 for abbreviations. From Vierte1 (1985); reprinted by permission of Springer-Verlag.

ANATOMY

115

24-36 SC1 ceh secrete synchronously in a manner similar to that seen in the ventral velum of tadpole Types 3 and 4. The extmion zones are prominent, and the secretory pits are not sunken. A transitional zone (fig. 5.14) lies between the dorsal velum and the ciliary groove. The ciliary groove of X. hmis does not extend between the filter plates (figs. 5.11B and 5.19A) as in tadpoles of Types 3 and 4. The ciliary groove is composed of ciliary celis and columnar (SC3) secretory celis in the center of the ciliary groove and pearshaped (SC4) celis in the transition zone (figs. 5.19D-E). Goblet celis are absent. The ciliary groove is enclosed by squamous epithelium underlain by supporting ceiis in the transition zone. The differences between the vela and secretory grooves and ridges and their positions and structure among Type 1 versus Types 3 and 4 tadpoles aiso apply to the structure of the iiter plates. Multiple branches of the middle plate of Xenupw more frequentiy fold into primary and secondary side folds that greatiy increase the surface area. As a result, filter niches are more numerous and the fdter canals form a more ramified system than in larval Types 3 and 4 (figs. 5.18A-D, 5.191-J, and 5.21C). The majority of organs in the fdter apparatus are construaed according to a uniform principie: bottle-shaped glandular celis are embedded in a single-layered squamous epithelium supported by connective tissue. These secretory ceils stand in groups and empty into secretory pits. This scheme applies to the dorsal and ventral vela, the glandular zone, and the branchial food traps (figs. 5.17E, 5.18G-J, 5.19C, and 5.20). Only inRrmphus, Rhinupblynw, the Discoglossidae, and Pelobatidae is the pattern of distribution of the SC1, and therefore of secretory pits, as disparate as in the ventral margin of the velum. 1 ; Rhinupblynus, the epithelium forms wavelike folds. In microhylids and aii Neobatrachia the secretory cells are arranged in rows and their secretion zones are situated on the exposed secretory ridges. In the Neobatrachia, SCl and their secretion zones are always grouped together. Fiiter plates, ciliary groove, and ciliary cushions in larval Types 3 &d 4 are org&zed differentiy from that described for Type 1 larvae. Arched iiter plates and the presence of middle and primary and secondary side folds increases the surface of the filter plates (figs. 5.18A-D, 5.191-J, and 5.21C). Two overlapp-ing layersof squamous epithelium are supported proximaiiy by connective tissue and the visceral skeleton. Apical celis protrude from the epithelium and form the distai lunits of the middle and side folds and also the lateral pustulations. The ciliary groove and ciiiary cushions consist of ciliary celis and goblet ceh (SCl and SC2; figs. 5.18E-F and 5.24A, B). Mucus is present on the ventral velum, branchial food traps, fdter r ~ & ,and ciliary cushions. The ultrastructure of the ciiiary groove, ventral vela, and dorsal vela in a Type 1 (Xenupush s ) tadpole is surnmarized fi-om Viertel(1985,1987). The SCl are taiier than in larval Types 3 and 4, and abundant endoplasmic reticulum (ER) is found around the nucleus. The apical pole is densely filled with granules. The SC3 celis are rich in mitochondria from the Gddle of the celi to the apical pole where vacuoles and extrusions occur. The nucleus is situated at the celi base. The

SC4 secretory ceiis are pear-shaped, rich in vacuoles, and densely iiied with ER except at the apical pole, the surface of the secretory zone is smaii, and their nucleus is unusuaiiy large. The ciliary celis, apicai ceiis, and squamous epithelial cels do not differ from those of larval Types 3 and 4. Examples of Types 3 and 4 larvae include Bombina variegata and Buf buf, B. calamita, and Rana tempmaria, respectively. Roughly bode-shaped SCl are rich in elearon-dense granules and vacuoles. There are many mitochondria in the extrusion zone of the apical pole and many longitudinal microtubules. Microvilh maintain contaa with the interstitial spaces, and a labyrinth of shaiiow folds forms the proximal limit of the ceiis. The shape of the ceii changes considerably accorduig to its vacuole content. Goblet celis, or goblet-celiiike mucous ceiis (SC2 secretory celis) lack microtubules, mitochondria and ER in the apical pole (fig. 5.21A). Mitochondria are found at the periphery of the celi, particularly proximaiiy in the mucous zone. The presence of both iiied and empty goblet ceiis indicates that mucus is being synthesized rapidly and extruded abruptly (fig. 5.21B). Ciliary celis are funnel-shapedwith a broad, flat, ciliated apical pole; the ceiis are rich in mitochondria and have an electron-dense qtoplasm caused by a high granular density (fig. 5.21A). The nucleus is basal. Squamous epithelial celis are especiaiiy flat in the fdter plates, and more or less triangular in cross section in the branchiai food traps and ventral velum (fig. 5.20). The nucleus is situated at the widest point and the cyoplasm is electron-dense. There is a single layer of vacuoles along the entire apicai surface of the ceil, and there is evidence of extrusion. Apical ceiis are papia- or ridge-shaped with the nucleus situated at the celi base and surrounded by a thck layer of ER (fig. 5.21C). Mitochondria and the Golgi apparatus are located in the middle and apical parts of the ceii. There is a high density of variously sized granules, a smaii number of vacuoles, and evidence of extrusion. Ontogenesis of parts of the fdter apparatus was described by Ekman (1913, Bombinapachypus, B u . bu., Hyh arborea, R esculenta, and R. tempuraria), Gegenbaur (1898-1901, Bombina), Gerhardt (1932, H. arborea and Pelobatesfscus), Goette (1874, Bombina), E. Marcus (1930, R. awalrj), and Naue (1890, R. esculenta, R. temporaria, and I?fscus). The origin of the iiter plates was the subjea of detaded discussion and controversy (Ekman 1913; Gegenbaur 18981901; Gerhardt 1932; Goette 1874; Greii 1905; J. G. Kerr 1905; Mangold 1936; E. Marcus 1930; H. Marcus 1908; Maurer 1888; Naue 1890). Viertel (1991, Bombina variegata, Bufa bu., B. &mita, R. tempurariu, andX. h s ) questioned the homology of the fdter apparatus with the pharynx, the evolution of anuran larvae and the fnaion of the iiter apparatus. Morphogenesis and histogenesis of the filter apparatus is based on studies of Bombina variegata, Buf mhmita, and Rana temporaria. The anatomy of the anuran pharynx in Gosner 17-18 is comparable to that of other vertebrates, especiaiiy those of embryonic fish, salamanders, and caeciiians. Viscerai pouches 1-5 and the mouth are not perforated before Gosner 20 (fig. 5.144). The anlagen of the transient

ANATOMY
Kg 5.18. Filter apparatus of I z t e m p s & (Gosner stage 32). bm -2) branchial arch IT,lateral view; note close position of fitter Left rvws and ciliary cushion and connection of branchiat food trap and vumal(B) C velum. * cushion touching the fitter rows. (C) Set-

117

'I

of filter rows. Sm-e a filter row. Surface d o n including ciliary and secretory cells. (F) Ciliary cushion qxned, connective tissue supporting the ciliary cells. (G) Ventral and velar papilla; position of branchid food trap, filter pLte, & \ - e n d velum in posterior view; secretory ridges of brachial f w d z n p s changing to typical secretory pits of ventral velum. h s e t g: Sec600of proximal ventral velum, cadaginous clasp supporn the thin squamous epithelium and a large vacuolized secretory cell. (H) Dorsal of ventral velum and velar papilla with a pattern of secretory pits. 1' Secretory pits of the velar edge, (J) Superficialsecretory p i s of the mnn-d ofthe ventral velum swmunded by polygonal squamous side cprthelial cells. (K) Surface anatomy of branchid f w d trap. Inset k : k f a c e of branchial food trap showing cells on a secretory ridge. kale = micrometersexcept millimeters in (A).See table 5.2 for abbrevia&. From Viertel(1985); reprinted by permission of Springer-Verlag.

cpidermal gills are visible as lateral protrusions. The buccal roof arena and buccal floor arena are still flat in Gosner 2021, and visceral pouches 2 4 perforate from stages 20-21 u@. 5.14B and 5.224 B). Visceral pouches 1and 5 do not perforate, and visceral pouch 1 later becomes the tuba eum hi The transient gills reach their maximum size at c i. Gosner 21 (fig. 5.14D) and regress during Gosner 24-25. The dstinction between endoderm (high density of yolk and lipid vacuoles) and ectoderm in the pharynx up to Gosner 23 (Viertel 1991) is corroborated by ontogenetic differentiation of the pharyngeal cells (i.e, increasing mitochondrial density) beginning at stage 23 (fig. 5.226, D). The cell structure, cytogenesis, and site of the visceral pouches show that the buccal floor, buccal roof, dorsal and ventral vela, anlagen of filter plates, transitional zone, and branchial food traps are of endodermal origin (Viertel 1991). The ventral velum at Gosner 22-24 starts as an archlike, transverse roll in the buccal floor anterior to visceral pouch 1 (figs. 5.14 and 5.22). In Gosner 24 the ventral velum covers the filter plates and gdl slits which lie between them. The surface of the secretory pits, into which the fully differentiated SC1 open, is visible from this time onward. The region between the ventral velum and the branchial food traps is termed the transition zone (figs. 5.14B, C, and 5.22F). In principle, this zone, the branchial food traps, and the ventral velum do not differ histologically ind cytologically. Ultrastructurally and histologically the anlagen of the filter plates are not different from the buccal floor (fig. 5.20). These anlagen mark the posterior limits of the pharynx and take up a progressively posterolateral position. The endodermally derived ciliary cushions and ciliary grooves clearly are dorsolateral projections of the esophagus and not ofpharyngeal origin (figs. 5.11A and 5.14D). Their surface anatomy and ultrastructure are the same as those of the esophagus. Ciliary and goblet cells in the epithelia of these organs do not occur in any other part of the buccopharynx. These cells, in contrast to the cells of the buccopharynx, are fully differentiated by Gosner 23.

Endoderm is represented largely in the proximal area of the filter plates and anterior to the branchial food traps. Apparently the endodermal anlagen of the filter plates are overlapped by a layer of epidermal cells proceeding from the anlagen persistent epidermal gills (fig.5.22F). The single layer of cells and the absence of ciliary cells and Type A and B cells indicate that this is a sensory layer (fig. 5.22D). This zone, where the epidermal cells meet the end~derm be(i.e., tween the persistent epidermal gills and the filter plates), is termed the ectodermal-endodermal transition zone (fig. 5.22E). The anlagen of the filter plates become progressively flatter and expand in a posterolateral direction. At Gosner 23, the anlagen of the middle folds of the filter rows are visible as pro~rusionson the surface of the filter plates. These increase in height from Gosner 24-25 and branch into the anlagen of the primary side folds. Secondary side folds are not differentiated by Gosner 25. After Gosner 23 some endodermal cells of the anlagen of the filter plates become oval or cube-shaped. The rate of growth of these cells does not increase until Gosner 25 when they become exposed as protrusions between the cells of the squamous epithelium. These are the anlagen of apical cells and continue to grow afier Gosner 25. Before Nieuwkooppaber 35-36, the pharyngeal development ofXempz4s taevis (fig. 5.144, B) is equivalent to the initial state in larval Types 3 and 4. From Nieuwkooppaber 38 the adage of the ventral velum arches dorsally between the medial line and the visceral pouches. The anlagen of the filter plates, anterior visceral pouches and the gill slits change their position longitudinally and less so transversely. The extension of the pharyngeal longitudinal axis does not occur until Nieuwkooppaber 47. From Nieuwkooppaber 4 2 4 3 the adage of the ventral velum has a serial arrangement of cells in three tiers, and secretory cells break through between the distal squamous epithelial cells. A full differentiation of the surface of the ventral velum takes place at Nieuwkoop/ Faber 46. Secretory grooves into which SCl flow are framed by secretory ridges and random squamous epithelial cells. The probable function of microtubules in the apical zone of secretory cells, which appear at Nieuwkooppaber 47, is to guide and transport vacuoles during extrusion (Berthold 1980). Mucus synthesis in the ventral velum is evident from Nieuwkooppaber 45, although no mucus is found on the anlagen of the filter rows in this stage. The epidermis of the ectodermal-endodermal transition zone ofXempzts at Nieuwkoop/Faber 38 begins to penetrate into and overlay the visceral pouches (fig. 5.23). Judging by the distribution of yolk and the absence of chary cells, the sensory layer overlays the anlagen of the filter plates. In contrast to the case of Bmbina ~ariegata,Bzlfo cakk.mita, and ma tmpmaria, the sensory layer in Xenopzls resembles n squamous epithelium as does the epidermal mantle of the gills that are present ventrolaterally only during stages 3847. From Nieuwkooppaber 42 the surface of the filter plates is enlarged by the dorsal arching of their anlagen and, from stage 47, by the appearance of the secondary side folds. The anlagen of the apical cells are formed by the endodermal

ANATOMY Fig. 5.19. Fiter apparanis of Xnops k (Nieuwkoop/Faber stage 54). (A) Fiiter apparanis and buccal floor arena with views of ventral velum, filter plates, and dorsal vela. Symbol: stipple = ciliary groove. (B) Right side, anterior view of cross section at the plane of fiiter plate III and branchial cavities I1 and I11 with epidetmis reflected. (C) Section of ventral velum showing different sizes of apicai ceii poles caused by different exmision phases of mucus. (D) Ciiiary groove and dorsal velum. Areas: Z1 = rim of squamous epithelium separating cihary groove and lateral filter plate, 2 2 = ciliary groove (also shown in E), 23 = tramitional zone of ciliary groove and dorsal velum (also shown in F), and 24 = dorsal velum with smooth epithelial cells (also shown i G). (H) Epithelium at base of filter plate III;see asterisk in (B). n (I-J) Morphology of fiiter row. (K) Fiiter rows of the ventral region with pustuiations at the base of a filter row. Scale = micrometers. See table 5.2 for other abbreviations. From Vieael (1987); reprinted by permission of Gustav Fischer Verlag.

119

first occurs and all larvai organs are present at least as an"

Ingestion of food by tadpoles of Xenupus laepis begins at Nieuwkoop/Faber 42, and by stage 46 ali ssructures required for active food intake are present except the iiter plates are not yet fuliy Werentiated (Viertel1991). Food particles are present in the esophagus of embryos by this stage, and chondrogenesis of the palatoquadrate and ceratohyals has been completed by this stage. Accordmg to Wassersug and Hoff (1979), these visceral elements, together with the connecting musculus orbitohyoideus, are responsible for the ventilation and irrigation of the iiter apparatus. The synthesis of mucus on the ventral velum and a t e r rows does.not begin until Nieuwkoop/Faber 45. At Nieuwkoop/Faber 46 the yolk reserves from the buccopharynx and the esophagus are exhausted, and those in the digestive tract are being deaniagen of the iiter plates, break through the squamous epi- pleted; the peribranchal chamber is dosed. In other species, thelial ceiis of the sensory layer in Nieuwkoop/Faber 47, and ingestion begins before the yolk reserves have been used up come to lie between the squamous epithelial ceiis. Both celi and before the peribranchial chamber closes. Beginning sustypes are supported by endodermal ceiis. pension feediG before the filter plates are enlarged anTd the In Bufo cafamita, ingestion begins prior to Merentiation keratinized mouthparts are developed in Type 4 tadpoles of the iiter plates and before Type 1 secretory ceiis of the is an example of functional heterochrony (Viertel 1991, iiter plates and ventral velum have become fuliy active (Vier- 1992. 1996). te1 1991). The steep rise in ingestion rate apparently results The origin and evolution of the buccopharynx are now from the presence of goblet and ciliary ceiis that have differ- considered. Except for the ciliary cushions and groove, ciiientiated by Gosner 23. The anlagen of the iiter plates are ary and goblet ceiis occur exclusively in the esophagus and brought continualiy closer to the chary cushions by ventiia- larval stomach (Ueck 1967; figs. 5.24 and 5.25A, B). The tion movements of the buccopharynx in Gosner 23-24. This presence and location of SC1 in the ventral velum, transition results in mucous contact between the goblet celis and ania- zone, branchial food trap, glandular zone, and dorsal velurn gen of the iiter rows. Threads of mucous that pass fiom dor- suggest that the branchial food traps and transition zone sal to ventral in the lumen of the pharynx form a mucous are ventral Merentiations of the ventral velum. The dorsal entrapment system. Because a mechanism of transport of arching and arrangement in rows of the SC1 changes the mucus-covered food particles is present in the ciliary ceiis branchial food traps of the pitted type (Type 3) into of the ciliary cushions, trapped food can be moved to the branchial food traps of the ridged type (Type 4). The iiter esophagus. The ciiiary cushions are the main producers of plates and gis are epidermal-pharyngeal organs. The phamucus for the functioning iiter apparatus. Although the ii- ryngeal endoderm is the largest component of the mass of ter plates also produce m u a , their main function is to pro- the filter dates. and e~idermal ectodem and mesoderm of vide an extensive surface area and support for the distribu- the capillaries predominate in the gis. tion and deposition of mucus from the ciliary cushons. The ceii structure of the ciliary groove ofxenupus laevij is M u a produced by one organ is deposited on the surface the same as that of the esophagus. Fox et al. (1972) and of another structure. Kindred (1927) described ciliary ceiis in the esophagus that The larvae of Bufo caiamita and Rana tempmaria become are identical to those of the ciliary groove from Nieuwkoopl vagile at Gosner 22, and after Gosner 23 the embryos ac- Faber 44. Fox et al. (1970,1972) described SC3 with apical tively search for and ingest food. This occurs prior to differ- vesicles in the eso~hams. I " Goblet teus do not differentiate in entiation of the keratinized jaw sheaths and labial tooth the esophagus untii the beginning of metamorphosis (Nieuwrows, which precludes the scraping of attached foods. These koop/Faber 65; Fox et al. 1972) when the ciliary groove has young individuals depend upon atration of smali, plank- already regressed. Fox et al. (1972) also corroborated the tonic, or sedimentary particles. Because the transition to a presence of rnicroviiious stubs on the ciliary ceiis of the swimrning stage in B. bufi embryos does not occur untii esophagus. Ciliary celis and SC3 are found only in the chary Gosner 23, it is not surprising that ingestion activity begins groove and esophagus. Their identical celiular composition at stage 24 rather than 23. In B. bufo, B. calamita, R. tempora- and L ~ositions relative to each other show that the charv ria, and other anuran species with free-swirnming larvae, groove is an anterolateral projection of the esophagus. there probably is strong selection to start feeding as soon as The aliocation of the ventral velum of Xenupus to the possible before the yolk material has been depleted. Early pharynx or buccal cavity is more difficult than with larval feeding aliows tadpoles to proceed as early as possible to in- ~ y p e3 and 4. Judging by its position relative to the visceral i dependent feeding and complement the dimuiishing yolk pouches (Nieuwkoop/Faber 38), the ventral velum is part of supply with a supplementary source of food. Perhaps the be- the buccal roof as in larval Types 3 and 4. Gradwell(1971a, p i n g of the larval phase and the end of embryonic devel- b, 1975c) thought that the ventral velum of larval Types 3 opment should be defined as that point at which ingestion and 4 originated similar to that ofxenopzts by proiiferation
I

ANATOMY
Fig. 5.20. Cytology of the filter apparatus of Rana temporaria (Gosner stage 32). (A) Cross section of ventral velum anterior to the velar edge. (B) Cross section of the edge of ventral velum. (C) Longitudinal section of the edge of ventral velum. (D) Longitudinal section of the ventral side anterior to the edge of ventral velum. (E) Section of secretory celi. (F) Cross section of branchial food trap; secretory ceils extendiig to the whole branchial food trap at SC1; squamous epithelial celis covering the secretory celis at SQC. (G) Cross section of branchial food trap. (H) Transverse section of branchial food trap of Eu$ bu$; secretory cells exposed plauily. Scale = micrometers. See table 5.2 for abbreviations. From Viertel (1985); reprinted by permission of Springer-Verlag.

121

of the area between the halves of the velum in the oral cavity. Neither surface anatomy nor celi repertoire provides evidente of the origin of the ventral velum inX. laevis. The transition of endodermal celis to secretory ceiis SC1 is conspicuously similar to that in the ventral velum, transition zone, and branchial food traps of larval Types 3 and 4. The arrangement of SC1 in rows also is strikingly similar to the pattern in the ridged branchial food traps of Type 4 lamae. It is therefore probable that the branches of the ventrai velum ofX. W s are homologous with the ventral velum and the branchial food traps of larvae of Types 3 and 4. The previous description of the filter plates in larval Types 3 and 4 also applies to X. laevis. The term "pharynx" warrants hrther discussion. The site of the ceratobranchials shows that the anterior pharynx bes at the leve1 of the buccai floor arena. The ectoderm that vaginates toward the aniagen of the iiter plates does not belong to the pharynx any more than does the esophageal tissue that penetrates the pharyngeal roof to form the ciliaty cushions and ciliary groove. These structures penetrate the embryonic p h a q m and contribute to the formation of the filter apparatus. Relative to the origins and sites of the visceral skeleton, branchial nemes, and visceral musculature, the original embryonic pharynx does not d i e r from that of other vertebrate embryos (i.e., plesiomorphic vertebrate pharynx). It is a supporting structure for the invaginating epidermis and the esophageal endoderm; an eaodermalesophageal iiter apparatus has appeared. If the anterior-posterior extension of the buccopharynx in X. laevis is a plesiomorphic charaaer, then the buccopharynx may have evolved according to the foiiowing scenario. A fold morphologically and ultrastructurally similar to the ventral velum of lamai Type 3, but whose location is similar to that of the Type 1 lama ofX. b i s , lirnited the posterior extension of the buccai floor of the primitive anuran larva. This primitive morphology gave rise to two lineages. In one, the anatomical position of the buccopharynx was maintained, while the types of ceii present in the ventral velum acquired their own pattern of differentiation and arrangement. This is the coniguration found in xenoanuran larvae (Type 1; Starrett 1973). In a second lineage, the longitudinal axis of the mouth-buccal floor-glottis was shortened, especially posteriorly, while the transverse axis of the buccal floor was extended. Fewer celi types diierentiated in the buccopharynx than in the xenoanuran larval type. This is the

configuration in lemmanuran (Type 3 of Orton 1953) and acosmanuran (Type 4) larvae of Starrett (1973). These two lineages probably developed independently w i t h different phylogenetic iines. Obviously the diierentiation of a filter apparatus by anuran larvae is unique within the chordates and involved a specialized modification and association of the pharynx or Kiemendann (i.e., giils and gut). The common ancestor of all anuran larvae probably was a filter feeder with the foiiowing a ~ t a ~ o m o ~ ~-char- s hou aaers: an ectodermal-endodermal transition zone (filter plates) and an esophagus extending anterolaterdy into the pharynx (ciliary groove and/or ciliary cushions). The position of the ventral velum in the tadpole ofX. laevis is primitive and plesiomorphic w i t h anuran lamae. Gradwell (1971a, b, 1975c) argued that the ventral velum in lamal Types 3 and 4 was foymed by proiiferation of the area between the Xenopus-like ventral velum. This ancestral velum probably was charaaerized by the simple combination of squamous epithelial celis and SC1. If so, the compiicated arrangement of SC1 and squamous epithelial ceiis in X. lamzs suggests that the ventral velum was difTerentiated apomorphically. Accordingly, the ventrai velum and branchial food traps of larvai Types 3 and 4 are apomorphous with respea to their locations, and the branchial food traps of larval Type 4 are apomorphous relative to the more highly differentiated arrangement of SC1 (branchial food traps of the pitted type). The epidermal covering of the aniagen of the filter plates (i.e., the eaodermal-endodermal transition zone of the viscerai pockets; figs. 5.24 and 5.25) is comrnon to Bombina variegata, B@ buf, B. calamita, Rana temporaria, Xenupus laevis, and probably all generalized larvae. Developmental association of two germ layers fits weii into our knowledge of the vertebrate head, which is composed of ectodermal (neural crest) and endodermal elements, and especially the development of the vertebrate pharynx (splanchnocranium conneaed with the endodermal wall of the anterior p t ) . Which characters of the anuran filter apparats were derived within Lissamphibia (i.e., autapomorphic), and which charaaers are common to Lissamphibia and outgroups? Statements (B. I. Balinsky 1981; Hibiya 1982) that epidermis overlays mesodermal giil tissue in bony fish allow comparison to the anuran lamal giils (B. I. Balinsky 1981; Fox 1986; Greven 1980; Lewinson et al. 1987). Because of certain synapomorphies (loss of sacculus vasculosus, presence of neurocranial endolymphatic sacs), the Dipnoi is considered to be the sister group of amphibians (Northcutt 1986; Rosen et ai. 1981). Fox (1965) and Rauthner (1967) mentioned a deep penetration of giii epidermis that overlies the endoderm in the first visceral pouch of Dipnoi. The ectodermai-endodermai transitional zone of anuran larvae could be interpreted as the result of the continuing development of t h s synapomorphous charaaer. If true, then the ancestor of anuran lamae had already developed a similar ectodermalendodermal transitional zone of unknown diierentiation. Because of their uniformity (Lannoo et al. 1987; Wassersug 1980; Wassersug and Heyer 1988; Wassersug and 1)yburn 1987), the surface structure and histology of ilter

122

BRUNO VIERTEL AND SUSANNE RICHTER

Fig. 5.21. Cytology of the filter apparatus of Rana tempmrh (Gosner smge 32). (A) Different conditions of merocrime exmision in the ciiiary cushion. Inset a: Cilia on the ciiiary cushion. (B) Cross section of ciiiary cushion. (C) Cross section of primary side fold of filter row. Scale = micrometers. See mble 5.2 for abbreviations. Frorn Viertel (1985); reprinted by permission of Springer-Verlag.

ANATOMY

123

plates do not lend thernselves to cladistic analyses within anurans (0. M. Sokol 1975). Although quite Merent among anurans, the supporting cartilaginous system also lacks characters usefd in a phylogenetic analysis. The esophageal derivation of ciliary grooves in X. laevts and of ciliary grooves and cushions in the other species is not found in Dipnoi, Urodela, or Gymnophiona and seerns to be unique to anuran larvae. Ciiiary grooves and ciiiary cushions therefore are best considered as ~utapornomhic characters of anuran larvae, and their evolutionary origins should be considered carefuliy. Does the different structure of the ciiiary groove in Xenopus cornpared with the ciiiary grooves and ciliary cushions of Bombina variegata, Bufo calamita, B. bufo, and R. tempmatia allow phylogenetic conclusions? Goblet celis are the typical secretory celi of the ciliary grooves and ciliary cushions in these four species. InXequs, goblet celis do not occur in the ciliary groove (or esophagus), and Type 3 and 4 secreto9 celis are present (Viertel 1987). There are two alternative explanations. (1) The ciliary grooves of all the species studied were derived frorn a common anuran ancestral larva ("earlier stock" of Tihen 1965). This common character was rnodified structurally in Merknt ways in Xenupus and other species, and the ciiiary cushions were derived from the ciliary groove. (2) The ciliary grooves in Xenupus and other species (families) originated independently fi-orn the esophagus via convergence or parallelisrn. In this case the ciliary cushions would be derived frorn the esophagus. The first alternative is supported by several other combrnon anuran larval characters, such as the positions of the operculum (fold of peribranchial chamber) and spiracle(s) and the simpiification of the rnouthparts rnodified in different ways in Xenupus and tadpoles of larval Types 3 and 4 (Inger 1967; 0. M. Sokol 1975; Then 1965). As for the ciiiary grooves, these organs probably were inherited frorn a common ancestor and rnodi6ed differentiv in the various larval types. As expected, corroboration of this alternative is found in the dendrograms of frog farnilies (Inger 1967; Trueb 1973) and the cladistic analysis of Dueliman and Trueb (1986); a ciliary groove was f&d in a cornrnon ancestor for all anurans and subsequentiy modified in Xenupus and other lineages. The alternative is corroborated by the first hypothetical phylogeny of frogs of Inger (1967) in which Pipidae and Rhinophrynidae (Type 1) on the one hand and all the other families (and larval types) on the other are derived independentiy from a common ancestor. In other dendrogr&s by 1nier (1967), ciliary grooves would have been derived independently three times, and in the other hypothesized phylogenetic relationship (Dueliman and Trueb 1986; Trueb 1973), four separate origins would be expected for the species studied here. If it is presumed that all anuran families developed ciiiary grooves, then seven independent origins would be required. This high number of independent derivations suggests that the first alternative is much more probable than the second (see Ax 1984). The ultrastructure and ontogenesis of the filter apparatus of Type 2 larvae (Microhylidae) also is in agreement with this discussion. Mucous cords are described by R. M. Savage
L ,

(1952) and pressure cushions by Wassersug (1980); rnucous cor& are alwavs associated with ciliarv nrooves. Whether the pressure cushions (Wassersua 1980) are uliated or not is unknown. The surface anatorny of &e filter plates is, in principle, identical with that of all other species (Wassersug 1980; Wassersug and Pyburn 1987). Based on a comparison of Scaphzuphryne larva with the larvae of the Ranoidea (Type 4) and Microhyiidae (Type 2), Wassersug (1984) concluded that the Type 2 larva was derived fi-orn Type 4 through developrnental truncation. Wassersup: viewed Ranoidea and Mi" crohylidae as closely related taxa. If true, then the scenario above is also true for Type 2 larvae (Viertel1991). The functional significance of Aise evolutionary features is the ability of anuran larvae to exploit a broad range of suspended and sedirnented particles and, in larval Types 3 and 4, attached organisrns. Most authors state that suspension feedmg in tadpoles is a primitive character common to all anuran larvae (Inger 1967; Slade and Wassersug 1975; 0. M. Sokol 1975). This feeding strategy rnust have had a tremendous iduence on anuran evolution (Viertel 1991,1992,1996).
i U

Dhestive Tat and Diverticulu rc The foregut begins at the esophagus and includes the ciliary cushions and ciliary grooves that extend into the buccopharynx. The esophagus starts in the sagittal plane and then loops to the right where the manicotto (= manicotto glandulare, larval strnach) begins in the anterior bend of &e second loop (I. Grfiths 1961, Rana tidibunda; figs. 5.12 and 5.26A, C, D). Mucus-producing goblet ceUs and ciliary c e h are common in the forigut, andthe ciliary celis are arianged in a species-spechcpattern. In most species (e.g., I. Grfiths 1961, Rana tidibunda, Taylor/Koliros 1 1 Rovira et al. 1; 1993). ciiia are still present in the left lateral loop of the ,, esophagus. After a cilia-free section, cilia are found in most species anterior to the manicotto in two bands that extend at an angle of 90" to those of the ciliated esophageal loop. The lumen of the rnanicotto is ciliated. Posterior to the rnanicotto, two ciliated bands from the midgut (Ueck 1967) extend laterally. Because the rnanicotto lacks any muscular layer that enable; peristalsis, the ciliary tracts 1lkefy promote food movernent at points where accurnulation might arise (see Naitoh et al. 1990 for mforrnation on rnotility of the larval intestine). The anatomical proportions of the foregut and the site of the liver and pancreas differ among genera (figs. 5.12 and 5.13). Philautusgryllus, in particular, deviates considerably from the general scherne. The esophagus of suspension feedmg anuran larvae is short with thin walls and longitudinal folds (fig. 5.261)). In rnost species it has a combination of ciliary and goblet ceUs (Barrington 1946; 1.Griffiths 1961; Viertel1992, Bufo bufo, B. calumita, and Rana temporaria);C. A. Brown et al. (1992) discussed vagal stimulation of the alimentary tract. Ciliary celis in Xenupus are developed by Nieuwkoop/Faber 44 (Fox et al. 1972; Viertel1992; see Buccopharynx).Viertel(1987, 1992) did not tind esophageal goblet celis in the ciliary groove of X. l m 5 , an anterior extension of the esophagus (Fox et ai. 1972: Ueck 1967). Viertel(1987) describes bulbshaped celis rich in ER and containing numerous hose-

, 3.22. Ontogeny of the iiter apparatus ofBufocahmita. (A) Left F

Pih- ot'buccal floor arena, anlagen of iiter plates, and transient epiderd (Gosner stage 21). (B) Lateral view of anlage of filter plates, of rentrai velum, and transient epidermai giiis (Gosner early 22 . C ) Cross section of undifferentiated endodermal ceiis of the phaorgans (Gosner 21). (D) Cross section of the ectoderm covciq die nansient epidermal gis (Gosner 22) with mitochondrialah Tlpe B ceiis. (E) Site of anlage of iiter plates, transient epidermai - aniage of persistent epidermai gills, and anlage of ventral velum & h e r lare 22). (F) Cross section of proximai base of transient epikmd nith underlying endodermai celis (Gosner 22). Scaie = znaometers. See table 5.2 for abbreviations. From Viertel (1991); -qrinced b!. permission of Springer-Verlag.

ANATOMY

125

shaped vacuoles in the ciiiary groove; these are probably the wsicular cells of Fox (1981) and Fox et ai. (1970, 1972). The esophagus of the carnivorous larva of Hymenochirus k r g r n . is long and forms diverticula. Strong fibrous cords ot-Gmective tissue develop posteriorly. ~ o b l ecells are the t hmuiant ceii type in the anterior epithelium, whereas ciliary d l s are found posteriorly almost to the exclusion of all other eii n-pes. ~ h esophag& evidentiy is elastic and secretes iaqe quantities of mucus, which permit large prey to be sn-alion-edwhole (Ueck 1967). I. Grfiths (1961) described a abrupt histologicai transition from esobhagus to mann h o in species in the Bufonidae, Discoglossidae, Hylidae, IxptOdactylidae, Pelobatidae and Pipidae; members of the Ranidae and Rhacophoridae have a gradual histologicai rransirion. The clairn that the manicotto glandulare (Lambertini 1929; figs. 5.11A1 5 . 2 6 4 and 5.27) is merely a storagem a c h (J. M. Dodd 1950) that lacks digestive fnctions or a precursor of the frog stomach (Siissbier 1936) has been Asproved (Ueck 1967).Branched tubular glands lie beneath a diated epithelium, and large quantities of tubules thicken d x n - d . The tubular glands empty into the manicotto lumen through an irregular system of gaps rather than duas @. 5.26E). Blood capiaries surround and penetrate an ourer generative layer that forms the distai limit of the tububr @and tissue. A t h n layer of muscles lies on top of the qillanes and generative layer (I. Griffiths 1961, Rana ridih&).There is a single layer of circular muscle cells and a * e layer of longitudinal muscle cells in Xenqus lueuis L-eck 1967), but the muscularis mucosae and submucosa ue not present. Othenvise, the anatomy ofX. h s is similar m that of R. &bunda. -inatomical and histologicai comparisons of the rnanKocco reveal no fundamental differences among members of nine farnilies, aithough a few species do not fit the general scheme (I. Griffiths 1961). Manicotto glands in Alytes obamiazns are found only dorsolaterally; a muscular foregut nphlosolis (tonguelike evagination of the gut wall) with ciiian-c e h and lined with ciiiary tracks is situated ventrdy. G b l e t cells are highly concentrated in the manicotto epitheh Bumbina variegata lacks manicotto glands (Wiiczynska . 1981). In Philautusgryllus (fig. 5.27A-C; I . Griffiths 1961), &e anatomicai deviation is even greater, and the anatomy of microhylid larvae is more complex than in other forrns. Ribosomes. vesicles.,and ~ a r d earravs of smooth ER ocl cur in actively secretory manicotto gland cells. The high denb

sity of ER forces the mitochondria to the base and periphery of the ceii. Enzymatic granules usually found in small numbers in the middle of the ceii suggest that mucus production is not the exclusive function of gland cells. Gland development is dependent upon the nutritionai state of the tadpoles. Nutritional deprivation reduces the totai diameter of the tubules. The gland ceiis are cuboidai or slightly cylindrical with small microvdh. By contrast, gland cells of weii-fed tadpoles are flattened to cuboidai with welldeveloped microvilli. The number of lipid vacuoles (neutrai fats) increases with the intake of fat and decreases with the absence of dietary fat. They disappear in starved animais after 36 days. The cytoplasm of starving cells becomes denser whde basophdic granules increase in size and number. This suggests that lipid vacuoles accumulate nutritive substances. Eating starch increases the size and number of lipid vacuoles more than eating dried pancreas tissue and yolk. The cytology of starved tadpoles that start feeding again soon approximates that of weil-fed tadpoles. This indicates that the gland cells of the manicotto glandulare are functionally linked to the digestive process and that the basophilic granules are involved in protein digestion. Cytological comparison with the gastric parietai or oxynic cell of frogs suggests that the larvai gland cell is an ion transporting cell. This cell produces and secretes electron-dense granules that can be distingushed from the pepsin granules of the Magenhauptzelle (= stomach main cells) of adult Xenopus. The pH of the manicotto is below 3, and lipase activity was described by I. Griffiths (1961, Rana temporaria and Xenopus h s ) . Hydrochloric acid kills microorganisms in the food, denatures protein to enable enzymatic breakdown, and dissolves saits from organic substances. The transformation of hemicellulose into sucrose in the cell walls of algae is especially important to suspension-feeding larvae. Because food passes through the manicotto quickly, the hydrochloric acid does not concentrate in the manicotto. Assuming that aggressive pepsin is not present, these protection mechanisms are sufficient. Inokuch et ai. (1991) found a pepsinhke enzyrne in larval R. catesbeiana that may be cathepsin E, and Carroll et ai. (1991) discussed pepsin in anuran larvae. I. Griffiths (1961) found evidence of proteolytic activity in early prometamorphic Xertopus lmis and Rana temporaria and tryptic activity in R. ridibunda up to Taylor/Kollros XX. Pepsin is present in the stomach of metamorphc Xertopus laevis at Nieuwkoop/Faber 63 (Fox 1984), beginning with Taylor/Kollros XXTI in R. &bunda (I. Griffiths 1961), and aiso in metamorphic R.p$iens (Kuntz 1924). Although pepsin or cathepsin may be present, many factors argue against a resorptive function of the manicotto (Ueck 1967), including the lack of direct contact between the food and the gland cells, the nonresorption of tryptophan blue, and the lack of alkaiine phosphatase. Cytologicai comparisons of the gland cells with the resorbing cells of the intestine shows that the latter are more densely concentrated than gland cells, contain Golgi bodes, and metabolize low molecular weight substances. Based on the systematic diversity of species studied, it is assumed that the manicotto glandulare is plesiomorphic for tadpoles and probably was present in their common ances-

BRUNO VIERTEL AND SUSANNE RICHTER

ETZ

Fig. 5.23. Ontogeny of the iiter apparatus ofXenopus iuevis. (A) Cross section at the plane of the visceral pockets and the anlagen of transient epidermal @s (NieuwkoopJFaberstage 38). (B) Magnified cross section from area in (A) of visceral pocket showing the sensory layer and endoderm. (C) Frontal section of left side of pharynx showing sensory layer and endoderm (Nieuwkoop/Faber38). (D) Paramedial section of the head region, peribranchial chamber opened, anlagen of iiter plates sectioned (Nieuwkoop/Faber 42). Schernatics of TEM section of anlagen of filter row, yo& endoderm overlapped by the sensory layer of the epidermis: (E) an early hatchlig (Nieuwkoop/Faber 40141) and (F) an early tadpole (Nieuwkoop/Faber45). Abbreviations: AD = aorta dorsahs, BII-I11 = branchial arches I to III, C = connective tissue, Cn= = connective tissue ceii, T I ENC = endodermal ceii, E Z = ectodermal-endodermaitransition zone, EV = ear vesicle, FPC = epidermal fold of peribranchial chamber (operculum),PC = penbranchial chamber, SL = sensory layer, SLC = sensory A layer ceii, TEG = transient epidermal gis, V = visceral arch, VP = viscerai pocket, VPl = visceral pocket 1, and W = yok vacuoles. From Viertel (1991); reprinted by permission of Springer-Verlag.

ANATOMY

.'.'.
IIII
"o .: ",
OO

Sensory layer on the endodermal AFP Endoderm Neurai anlage Blood Muscles

/
,e d'

Periderm Sensory layer Periderm Sensory layer

Epiderm

Primordial cranium ... ., and visceral skeleton ..::.

::: . .. . . ..::.

SensOr~ layer penetrating endoderm

Endoderm Mesoderm

.,$&
I
L

[mmn

Chorda dorsalis

tor. Presurnably the manicotto of carnivorous tadpoles is an apomorphous modification of the manicotto glandulare of the omnivorous suspension feeder (I. G r a t h s 1961; Ueck 1967). The manicotto epithelium of the carnivorous larva of H ~ n o c h z m boettget4 lacks ciliary cells and has a thick mus cous layer containing acid phosphatase that protects the manicotto from ingested food. Branched tubular glands in the anterior part of the manicotto have cells similar to those in X. h m i s and R. temporaria, but glands are absent in the posterior region. The posterior section of the manicotto is larger than any other part of the foregut (Ueck 1967) and functions as a storage area; this is functionally similar to the foregut in other animals that swallow and store their food whole. Like Xenopus, H. boettgmi lacks both musculans mucosae and submucosae, and the longitudinal and circular muscles (Ueck 1967) are poorly developed. Other carnivo-

rous forms develop a thick muscular coat, although there are no ciliary cells and few crypts (I. Griffiths 1961, Lepzabbatrachus laevi'r, Occidozyga b i s , and 0.lima). Griffiths ( 1961) I. suggested that the manicotto is a grindiig chamber in these species. The branched tubular glands differentiate at an early stage with the development of deep folds in the manicotto epithelium. Short tubules appear and then branch (Kopsch 1952, Rana temporaria).The synthesis of hydrochloric acid and the extrusion of lysosomes disprove the conclusion of Barrington (1946) that the ciliary cells, goblet cells, and Schhimkopfchenzelhn (= mucous-headed cell) contribute to the formation of the gland cells. The lectin hstochemistry of the manicotto epithelium is fundamentally diierent from that of the tubular glands (Ishizuya-Oka and Shimozawa 1990). I. G r a t h s (1961, Alytes ~ b s t ~ c a n s Rana ridibzmda) and

128

BRUNO VIERTEL A N D SUSANNE R I C H T E R

posterior of buccal cavity

, _ _ _ _ _ _ _-_ - - - - - - - - - - - = - - _ _ _ _ _ _ _ _ _ _ _- branchia - _

endoderm of esophagus

DV

: BRA

BFA , ' , .-------_-__-____---

W-

endoderrn of buccopharynx cells rich in yolk and lipid endoderrn A vacuoles

rciliary cushion

: BFT+FP

.. .

V ~

--------------

--- --- --- - -------

-.r-----------

transition coat of the transient ', zone and persistent gills , ' ......................... epidermis

-------------chamber fold peribranchial

t /
+

ectoderm cells flat and rich ir1 lipid C U O ~ ~ S melanin granules, and some cells with cilia

Fig. 5.24. Origin and morphogenesis of the filter apparatus. The filter plates (ectodermal-endodermaltransition zone) and epidermal gis form one branchia. The parts of different origins (i.e., ciliary cushions, branchial food traps, and filter plates) are shown as a functional unit for mucus entrapment and ciliary transport. Symbols: arrows indicate

change in morphogenesis, the box endoses functional units for production and transpon of mucus, and dashed lines enclose unit of one organ. See table 5.2 for abbreviations. From Vierte1 (1991); reprinted by permission of Springer-Verlag.

/ CG \

EYEDVI

PCW

'

CG FP

1 w
PET
2 mrn

I FP II

\ HY

Fig. 5.25. Dorsolaterai aspects of the buccopharynx of tadpoles. Right, dorsai halves of heads dissected and lifted. (A) Rana tempouariu ( e (Orton's larval Type 4, Gosner stage 36). (B) X ~ u s u e v i s T j ~ 1, Nieuwkoop/Faber stage 54). Symbols: crosshatch = ectodermal-

endodermal transition zone, stipple = endoderm, and diagonal lines = epithelium of esophagus (ciliary cushions, ciliary grooves). See table 5.1 for abbreviations. From Viertel(1991); reprinted by permission of Springer-Verlag.

ANATOMY

129

suggested that the glandular part of the manicotto originates hom elements of the ventral pancreatic anlage. Pancreatic tissue is found in the submucosa of the Dipnoi and Agnatha (Barrington 1942, 1957; J. G. Kerr 1919). Ceiis identical to those of pancreas ceh in the manicotto apparently are not of pancreatic origin (Fox et ai. 1972; Ueck 1967). The lytic regression of the gland ceiis and the neogenesis of a stomach

during metamorphic climax show that the manicotto glandulare is a typicai larva1 organ (see Carver and Frieden 1977 and Janes 1934 for further discussion of metamorphic regression of the gut). Anuran larvae lack a pylorus, so the rnidgut begins irnmediately posterior to the manicotto and ends at the posterior opening of the hepatopancreatic duct (fig. 5.264 C). This

Fig. 5.26. Form and function of the foregut in Rana dibunda. (A) Ventral view. (B) Foregut reconstruction; left, anterior half of gut tube removed (Shumway stage 22). (C) Ventral view of foregut (Shumway 25). (D) C i l i q patterns of the larvai foregut: (a) Ventral view of foregut opened along ventral midline and edges reflected lateraiiy, c d q tracts in black, (b-e) diagrammatic cross sections at levels 1 4 in (a), and (f) lateral view of foregut showing topographic relationship of the planes w, x, y, and z in (a). (E) Manicotto of a larva (Taylor/Koiiros stage V). Numerals 1-6 = successive stages in the development of manicotto glands. Abbreviations: B = manicotto-pancreatic bridge, C = manicotto crypt, CL = ciliated border of crypt opening, D = distal opening of hepatopancreatic duct, DP = dorsal pancreas, EP = esophageal plug, ES = esophagus, F = foregut, G = gaiibladder, GC = gland cord, GL = gland lumen, GR = glandular rupturepoint in the epithelium, HP = hepatopancreatic duct, I = intestine, L = gut lumen, LB = left lung bud, LV = liver, M = manicotto, MC = manicotto cells, OC = opening into manicotto crypt, OG = outer generative layer, P = pancreas, PH = pharynx, R = recurrent hepatopancreatic ducmles, S = secretory granules, SM = serosa and attenuated muscle layer, T = transverse septum, and W = ventral pancreas. Orientation marks as in fig. 5. 1.After I. Grif5th.s (1961); reprinted by pemssion of The Zoological Society of London.

BRUNO VIERTEL AND SUSANNE RICHTER

Fig. 5.27. Form and function of the foregut in Phzlautusgryllus. (A) Ventral view of foregut (Taylor/Koiiros stage V). (B) Lateral view of foregut, half of manicotto capsuie and parts of foregut wall removed (Taylor/KoIlros V). Region marked as xy is that shown in (C) a and d. (C) Ciliation panern of larva1 foregut: (a) Dorsal view of foregut opened along the dorsal midline and edges reflected laterally; ciliary tracts in black and manicotto tissue stippled, (b) diagrammatic cross section at "i" in (a), (c) diagrammatic cross section at "i?' in (a), and (d) lateral view showing topographic relationships of points indicated as v, w, x, y, and z in (a). Abbreviations: CC = lateral chated canals,

F = foregut, FC = foregut cavity, FG = ciliated collecting grooves, FP = foregut plug, FT = foregut tongue, H = hepatopancreatic duct, I = intestine, I 0 = lateral opening of foregut tongue, K = muscular capsule of the manicotto, L = gut lumen, M = manicotto, MC = manicono cells, MCH = manicotto chamber, MD = midgut diverticuium, MF = manicotto flap, MT = muscle of foregut tongue, P = pancreas, R = recurrent hepatopancreatic ductules, RV = reservoir, T = transverse sepmi, and TM = transverse muscles. After I. Grfiths (1961); reprinted by permission of The Zoologicd Society of London.

part of the second gut loop descends posterior to the manicotto and rises again anteriorly. The duodenum begins at the orifice of the hepatopancreatic d u a and extends dorsay to form a loop that descends on the right and rises on the left. N. E. Kemp (1949, 1951) discussed how body shape influentes the coiling pattern of the intestine (see Ichizuya-

Oka and Shimozawa 1990 for development of gut connective tissue). Two spirais are formed, and a fibrous or muscular rnfolding of the intestinal wa (i.e., typhlosolis; Ueck 1967; see Rovira et ai. 1993; Sesama et ai. 1995) enlarges the anterior secretory and posterior resorptive surface of the intestine, controls clearance time of food through the gut,

ANATOMY

131

causes a narrowing of the intestinal passage, and slows rnixing of the pancreatic enzymes with the food mass. The orifice of the hepatopancreatic duct lies next to the anterior end of the typhlosolis. The lack of secretory digestive granules in the duodenal ceils suggests that this part of the intestine does not produce its own digestive enzymes but resorbs substances metabolized by pancreatic enzymes (Fox et al. 1970). Typical duodenal celis are columnar, mitochondria-rich, microvdious, ciliated celis (Fox 1981; Fox et al. 1970; Jorquera et al. 1982; Ueck 1967, Brwpus luevis, Rhinoewna danuinii, and R. mfim). Calciform ceils have been observed in R. mfim. ~oncentrationsof lipid vacuoles are high during embryonic and early premetamorphic stages (Jorquera et al. 1982). The yolk and lipid vacuoles and the ciiia in Brwpus disappear by Nieuwkoop/Faber 65. The boundarv between the duodenum and the iieum is indistinct. The ilium extends ventrdy in a left-handed outer spiral and then from posterior to anterior in 4-8 counterclockwise coils (figs. 5.11-5.13). There are no diverticula, villi, or caeca in the ileum; and the muscularis is thin. The diameter of the intestinal tube and the thickness of the epithelium decreases posteriorly from 70 to 10 Fm. Goblet and microvilious adsoritive cells are present, and endoreduplication is common (Ueck 1967). The iieum resorbs metabolites digested by pancreatic enzymes, and the microvilious celis probably secrete various enzymes (Ueck 1967). High catalase activity found (Daua et al. 1980, R. catesbeiana) in the "anterior intestine" probably relates to the iieum. Enzyme activity decreases during metamorphic climax and resumes dter metamorphosis. The colon (= rectum, coprodaeum) comprises the last two loops of the intestinal spiral (Ueck 1967, X. luevis; figs. 5.11B and 5.13) and can be distingushed from the ileum by its greater diameter. Intense enzyme activity and the presente of microorganisms in a characteristic layer near the epithelial celis suggest the symbiotic breakdown of plant celi w d s . Symbionts are not found when larvae are fed only yolk (Ueck 1967), and Piavaux (1972) found no evidence of cellulases in the intestinal tract. The rectum is positioned longitudindy and dorsdy. Its cuboidal, microvilious, ciliary celis (Nieuwkoop/Faber 47 X. laevts, Fox et al. 1972) indicate resorptive and transport functions, especidy water (figs. 5.11-5.13). Microvdious celis rich in mitochondria, smooth vesicles, and ribosomes are found exclusively in the anterior rectum in X. laevik (Nieuwkoop/Faber 47); during stages 54-58, they are present in the entire rectum (Fox et al. 1972; Ueck 1967). The rectum, bladder, and ureters lead into the cloaca. In contrast to the intestine from the foregut to the colon, the cloaca is not exclusively endodermal in origin (B. I. Baiinsky 1981; Monayong Ako'o 1981). Mesoderm contributes to the nephric ducts. In early larvae (Nieuwkoop/Faber 47), the urodaeum, the dorsal orifices of the nephric ducts, and the posterior proctodaeum are visible. The urodaeum is situated at the posterior margin of the coelom, and the proctodaeum is beyond the splanchmc w d of the splanchnotome (Van Dijk 1959, Ascaphus tntei). In A. tntei (Nieuwkoop/ Faber 51) a ventral thickening of the urodaeum indicates the development of the bladder aniage. This grows ventrdy, so

that an externally unpaired and internally divided, saclike extension is present by Nieuwkoop/Faber 56. The bladder remains externally paired at stage 58, and both parts have a common comection to the ventral urodaeum (Van Dijk 1959). The liver and gdbladder (fig. 5.26A, B, C) lie between the two foregut loops and the midgut loop. The liver surto rounds the manicotto and lies ventral and ~osteroventral I the trachea and esophagus. A large gdbladder is in the anterior region of the liver, often extending beyond the liver in many species. The short neck of the gallbladder and the ductus cysticus lead into the bile duct, which courses laterally along the liver margin and crosses the liver posteriorly. At the pancreas, the bile duct foliows the course of the two pancreatic duas for a short distance before all three duas merge to form the he~ato~ancreatic duct. The development of the liver begins as a ventromedial diverticulum of the midgut aniage between the heart and duodenum aniagen (Nieuwkoop/Faber 35-36, Xenopus luevis). The pocketlike primary hepatic cavity corresponds with the gut cavity and has an outlet into the submesodermal space. Anterior diverticular folds appear and narrow d but the posterior region of the hepatic cavity (Nieuwkoop/Faber 3740). At the same time, the area comected to the midgut narrows and forms the bile duct (Nieuwkoop/Faber 41). The remaining parts of the primary hepatic cavity form the main hepatic duas that transport bile from the liver to the g d bladder. Rernnants of the primary liver cavity form the g d bladder and ductus cysticus (B. I. Balinsky 1981; Nieuwk o o and Faber 1956). ~ fiolds of the anterkr diverticulum break into strands of ceils corresponding to the primary hepatic cavity that are traversed by blood vessels and the sinuses of the viteliine veins (B. I. Baiinsky 1981). The basic histological organization of the larval (and adult) liver consists of a parenchymatic frame of hepatocytes with an imer surface lined by endothelial and Kupffer ceils. The perisinusoidal space (= space of Disse) lined by endothelial celis lies between the hepatocytes and the endothelial ceils. Precursor celis of liver parenchyma are arranged in strands, rich in yolk and lipid vacuoles, held together by desmosomes. and often undergo mitosis. These celis form lob" ules, and the enclosed interceilular space becomes a bile capiilary. The celis increase in size, there is a marked increase in cristate mitochondria (yolk mitochondria), and some rough ER is present (Gosner 20-23). The yolk content is exhausted by Gosner 24, but lipid droplets are present. Liver tissue remains firm during and dter metamorphosis. and the spaces of Disse contain-numerous microvil~ and large nucleoli. Parenchymal celis of Rana catesbeiana change in fine structure in response to thyroid hormone (Cohen et al. 1978 and Fox 1984, Bufo bufo and R. nigromaculata). Peroxisome and micro~eroxisome~densities i&rease in vitro under thyroxine iniuence, and catalase activity is increased by four times (Daua et al. 1980; Daua and Hourdry 1983). Hepatocyte cytology remains unchanged in X. luevis during metamorphosis (Fox 1984). Columnar celis lining the lumen of the gdbladder have short microviiii, ribosomes, numerous smooth-wded veL L

132

BRUNO VIERTEL AND SUSANNE RICHTER

sicles, iipid vacuoles, a high density of mitochondria at the apicai pole, and invaginate intercellular junaions. The wali of the bile d u a is lined bv ciliated and microviiious cuboidal celis and a smaii number of ciliarv celis with microdious stubs. A fuzzy coat suggests that secretion is probable (Toner and Carr 1968). Ciliary celis extend posterior to the nonciliated neck as far as the orifice of the hepatopancreatic d u a (Fox et ai. 1970, X. k s at Nieuwkoop/Faber 47). The pancreas (figs. 5 . 2 6 4 C, and 5.27A, B) iies inside the second foregut loop with its distai region adjacent to the manicotto. The two pancreatic duas pass to the left and right of the subintestinalvein and subsequentlyflow into the bile d u a (Fox et ai. 1970). In many species the hepatopancreatic d u a branches distay into hepatopancreatic ductules and a distai hepatopancreatic orifice that marks the end of the midgut. The pancreas develops from three endodermai aniagen. The first is a divertidum at the connection between the primary hepatic cavity and the midgut (hture orifice of the bile d u a ; Nieuwkoop/Faber 37-38, X. laevis). This divertidum grows into two parts that surrounds the bile dua. The third part develops next to the dorsai midgut waii and, by Nieuwkooppaber 39, contaas the first aniage and separates from the midgut waii. The ventral pancreatic aniagen each differentiate into a pancreatic dua; the dorsai anlage does not. Differentiation of acinar celis starts at Nieuwkoop/Faber 41 in X. h i s . A single columnar celi layer lines the ducts (Fox 1984; Fox et ai. 1970). Concentrically arranged pyramidal ceils form acini and line the branching ducts. The apicai pole of each celi projects into a common central lumen (Fox et al. 1970). In Diswglosssus the exocrine pancreas does not difTerentiate until the p ceils of the endocrine pancreas are visible (after Gallien/Houllion 28; Beaumont 1970). It is not known if the larvai exocrine pancreas synthesizes lipases, nucleases, peptidases, and prote&es as in ddts and other Amphibia (Karlson 1972). During metamorphic climax (Nieuwkoop/Faber 61 ofX. l&), 70%-80% of the total size and weight (Atkinson and Little 1972; Race et ai. 1966 in Fox 1984) of the exocrine tissue degenerates. After metamorphosis the pancreas is reconstructed with a higher proportion of endocrine ceils.

therefore be regarded as an aspect of regional differentiation in the larvae. The close morphologicai relationship of the excretory and reproductive systems is mainly based on embryonic propinquity, especialiy in males, where iudney tubules transport sexual produas. The anatomy and development of the urinary and reproductive systems are treated in sequence below.

Urina? System: Anatomy Most anuran larvae iive in water, but even those larvae that develop outside of water occur in moist environments, and fluid exchange occurs between the lamae and the environment. The maintenance of acid-base balance, electrolytes, and fluids requires coordmation between renal naion and ion and water movement across the digestive system, gih, skin, and in later stages, the urinary bladder (Bentley 1971; Deyrup 1964).Tadpoles are good osmoregulators (Candelas and Gomez 1963, Leptodactylus aalbilabris and Limnonectes cancrivoms; M. S. Gordon and Tucker 1965, Limnonectes canmivoms; R. M. Jones 1980, Spea multiplicata; McClanahan 1975; see chap. 8). The nitrogenous excretion rate is relatively h g h soon after hatching, declines during larvai development, and increases again at metamorphosis. Larvai anurans usualiy are ammoniotelic, and in response to enzyrnatic changes in the liver (J. B. Balinsky 1970; Cohen 1970, Rana catesbeiana), tadpoles shift to ureotely at metamorphic climax (Cohen 1966, Hylageographica, 1970, R. catesbeiana; M. S. Gordon and Tucker 1965, Limnonectes cancrivms Munro 1953, Bufo bufo and R. temporana; Tahin et ai. 1979, Hylageographica; Zamorano et ai. 1988, Caudiverbera caudiverbera). In the uricotelic Phyllomedusa sauvagii (Shoemaker and McClanahan 1982), Scaphiopus couchii and Spea multiplicata (R.M. Jones 1980), arnrnoniotelic Pipa carvalhoi (Tahin et ai. 1979), and probably severai leptodactylids (Candelas and Gomez 1963, Leptoaktylus abilabris; Shoemaker and McClanahan 1973, Leptoda+us bufonius) and myobatrachds (A. A. Martin and Cooper 1972, Geominia viccmiana), urea predominates i the premetamorphic n stages (see Ashley et ai. 1968, Rana catesbeiana). In Xenopus, the ammonia excretion rate remains alrnost unchanged throughout development (J. B. Balinsky et ai. 1967, 1972; MUNO 1953; see E M. Wright and Wright 1996, R. catesbeiana). Urogenital System The excretory system of anuran larvae likely evolved from The excretory and reproduaive systems of anurans show a hypotheticai holonephros (Price 1897) derived from the strking morphologicd interre~ationsh~s dependen- entire nephrotomai plate and with one nephron derived and cies as a resdt of common mesodermai (i.e., intermediate from each nephrotome. Differentiation of the nephrons plate = nephrotome or mesomere) origins and similar trans- progresses in an anteroposterior direaion, and anterior port nc&s. The nephrotome, desGte its irreversible de- nephrons are less organized than those that develop later. termination in the late n e u d a (Bossard 1971, Bombina Thus, in anamniotes, a hypotheticai uniform holonephros varigata; Burns 1955; Gipodoux and Hakun 1978, Bufo may have developed into a pro- and opisthonephros adapted bufa; Hakim and Gipodoux 1978; B. bufa; Tung 1935, to special developmental and naionai conditions (Ciantar ~ ~ s c o g l opictus), hs few developmental options in its 1983; Fraser 1950; R W. Haii 1904; Richter 1992; Van ~s~s responses to the surroundmg endoderm and chordameso- Den Broek et ai. 1938; M. H . Wake 1979). Both types of derm. The onset and charaaer of development of the uro- kidneys are moderate-size structures lying retroperitoneaiiy genitai system are modified by the position of its blastema against the dorsai body wa (fig. 5.28A). in the urogenitai field. The spatiotemporai appearance of the The paired, triangular pronephric kidneys (= head kidreproductive, pronephric, and opisthonephric system must ney) generaiiy are located ventral to postotic somites III-V

ANATOMY

/"*

CLT

Fig. 5.28. Anatomy and development of the larvai kidney. (A) Schemauc representation of the configuration of the larva1 kidneys. (B-D) Pronephros of Rana esnrlenta (Gosner stage 27); l a and 1b in (B) connect with l a and l b in (C), and 2 in (C) connects with 2 in (D). Development of opisthonephros of R. esculenta. (E) Developing nephrons

showing renal vesicle (upper right) and S-shaped body ( h e r right). (F-G) Young opisthonephricnephron; 1 in (F) connects with 1 in (G). Arrows indicate diiection of flow; double-headed mows mark boundaries of tubular parts with an open lumen that connect elsewhere. Ser table 5.3 for abbreviations.

(Fox 1962b) or slightly anterior (M. I. Michael and Yacob 1974, Hyla arborea samignyi). This kidney begins to function at the muscular response stage (Carnbar 1947a, Rana dalmatzna; Rappaport 1955, R. pipiens) shortly after the pronephric duct joins the cloaca (Deymp 1964; J&ee 1954a, R.pipiens; Richter 1992,R. esculenta). This systern functions

during most of larvai life and gradudy disappears during metarnorphic climax. Each kidney consists of a coiled tubular complex, a large posterior cardinal sinus, interstitiai hematopoietic tissue, and a glomus (figs. 5.28A and 5.29B). The ceil structure of the tubular complex is similar arnong anurans (Ciantar

134

BRUNO VIERTEL AND SUSANNE RICHTER

ANATOMY Fig. 5.29. Anatomy and development of the pro- and opisthonephros. Pronephros in Rana escuenta: (A) Glomus and tubular c~mplex ( ~ o s n c 27); arrow indicates a mesangial ceii; scaie = 50 p k (B) Visr ceral podocytic epithelium of the glomus; scale = 5 pm. Opisthonephros of Rana esmenta ( G D and F-G), and Bufo bufo (E): (C) Sshaped body; arrow indicates an invading mesangioblast; scaie = 30 Fm. (D) Peritoneal fumel; scaie = 50 pm. (E) Immature glomerulus ofBufi, h oscaie = 20 pm. From Naito (1984); repcinted by permisf; sion of Niigata University School of Medicine. (F) Glornerulus and tubular cornplex; arrow indicates rnesangial celis; scale = 40 pm. (G) Viscerai podocytic epithelium of glornerulus; scaie = 10 pm. See table 5.3 for abbreviations.

135

1983, Xenupus l e ; Fox 1963,1970, 1971, Rana temporaria; Fox and Hamilton 1971, X. hms; Gibley and Chang 1966, R. pipiens; Hurley 1958, R. sylaatica; JafTee 1954a, R. pipiens; Kindahi 1938, R. tempmaria; Matsurka 1935, Bufofomzosus; M. I. Michael and Yacob 1974, B. v. viridis, H. a. sam>nyi, and R. 1: dibunda; Richter 1992,R. esculenta; Schiuga 1974, B. bufo and Pelobates&scus). Each tubuiar complex (figs. 5.28B-D, 5.29A) is composed of three nephrostomal segments that lead from the coelom, via nephrostomes, to three short proximal tubuie segments. The cuboidal epithelium of the ciliated nephrostomal tubuie segment possesses straight intercelluiar junctions and irregularly shaped mitochondria. The proximal segments unite to form a common proximal tubuie segment. Micropinocytotic vesicles are visible between the microvilli of the apicai cell membrane of the proximal tubuie cells. Proximal tubuiar reabsorption, large numbers of celluiar infoldings and lateral interdigitations, and the presence of enzyme systems have been demonstrated (Fox 1984, &nopus h s ; Grard and Cordier 1934a, b, Alytes obstez%mns, B. bufo, Disc~lossuspictus, and R. temporaria; Hah 1974, Bombinu oriental& JafTee 1954b, 1963, R.pipiens; Piatka and Gibley 1967, R.pipiens). Golgi networks are scarce in R. temporaria (Fox 1970) but always present in X. laePis (Fox and Hamilton 1971). The common proximal tubuie segment connects via an intermediate ciliated tubuie segment that more distdy becomes the conneaing tubuie segment. The distal tubuie segment has a flat, striated epithelium where water is reabsorbed (Grard and Cordier 1934a). The connecting tubuie empties into the pronephric duct. The glomus is a slightly elongated mass of capiilaries and intercapillary cells w i h n a spongy mesangial matrix and covered by a visceral podocytic epithelium (fig. 5.29B). Podocytes have well-detined, interdigitating prirnary, secondary, and tertiary processes and pedicels. The afTerent vessel of the capdary tufi is derived from the dorsal aorta, and the efferent vessel is derived from the posterior cardinal vein (figs. 5.28B and 5.29A). Epithelioid-like cells around the afTerent vessels may regulate arterial pressure (Richter 1992; Richter and Splechtna 1990, Rana esculenta). The space between the variou tubuie segments is occupied by sinusoids of the posterior cardinal vein that form an extensive vascular network around the tubuies. The pronephric sinus plays an irnportant role in hemopoiesis (K. L. Carpenter and Turpen 1979, Ranupipiens; Foxon 1964, G. Frank 1988, R. es-

cuhnta; Hollyfield 1966, J. D. Horton 1971b, R. pzpns; Latsis and Saraeva 1980, R. tempmaria; Plytycz and Bigaj 1983; Turpen et al. 1981a, b; Turpen and Knudson 1982, R. pipiens). The pronephric kidneys are drained by the pronephric duas (= Woiffian ducts), which aiso serve as the vasa deferentia. The duas originate from the anterior mesodermal d u a priiordiunl along with a minor contribution from the ectodermal cloacal diverticulum (Fox 1963; Fox and Hamilton 1964; Poole and Steinberg 1977, Xenops laevis; Richter 1992, R. esculenta). The duct epithelium of polygonal cells joined by junctional complexes is underlain by a basal membrane and a fibrous extracelluiar matrix. Except for a few short microvilli, the lumen w d is almost smooth (Ciantar 1983, X l m i ; Delbos 1975, R. dalmatina; Fox and Hamilton 1971, X. laeozs). The oval, opisthonephric kidneys are generdy asymmetrical and d i s s d a r in developmental rate and shape (Kindahi 1938, R. tempmaria; Nodzenski et ai. 1989, Leptobracbium montanum, Leptoiahx~racilis,and Mgopblys nasuta). These kidneys lie to either side of the dorsal mesentery, are covered ventrdy by the adrenal glands, and start to function shortly before hind lirnb development (Richter 1992, R. esculenta). The opisthonephric kidneys are drained by the pronephric ducts (= Woiffian or opisthonephric ducts) and can be divided into lateral (primary nephrons), medial (younger nephron generations), and dorsomediai (nephric anlagen and opisthonephric blastema) mnes. Malpighian corpuscles are aiways in the ventral mne. Ciliated peritoneal furinels restricted to the ventral kidney margin are numerou and prominent during larval life. An opisthonephric nephron (figs. 5.28F-G and 5.29F) includes a neck segment at the urinary pole and an intermediate segment interposed between the thicker proximai and distai tubuie segments. Both have cuboidai, ciliated epithelia. The proximal segment, which reabsorbs solutes, is compnsed of cuboidal, granuiated cells with a dense luminal b m h border, and the epithelium has a petmeability gradient (Grard and Cordier 1934b,Alytes obst&cuns, Bufo bufo, Dzs~~lossuspictus, Ranu temporaria) that correlates with the and pattern of tight junctions along the proximai epithelium in aduit R. esculenta (Taugner et al. 1982). This segment shows alkaline phosphatase activity (Hah 1976, Bombina orientalzs). The cuboidal cells of the distai tubuie possess a smooth, apical plasma membrane and basal striations at Nieuwkoop/ Faber 49. The connecting tubuie segment and the colleaing duct of larval and adult pipids are characterized, apart from the occurrence of light cells, by bottle-shaped cells (Bargmann 1937, 1978; Jonas and Rohlich 1970; Jonas and Spannhof 1971) differentiated from light cells at Nieuwkoop/Faber 49. Because their secretory material disappears in saltwater-acclimated tadpoles, these cells probably function in ion regulation (L. Spannhof and Jonas 1969). The cells contain a highiy developed intercelluiar channel system fiiied with acid mucopolysaccharides (L. Spannhof 1956; L. Spannhof and Dittrich 1967). Each renal tubule is associated with a large oval Malpighian corpuscle (fig. 5.29F). The visceral epithelium of Bowman's capsuie is composed of

BRUNO VIERTEL A N D SUSANNE RICHTER

spherical podocytes (fig. 5.29G), and the ceils of the parietal epithelium are flat. The afferent artery, surrounded by epithelioid-like cells (Richter 1992, Rana escuknta), branches into a large glomedar tuft direaly from the renal artery. The efferent vessel gives rise to a rich capiiiary network ensheathmg the tubuiar epithelium (Naito 1984, R. cutesbeiana; Richter 1992, R. escuknta). Data are not available on the uitrastructure of the larval juxtaglomedar apparatus. Nephrons are fiequently innervated by myelinated nerve fibers. Occasiondy, synaptic, vesicle-laden varicosities with dense grandes are associated with both the arteriolar smooth muscle ceils and the distal tubuie ceils (Tsuneki et al. 1984, R. cutesbeiana). The nephrons are surrounded by the venous sinus of the renal portal system where waste material and excess water are removed (Engels 1935, R. temporaria; Miiiard 1949, X. laepis; Richter 1992, R. escuknta). Arterial blood flows fiom renal arteries branching fiom the dorsal aorta (Miiiard 1945, X. h Lymphomyeloid (J. D. Hor) . ton 1971b, R.pipiens; Manning and Horton 1969, X. laepis; Plytycz and Bigaj 1983) and granuiopoietictissues are extensive (G. Frank 1988, R. escuknta; Meseguer et al. 1985, R. ridibunda) in the intertubuiar spaces. Peritoneal nnels usudy present in the opisthonephros (fig. 5.29D) open into the coelom, but their conneaion with Bowman's capsule is often lost before metamorphosis. In most larvae the peritoneal nnels open into the opisthonephric venous sinus, but in Alytes obstetricans and Dtscoglossuspictus (Grard and Cordier 1933, 1934a) and X. laevis (De Waal1973; Kiuist 1936), they retain a conneaion with the nephrons throughout larval life. Peritoneal nnels are densely ciiiated and carry liquid or particles from the coelomic cavity into either the veins or the nephrons (Grard and Cordier 1933,1934b). The urinary bladder originates from a diverticuium extending anterolaterdy from the cloaca at late prometamorphosis and then increases in developmental rate at the beginning of metamorphic climax (Nieuwkoop and Faber 1967, Xenopus hmi, Nieuwkoop/Faber 59; T. L. Poweil and Just 1985, 1987, Rana cutesbeiana, Taylor/Koilros XX-XXV, Gosner 41/42-46; Richter 1992, R. escuknta, Gosner 40). The bladder is suspended in the coelomic cavity by two strands of conneaive tissue and communicates with the cloaca through a wide orilice. The flat epithelium has a smooth
Table 5.3 Abbreviations used in figures 5.28-5.31

luminal surface. A thin muscular layer develops within the bladder w d at metamorphic climax. The bladder accomplishes at least passive transport of water and nitrogenous excretion products and active sodium transfer. The major permeabiiity barriers are the apical and lateral plasma membranes (DiBona et al. 1969, Bufo marinus). Net sodium transport and osmotic water flow across the bladder is stimuiated by hormones (Leaf 1965). Water retention is lowest in young tadpoles and increases at metamorphic climax (Aivarado and Johnson 1966; Bentley and Greenwald 1970; T. L. Powell and Just 1987, Rana catesbeiana). Urinaty System: DevelUpment The development of the paired pronephric kidneys is a progressive event beginning within an anteroposterior wave of tubuiogenesis.The first pronephric anlage inXenopw (Nieuwkoop/Faber 21; Richter 1992, Rana esculenta, Gosner 1617) is a slight thickening of the intermediate plate. Tubuiogenesis involves smali, dorsolateral, metameric outgrowths f the somatic layer of the nephrotomes (fig. 5.29); &ese are initidy blind, then lengthen, bend backward, and unite at their distal ends to form a longitudinal duct that Werentiates into the tubular and ~ r o n e ~ h rduas. The ~ r o n e ~ h r i c ic tubuies increase in length and become highiy convoluted. Celiuiar Werentiation of the tubuiar complex and the pronephric duct occurs in the opercuiar-fold stage (Rana esculenta, Gosner 23; Xenopus, Nieuwkoop/Faber 33-34). The tubuiar complex consists of three ciiiated nephrostomal tubule segments that open into the coelom via three nephro@ stomes (figs. 5.28B and 5.29A). In some species, a vestigial fourth nephrostome may be present (Fox 1962a,R. tempratia; Schiuga 1974, Bufo bufo and Pelobatesfiscus). As development proceeds, the nephrostomes eventudy fuse into a shgle uGt. The nephrostmal tubuies join into three proximal segments that-findy unite to a common proximal tubule segment. The latter continues as an intermediate, ciliated segment, and a nonciliated, distal portion leads into the conne&ng tubuie segment that distd? becomes the nephric duct. At the sarne time the posterior cardinal vein gives rise to a series of s m d capiilaries that rami+ arnong the coiling pronephric tubuies; they rejoin and merge to open into the duct of Cuvier (R. escuknta, Gosner 22; X. h s , Nieuwkoop/Faber 35-36, Miiiard 1949).
I I

A = abdominal cavity, AD = aorta dorsalis, AX = auxocyte-like cell, B = basal Iamina, BC = Bowman's capsuie, BP = buccopharyngeal cavity, C = capillary of the glomular (glomedar) tuft, CE = coelomic epithelial ceii, C1 = inner epithelial mass, CLT = coiiecting tubuie, C 0 = coelom, CPT = common proximal tubuie segment, CS = surface epithelium, CT = comecting tubuie segment, CV = capiiiary of the venous system, CX = cortex, D = degenerating auxocyte, DC = duct of Cuvier, DT = distal tubuie segment, E = endothelial ceii, ED = mdimentary efferent dum, EP = epitheloid ceii, F = foilidar ceiis, G = glomus, GA = glomedus anlage, GC = gonial ceii, GE = gerrn cell, GL = glomeruius, GS = secondary gonium, GT = glomedar tuft,H = interstitial hematopoietic tissue, I = intermediate tubuie segment, M = rneduiia, T MD = macula densa, ME = mesenchymal tissue, MS = mesentery, N = nephrostomal tubuie segment, NS = neck segment, O = oogonium, OC = ovarian cavity, OD = oocyte in diplotene stage, OE = early oocyte (leptotene-pachytene), OP = opisthonephros, OPD = opisthonephric duct, OT = opisthonephrictissue, P = peritoneal funnel, PA = anlage of the peritoneal funnel, PC = pedicels, PD = pronephric dua, PE = paneta1 epithelium of Bowman's capsuie, PF = primary foot process, PG = primitive glomedar capiiiary intercalated benveen afferent and efferent vessel, PGC = primordial germ ceii, PO = podocyte, PRO = pronephros, PT = proximal tubuie segment, R = testicular rete, RA = renal artery, S = sinus of the posterior cardinal vein, SC = spermatocyte, SF = secondary foot process, SG = spermatogonium, ST = seminiferous tubule, T = tubuie anlage, VA = vas aferens, VB = brachial vein, VCA = anterior cardinal vein, VCP = posterior cardinal vein, VE = vas efferens, and VJ = extemal jugular vein.

ANATOMY

137

In the pronephros, the Werentiation of the Malphghian corpuscle is independent of tubulogenesis (Gipouloux 1957; Gipouloux and Girard 1986). The visceral epithelium of the Maphighian corpuscle is formed by the fusion of three successively enlarging folds of the splanchnic layers of the nephrotomes (Richter 1992, R. esculenta). Blood vssels from the dorsal aorta and posterior cardinal vein eventuaily give rise to the glomular tuft and the afTerent and efferent glomular vessels (Richter 1989, 1992, and Richter and Splechtna 1990, R. esculenta). In Xempus, the glomular blood supply starts at Nieuwkoop/Faber 35-36 (R. esculenta, Gosner 19). The entire pronephros of Xenupus becomes functiona at Nieuwkooppaber 37-38 (R. esculenta, Gosner 24) and attains its maximum development at stage 56. In R. temporaria, pronephros size increases from Cambar/Marrot 2 9 4 7 (Nieuwkooppaber 41-55) and decreases afier stage 49 (Fox 1962a). The pronephric nuclear population and the total pronephric tissue volume of R. esculenta increase with larval length (Fox 1961). InXenupus the mitotic index of pronephric tubule celis is reduced afier Nieuwkooppaber 51, although tubule length and lumen volume increase until stage 55 (Fox 1984). Chopra and Simnett (1969a, b, 1970a, b, 1971a, b) and Simnett and Chopra (1969) suggested that the pronephric growth rate in Xenupus is regulated by organ specific, antigenic mitotic inhtbitors synthesized during the development of the pronephros. The onset of pronephros regression apparently is variable among species but generaiiy begins midway through prometarnorphosis (Fox 1984). Histological changes in thyroid tissue and cirdating thyroxine seem related to the degeneration process (M. H. I. Dodd and Dodd 1976; Fox 1971, Rana tempmaria; Hurley 1958, R. sylvatica; M. I. Michael and Yacob 1974, Buf v. Ytridis, Hyh arbmea savignyi, and R. 1: ridibunda; Richter 1992, R. esculenta). Reduction of thyroid activity with thyrostatics (Fox and Turner 1967, R. temporaria andX. laevis; Gardener and Peadon 1955, Eleutherodactylus martinicensis; Hurley 1958, R. sylvatica), thyroidectomy, hypophysectomy (Fox 1963), or the administration of prolactin (Vietti et ai. 1973, R. temporaria) inhibits pronephric degeneration. Metarnorphic degeneration of the pronephros is uneven and variable in rate (Fox 1970, R. temporaria) among adjacent tubular ceiis. The tubules become somewhat swolien and the expanded lumens often contain necrotic tissue. The nephrostomes, whch fused during the premetamorphic stages, are bordered by connective tissue (Nieuwkoop and Faber 1967, X. W). blood vessels and the arnount of The hematopoietic tissue increase, and the Malphghian corpuscle becomes increasingly dense and smaller. Tubular celis contain large degeneration bodes in which mitochondria or other organeiies undergo lysosomal hydrolysis. Synthesis of DNA ceases, whle RNA synthesis, perhaps involved in the synthesis of lysosomes, proceeds unchanged (Fox 1970, 1971,Rana temporaria 1983).The brush border of the distal tubule segment, the cilia of the nephrostomal and intermediate segment, the tubular basement membrane together with its overlying coliagen layer, and infoldings of the plasma

membrane tend to persist almost to the final stage of tubule degeneration. Necrotic tubules first change into solid strands and are then ingested by leucocytes (Fox 1970, R. temporaria; Richter 1992, R. esculenta). In Xenupus, the compact anlage of the pronepM~c duct begins to grow posteriorly (Nieuwkoop/Faber 26; Rana esculenta, Gosner 17-18) by their own independent growth along the lateral mesoderm (Cambar 1952a, and Cambar et al. 1962, R. dalmatina; Carnbar and Gipouloux 1956b, Bufo bufo; Fox 1963; Richter 1992, R. esculenta; Vannini and Giorgi 1969, B. bufo). The celis of the latera mesoderm and the duct tip possess filopodia (Ciantar 1983,X. laems and Delbos 1975, R. dalmatina). Essentially, duct elongation appears to be a migratory phenomenon (Gipouloux and Cambar 1961, B. bufo and Overton 1959, R. @p"ns) occurring by celi rearrangement. The migration of the anlage of the pronephric duct is stimulated and guided by the presente of morphogenetic factors, probably protein substances. These substances located on the celi surface of the lateral rnesoderm (Cambar and Gipouloux 1973, R. dalmcttina) create a predetermined pathway in space and time (Carnbar and Gipouloux 1970a, B. buf, Dism~lossuspictw, R. daland matina and Tung and Ku 1944, B.~agarizans R. nigroand mdcuhta). The direction of duct elongation is not d u e n c e d by the somites (Gipouloux and Cambar 1970a, B. bufo and R. esculenta). The rectal divertida irst appear in Xenupus at Nieuwkoop/Faber 32 (R. esculenta, Gosner 20) as srnaii grooves in the dorsal proctodeal w d . They grow anteriorly but do not exert an inductive stimulus on the growing tips of the duas (Cambar 1952b, R. dalmcttina), which join the cloacal divertida at Nieuwkoop/Faber 35-36 (R. esculenta, Gosner 22). The celiular Werentiation of the duct epithelium is completed at Nieuwkoop/Faber 39 (R. esculenta, Gosner 23), and that of the rectal divertida is completed stage 43 (R. esculenta, Gosner 25). The pronephric ducts between the pronephros and the opisthonephros begin to degenerate at Nieuwkoop/Faber 59 (R. esculenta, Gosner 41) and disappear at stage 60 (R. esculenta, Gosner 42). The posterior regions of the duas, which belong to the opisthonephric anlage, thcken and the epithelium becomes stratified. At Nieuwkoop/Faber 58 (R. esculenta, Gosner 45), the wail consists of 2-3 celi layers enveloped by connective tissue. A functional pronephros is essential for the development and maintenance of the pronephric duct (Carnbar 1954, R. dalmatina; Vannini and Giorgi 1969, B. buf). If the pronephros is extirpated, the anterior part of the duct reduces to a solid strand, while the posterior, tubular haifnctions as an inducer for the opisthonephric blastema (Carnbar 1947b, R. dalmatina). This inductive stimulus and the required self-differentiationpotential of the blastema are indispensable for the development of the opisthonephric tubules (Carnbar 1948, Alytes obstetrimns, B. buf, R. dalmcttim, and R. esculenta; Cambar and Gipouloux 1956c, B. buf; Van Geertruyden 1946, 1948, R. temporaria). In B. bufo and R. dalmatina the inductive activity of the pronephric duct is attributed to extraceliular substances (Delbos 1975 and Gipouloux and Delbos 1977). Both the inductive distance

138

BRUNO VIERTEL A N D SUSANNE RICHTER

benveen the duct epithelium and the blastema (Cambar and Gipouloux 1956b, B. bufo and Dism~lossuspictus Gipouand loux and Cambar 1970b, B. bufo, D.pictus, and R. dalmatina) and the duration of inductive potency (Cambar and Gipouloux 1970b, B. bufo, D. pictus, and R. dalmatina) depend on the pronephric duct and vary among species. The opisthonephros develops in the posterior region of the intermediate plate where the nephrogenous cor& develop from dissolved nephrotomes and gives rise to a set of distinct condensed ceii aggregations that soon become renai vesicles (Gipouloux and H a k h 1976, E. bufo and X. laevis; E Gray 1930, Rana temporaria; Richter 1992, R. escuhnta; Schluga 1974, E. bufo and Pelobatesfiscus; Vannini and Sabbadin 1954, R. escuhnta). The renal vesicles are dose to the epithelium of the nephric duct and receive an inductive signai from the latter. Their arrangement is not correlated with the segmentation of the myotomes, but there is a tendency for the primary units to develop at regular intervds. I n z n o pus, the first renal vesicles, which are anlagen of primary nephrons, are present at NieuwkoopFaber 47 (R. escuhnta, Gosner 25). Functional peritoned fumeis appear at Nieuwkooppaber 48 (R. escuhnta, Gosner 26). The vesicles successively develop into tubules joining the pronephric duct, which becomes the opisthonephric duct. The remaining ceiis of the nephrogenic blastema proliferate, extend dorsomedially and give rise to new generations of nephrons. As new nephrons are added, the nephric duct and the primary units are pushed toward the lateral margin of the kidney (Arnauld and Cambar 1970 and Cambar and Arnauld 1970, R. dalmatina; E Gray 1930, R. temporaria). Coiieaing duas are formed from bulges in the wall of the nephric d u a (Hirsch 1938; Richter 1992, R. escuhnta). Younger nephric anlagen inductively stimulated by older nephrons join the epithelium of the coiieaing ducts at the dorsai base of the more differentiated nephrons. Morphogenesis of the nephron starts with the accurnulation of fibroblast-like ceiis (Ciantar 1983, X. laevts; Delbos 1975, R. dalmatina; Richter 1992, 1995, R. escuhnta) arranged into hoiiow renal vesicles (fig. 5.28E) that probably require an extraceiiular matrix and a basement membrane for development of polarity (Ekblom 1989; Saxn 1987). ?fie formation of the S-shaped body (figs. 5.28E and 5.29C) is generally attributed to two opposed invaginations caused by diierential growth of the epithelium. The nephron anlage subsequently joins the uriniferous dua. Further elongations and ceii dierentiations of the epithelium lead to the estabiishment of tubular segments and to the formation of Bowman's capsule. The segment of the tubule attached to the visceral epitheiium of Bowman's capsule becomes the macula densa (figs. 5.28F and 5.29F; Richter 1992, 1995, R. esculenta; Schluga 1974, E. bufo and Pelobatesfiscus). The forrnation of the macula densa is correlated with the development of the glornerular capillary systern (Richter 1992, 1995, R. escuhnta). The glomenilar tuft develops from capillary sprouts of interstitial arterial and venous vessels that grow into the anlage of the Bowman's capsule, merge, and form a conglomerated capillary network by the fusion of freeending protrusions (fig. 5.29E; Ditrich and Lametschwandtner 1992,Xenopw b i s ; Naito 1984, R. catesbeiana; Richter

Fig. 5.30. Schematic representation of the developing gonads (both seles) and Bidder's organ. (A) Formation of genital ridge. Differentiaaon of gonads: (B) proliferation of coelomic epitheiiai celis of indifferent gonad and (C) cortex and medda pattiaiiy separated. Ovarian development: (D) cortex and medda separated by mesenchymal tissue, (E) meddary tissue reduced and ovarian cavities dzerentiated, and (F) development of follicles. Testicular development: (G) cortex and medda closely associated, (H) cortex separated into an inner epitheliai mass and an outer epitheiium (= surface epithelium),and (I) meddary mass dfierentiated into testicular rete and efferent ducts; formation of seminiferous tubules. Differentiation of Bidder's organ: (J) proliferation of sornatic celis, (K) no definite cortico-meddary structure evident; germ ceiis scattered among sornatic celis, (L) germ ceiis enlarge into auxocyte-like ceiis and somatic celis p d d y degenerate, (M) dzerentiation of auxocyte-like celis, and (N) auxocyte-like ceiis degenerate, and oocytes in meiotic stages. See table 5.3 for abbreviations. Redrawn from Tanimura and Iwasawa (1986, 1988); reprinted by permission of Science Report @&ata Univmity and Japanese Society of Developmental Biologists.

1992 and Richter and Splechtna 1990, R. escuhnta). The vessels are accornpanied by immigrating rnesenchymal ceiis distributed between the capillaries of the glornerular tuft and establish their position as rnesangial, lacis, and epithelioid ceiis (Richter 1992, 1995; Richter and Splechtna 1990, R. escuhnta). Peritoned fumels develop as ventral outgrowths of the vesicle (fig. 5.29D). They either detach from the S-shaped body (Engels 1935, Rana tempovaria; Grard and Cordier 1933, 1934b, E. bufo and R. tempovaria; E Gray 1936 and Kindahl 1938, R. temporaria; Richter 1992, 1995, R. escuIenta; Schluga 1974, E. bufo and Pelobatesfiscus) or remain comected with the neck segment until rnetamorphosis begins (Grard and Cordier 1933, 1934b, Alytes obstetricans and D~m~lossuspictus). the latter case, they separate frorn In the tubules by the destruction of their proxnal epitheiium. In Xenupus, communication between the nephron and the peritoned fumei persists throughout life (De Wad 1973; Kunst 1936). The peritoneal fumels usudy join the renal veins and, clustered near the ventromedial margin of the kidney, unite at their distal ends with the ventral peritoneum (Kunst 1936). At premetamorphosis the prirnary nephrons lose their connections with the pronephric duct and inally degenerate at metamorphic clirnax (E Gray 1930, 1932; Hirsch 1938; Kindahi 1938, Rana temporaria; Richter 1992, R. escuhnta). In Xenupus, degeneration starts at Nieuwkoop/Faber 55 (R. escuhnta, Gosner 32), and the ceiiular structure of remaining nephrons changes at metamorphic clima. The number and metaboiism of peroxisomes and catalase activity increases in the proximal tubule segment of R. catesbeiana at metamorphosis. No change in catalase activity or peroxisome number is deteaed inXenps, which remains aquatic and ammoniotelic throughout development (Daua et ai. 1982, 1983). The nephrons and the renal vascular system are reorganized in different zones within the kidney; each zone is cornposed of distinct nephron segments and special blood vessels, and the architecture of the metamorphic ludney now resembles the definitive adult ludney (Barch et al. 1966, Bufo bufo;

140

BRUNO VIERTEL AND SUSANNE R I C H T E R Fig. 5.31. Gonadai Merentiation of Rana nigrommulata. (A)Primordial gonad (Gosner stage 24), scaie = 50 pm. Sexualiy inMerent gonads at (B) Gosner 26 and (C) Gosner 27; cortex and medda clearly distinguished, direct contaa (awms) between corticai and medd a r y ceiis; scaie = 20 pm. Rudimentary testis: (D) Gosner 30, corticai and meddary regions closely associated (nrtenskr); mesenchyrnai (awmheadr) are evident; scale = 40 ceiis ( a m s ) and blood cap~laries pm; (E) Gosner 35, inner epithelial mass separated from surface epithelium, shows cordlike structure; scale = 50 pm; and (F) Gosner 46, rudmentary seminferous tubuies ( a m s ) formed from the inner epitheliai mass, and testicular rete connected to rudimentary seminiferous tubuies (awmheadr); scaie = 100 pm. Rudimentary ovary: (G) Gosner 30, cortex and medda separated, rnesenchyrnal A s ( a m s ) and blood capiaries (awmhead) visible; scde = 40 pm; (H) Gosner n 40, formation of ovarian cavity evident i meddary celi mass, primitive follides (awms) visible in cortex; scale = 100 pm; and (I)Many developing foiiicles present; scaie = 100 pm. See table 5.3 for abbreviations. From Tanimura and Iwasawa (1988); reprinted by permission of the Japanese Society of Developmental Biologists.

Bargmann et al. 1955, X. 1m3; Berton 1964, B. b . Hyh u, arborea, R dalmatinaJ and R . esculenta; Lametschwandtner et al. 1978, Bombina variejata, B. buPJR. dibunda, and X. l m 3 ; Linss and Geyer 1964,R . esmlenta; Morris and Campbell1978, B. ma&us; Ohtani and Naito 1980,B. buf; Richter 1992, R. esmlenta).

Reproductive System: Anatomy The gonads of male and female larvae are paired structures that develop from sexually undifferentiated primordial gonads located near the presumptive kidneys. Gonadal differentiation in Xenupw starting at Nieuwkoop/Faber 49, Gosner 29 in Rana nigromacuhta (Tanimura and Iwasawa 1988) and Gosner 26 in R. &bunda (Ogielska and Wagner 1990) is completed in young frogs. The primordial gonads appear as longitudinal bilateral thickenings covering the ventromedial region of the opisthonephros (fig. 5.30A) and are attached by the mesogonium to the inner body wall. The central region (= mesogoniurn) gives rise to the gonad proper, and the anterior region (= progonium) forms the fat bodes. The primordial gonad, differentiated from anterior to posterior, is composed of meduiiary and cortical regions covered with a basa lamina (Iwasawa and Yamaguchi 1984,xnupusb s ; Lamotte et al. 1973, Nimbaphrynodes occdentdis; Lopez 1989, Bombina orientalis; Merchant-Larios and Vdalpando 1981, R . pipiens andX. h Ogielska and Wagner 1990,R . ridibunda; Tani; mura and Iwasawa 1986, B u . japonuus fomsus, 1988, 1991, R . n~romaculata; Tanimura and Iwasawa 1989, Rhaco~hows arboreuss'). lack of ultrastructural differences beA &een medull4 and cortical ceiis of the primordial gonad indicates a cortical origin of the meduiiary tissue. In later stages, the mesogonium is invaded by blood vessels, nerves, and opisthonephric interstitial tissue. The primordial germ cells (= PGCs) located in the cortical region of the undifferentiated gonad form a primitive germinal epitheiium together with surrounding prefoliicular ceils. The PGCs commonly become smaller after they enter the genital ridge in association with proliferation and digestion of yolk platelets (Lopez 1989, Bombina orientafis; Ogielska and Wagner 1990, R. ridibunda; Zust and Dixon 1975, 1977, X. W s ) . Tight junctions between PGCs and prefohcular cells may prevent the passage of molecular substances between the two ceii types. Prefohcular ceiis contain a flattened nucleus and few mitochondria (Delbos et al. 1984,R.pipiens). Gonadal sex differentiation dders tempordy among species but usudv coincides with metamor~hic climax. Female , and male gonads are distinguishable by the number and sim of germ ceiis and the amount of meduiiary tissue (figs. 5.30D-I; Ijin and Egarni 1975, X. L ; & Kobayashi 1975, Ranu n@ro&uluta; Shirane 1986, R .japonica i d R. n e o m u l a t a ; Tanimura and Iwasawa 1987,Bufjaponicusfmmosus; Zaccanti et al. 1977, R . dalmatina, R esculenta, and R . latartei).Male gonads can develop in a differentiated, semidifferentiated. or undifFerentiated mode. The undderentiated and semidifferentiated types often express spontaneous early female sexuality (Tanimura and Iwasawa 1989, Rhacophms arbmeus; Takahashi 1971, Rana rugosa; Zaccanti et al. 1977, Rana dafmutina and R . bastei). In Phyllomedusa sau-

vagii, female gonadal differentiation initially displays a testicular trend (Rengel and Pisan 1981). Larval ovaries (figs. 5.30D-F and 5.31G-I) are larger than testes and have a slightly irregular outline. They usudy consist of meduiiary and cortical tissue in which the proliferating gonial ceiis form several groups of oogonia and oocytes (Iwasawa and Yamaguchi 1984, X. W s ; Lamotte et al. 1973, Nimbaphlynodes occdentalis; Lopez 1989, Bombina orientalk; Ogielska and Wagner 1990, R . ridibundu; Tanimura and Iwasawa 1986,B u .japonicusfomsus 1988,1991, R. nigromaculuta 1989, Rhacophows arboreus). The cortex is distinctiy separated from the meduiia by an aceiiular coliagenous layer, and the meduiiary tissue has degenerated and lost its sex cord Dattern. The cords. which are continuous with the mdiments of the rete ovarii svstem. dissolve. Small lumens formed in this way enlarge k d finhy fuse to the ovarian cavity. The meduiiary connective tissue thus becomes a simple epithelial layer the ovarian cavities. Oogonia measure 17-25 pm in diameter and cytologic d y resemble PGCs except for the lack of pigrnent granules and abundant ER. Oogonia undergo periodic proliferation to replenish oogonial nests of about 16 gonial cek (Coggins 1973, Xenupw W s ; Lopez 1989, Bombina oriental&; Redshaw 1972, X. W s ) .Intercellular bridges appear to be responsible for synchronous developmeni d&Gg oogenesis (Coggins 1973, X. W s ; Ruby et al. 1970, Ranup@iens). The transition from primary to secondary oogonia (= primary oocytes) is characterized by a dark cytoplasm (Eddy and Ito 1971. X. hmi) and a channe from lobed to round nuclei. Oogonia and primary oocytes contain germ plasm (Al-Mukthar and Webb 1971; J. B. Balinsky and Devis 1963; Coggins 1973; and Kalt 1973a,X. W s ; Delbos et al. 1971, R. dalmatina; Eddy and Ito 1971, R . catesbeiana, R. clumitans, and R . pipiens; Kress and Spornitz 1972, R. esmlenta and R. temporaria). Prophase of primary oocytes is initiated in larvae prior to metamorphosis (Al-Mukthar and Webb 1971; Coggins 1973; Jmgensen 1973; Lofts 1974; Lopez 1989; Ogielska and Wa~ner 1990: Van Gansen 1986). At metamor~hic climax larval ovaries exhibit oogonia and leptotene, zygotene,
V

142

BRUNO VIERTEL A N D SUSANNE K I C H T E R

pachytene, and diplotene oocytes. Near the end of leptotene, oogonial chromosomes attach to the inner surface of the nuclear membrane (= bouquet arrangement) at a site adjacent to the juxtanuclear mitochondrid aggregate, which &equently forms a cap over one end of the nucleus. The zygotene stage is characterized by the formation of short, axiai chromosomes. Paired homologous chromosomes form synaptonemai complexes, and s@apsis is completed in pachytene. Extrachromosomai rDNA, functioning in rRNA synthesis, increases, and multiple nucleoli are formed. The oocytes of a celi nest separate and become surrounded by follicular celis that differentiate into steroid-producing follicd a r epitheiium (Redshaw and Nicholis 1971, Xenopzs laevis; Saidapur and Nadkarni 1974, Euphlyctis cyanuphlyctis, H~oplobatrachus ti3eYinus, Limnonectes keralensis, and Uperodon systoma; Thibier-Fouchet et ai. 1976, X. laevis). During folliculogenesis, interceliular bridges disappear and oocytes begin to develop asynchronously. The cytoplasm and nucleus enlarge in diplotene oocytes, synaptonemal complexes are absent, and the chromosomes are in the lampbmsh stage (Coggins 1973; Wischnitzer 1976, X. laepis). During early diplotene, the Balbiani body is formed by condensation of the juxtanuclear rnitochondrial aggregate. In the late diplotene oocyte, dispersed germ plasm is stored in the cortex of the oocyte (J. B. Balinsky and Devis 1963, X. laevis). In most anuran larvae, testicular Merentiation starts at premetamorphosis (figs. 5.30G-I and 5.31D-F ; Iwasawa and Kobayashi 1976, Rana n&rmaculata; Iwasawa and Yamaguchi 1984, X. laevis; Lamotte and Xavier 1973, Nimbaphrynoides occidentalis; Lopez 1989, Bombina onintalis; Ogielska and Wagner 1990, R. &bunda; Tanimura and Iwasawa 1986, Bufojaponicus f m s u s , 1988,1991, Rana niyromaculata, 1989, Rbacophoms arbmeus).The primitive germinal epithelium disintegrates to a simple peritoneal surface epithelium. The male gonial celis or spermatogonia, together with their surrounding follicular cells, are scattered throughout the compact meduilary tissue, which becomes mixed with interstitial mesenchyrnal celis. In the center of the meduila, a network of tubules becomes increasingly distinct and 6naiiy Merentiates into seminiferous tubules, which are attached to the opisthonephric tubules by the vasa efferentia. In R. nZgromaculata (Tanimura and Iwasawa 1988) and Rbacophms arboreus (Takasu and Iwasawa 1983; Tanimura and Iwasawa 1989), seminiferous tubules are derived from the inner layer of the cortical epithelium. Medullary cor& near the hilus of the testicular primordium form the rete testis. Spermatogenesis is virtuaiiy identical among species (Deuchar 1975; Kalt 1973a, b; J. B. Kerr and Dixon 1974; and Reed and Stanley 1972,Xenupus laevis; Iwasawa and Kobayashi 1976, Rana n&romaculata; Rastogi et al. 1983, R. esculenta; reviews by Grass 1986 and Lofis 1974). Two types of spermatogonia are present in premetamorphic larvae; p r w single spermatogonia are characterized by a lobed, pal nucleus with diffuse chromatin, and secondary spermatogonia are derived from proliferating prirnary spermatogonia. Secondary spermatogonia usuaiiy occur in clus-

ters (= germina1 cyst or nest) of celis of similar size and appearance that develop at a uniform rate (Fawcett et ai. 1959, X. laevis; Rastogi et ai. 1983, R. esculenta). Meiosis usu-allv starts in late metamorvhosis with the transformation of secondary spermatogonia into primary spermatocytes. Leptotene spermatocytes occasionally observed in larval gonads (Iwasawa and Kobayashi 1976, R. nigromaculata; Kalt 1973a and Kalt and Gall1974,X. laevis) have large nuclei with siightly condensed chromatin. Zygotene spermatocytes are characterized by a prominent Golgi complex, a few flattened vesicles, and partiai synapsis and condensation of sister chromatids (= bouquet configuration). Synapsis is completed in the pachytene stage. Germ plasm disappears in pachytene spermatocytes (J. B. Kerr and Dixon 1974,X. l m i ) , and chromatin bivaients become separated in diplotene spermatocytes. The Woffian and Mderian duas of anurans transport gametes and, especially in males, nitrogenous wastes andrater. Therefore, the anatomical structure of the genital ducts reflects the functions of water economy, egg protection, and nurturing of developing young (M. H. Wake 1979). The Woffian duas arise from the anterior rezion of each kidnev and converge irnmediately behind the l&heys as they prOceed to the cloaca. Woffian duas persist in both sexes to develop initially as primary nephric ducts. In females they retain their excretory function. In males they also serve as urogenital ducts with remarkable modifications, especially near the cloaca where a seminal vesicle develops in some species (Amer 1972, Bufo r&aYis; Beigl 1989, Bmbina vaYieBata, Bufo bufo, Disco~lossuspictus, Rana esculenta; Bhaand duri 1953; Bhaduri and Basu 1957). In young female anurans, the Muilerian duas or oviducts usudy arise at metamorphosis from tissue near the pronephric funnel (Richter 1992, R. esculenta, Gosner 43). They are attached to the dorsal body w d , extend posterolaterally to the Wolffian ducts, and connect to the cloaca. In larvae. M d lerian duas are s&aight, h n - w d e d tubules, and in dults, they are extraordinarily coiled, especidy during the breeding season. Each oviduct includes an ostium, a short, straight pars recta, and a long, coiled pars convoluta; the posterior sections are eniarged as the ovisacs (Amer 1972, Bufo regularis; Bhaduri 1953; and Bhaduri and Basu 1957; P. Horton 1983. Rbeobatrachus silus). In males. Mullerian duas either degenerate or remain vestigial and apparently functioniess. Bidder's organ, present in male and most female bufonids (Amer 1968-1969, B. regulans; Bhaduri 1953; Di Grande 1968, 1987; and Di Grande and Marescalchi 1987, Bufo bufo; Tanimura and Iwasawa 1986, 1987, Bufojaponuusformosus; Zaccanti and Vigenti 1980, E. bufo), is formed from the anterior mesogoniaiportion of the g e ~ t aridge. Because l neither corticomeduilary structure nor primitive gonial cavities develop, Bidder's organ, which retains the potentiaiity of an ovary in adults of both sexes, consists of a cortical layer of flattened ceiis, peripherdy located gonial celis, a few 60cytes, and large auxocyte-hke ceils that partiaiiy degenerate postmetarnorphicaiiy (figs. 5.30J-N). Gonial celis and oocytes increase in number at metamorphic climax. The num-

ANATOMY

143

ber and mitotic index of these celis is significantly hgher in quently pPGCs bear a single (15-30 Fm long) pseudopofemales (Tanirnura and Iwasawa 1987, B.j.fomsus). The fat diurn at one end and a short ( 3 4 Fm long) celular process body, derived from the progonium, is characterized by pro- at the other extremity. Both are missing in PGCs situated at liferating somatic cells that form fingerlike projections the genital ridge. The pPGCs show membrane adenylattached to the gonads. No data are available on the function cyclase activity (Delbos et al. 1981, R. esculenta andX. W ) of fat body in larvae, but in the adults, fat body mass and and are separated from the smali, polygonal somatic endogonad mass are inversely correlated (Chieffi et al. 1980, R. dermal celis by a large intercellular space (Ikenishi 1982, X. esculenta; Jargensen 1986 and Jargensen et al. 1979, B. bufo; h s ) . The cells are rounded and contain many large yolk Pancharzma and Saidapur 1985 and Saidapur 1983, 1986, platelets, abundant amorphous lipid droplets, cortical granEuph(yctis qanuphlyctis), and fat body excision may (Cheffi ules, and germ plasrn that is characterized by a mass of large, et al. 1975, 1980, R. esculenta and Kasinathan et al. 1978, electron-dense mitochondria (Cwvwska 1972; Ikenishi et Euph(yctishexadactyllw) or may not (Kanamadi and Saidapur ai. 1974; Ikenish and Kotani 1975; Kait 1973a; and L. D. 1988, E. qanuphlycts) result in testicular regression and sper- Smith and Wilhams 1979, X. iuevis; Mohawald and Hennen matogenesis. In females, the fat body sustains vitellogenesis 1971 and M. A. Williams and Smith 1971, R. pipiens). The round nucleus is, in contrast to the endodermal somatic nu(Prasadmurthy and Saidapur 1987, E. qanoph(yetis). cleus, totipotent (L. D. Smith 1965, R. pipiens and Wylie Reproductive System: Devebment et ai. 1985, X. lmir) and contains a prominent nucleolus. Gonadal development starts with the accumulation of germ Particles of P-glycogen supply the energy resource for migraplasm at the vegetal pole of the fertdized egg. During cleav- tion (Delbos et al. 1981, R. esculenta andX. k s ) . J. B. Kerr age the germ plasm is distributed among a few cells that be- and Dixon (1974) suggested that germ plasm activated in come progenitors of the germ cell line (Rournoure 1937 and germ celis between gastruiation and meiosis controls migraBournoure et 4 . 1954, Rana temporaria; Buehr and Blackler tion, mitosis, and meiosis. 1970 and K. E. Dixon 1981, Xenopus k; Bernardino Di During migration, pPGCs show three main types of 1961; R. pipiens; Gipouloux 1962, 1975, DiscoglossuspiEtus, Merentiation: polarized pseudopodial projections conPhrynobatraEhus natalensis, R. pipiens, and R. temporaria; taining microfilaments, accumulation of extraceliular matrix L. D. Smith 1965, 1966; R. pipiens; Wakahara 1978; Whit- around these projections, and junctions and desmosomes ington and Dixon 1975,X. h s ; reviews by Beams and Kes- between the pPGCs and the substrate cells (Delbos et al. se1 1974 and Eddy 1975). 1981, R. esculenta andX. h Delbos et al. 1982, B. bufo ; Presumptive primordial germ cells (= pPGCs), giving and R. dalmatina; Heasman and Wylie 1978; Kamimura et rise to a11 gametes, subsequently undergo a smaii number of al. 1980; Wylie and Heasman 1976a; and Wylie and Roos cloning divisions during the migration to the genital ridge 1976,X. laevis). Migration of the pPGCs from the endoderm (Whitington and Dixon 1975, X. k ) . After gastruiation, to the genital ridge likely is achieved by passive segregation these celis (= pPGCs) migrate from the embryonic gut to of the pPGCs from other endodermal cells, by the fusion the'root of the dorsal mesentery (Lopez 1989, Bombina ori- of lateral mesodermal plates to form the dorsal mesentery entalis, Gosner 22) and then laterally across the dorsal ab- (Giorgi 1974 and Vannini and Giorgi 1969, B. bufo), and dominal wall to the gonadal ridge (figs. 5 . 3 0 4 B; Iwasawa by the attraction exerted by the genital ridge on germ cells Hyla and Yamaguchi 1984; Kamirnura et al. 1976; and Zst and (Gipouloux 1970, B. bufo, Dis~~lossusputus, arbmea, R Dixon 1977, X. laeais, Nieuwkoop/Faber 41-46; Lopez dalmatina, R. esculenta, and R. temporaria and Gipouloux 1989, Bombina mientalis, Gosner 23) where they Merenti- and Girard 1986). The basement membrane of the epithelial ate into primordial germ cells (= PGCs). Between the early cells that line the migratory route of the pPGCs may be an gastruia and Merentiation from the endoderm, pPGCs di- additional factor. T h s is believed to regulate the movement vide twice (Dziadek and Dixon 1977 and Whitington and of pPGCs from endoderm to mesentery by restricting those Dixon 1975, X. b i s ) . They synthesize DNA between ceils to the lateral plate (Heasman et al. 1985, X. laevis). The Nieuwkoop/Faber 10-33 and RNA from the gastruia to pPGCs also may be guided by diffusible substances released early tail-bud-stage (Dziadek and Dixon 1975, 1977, X. from the chordamesoderm (Giorgi 1974, B. bufo; Gipoub i s ) . The transcription of hnRNA, rnRNA, and DNA is loux 1964, 1970, B. bufo, D. pietus, H. arbmea, R. dalmatim, suppressed during Nieuwkoop/Faber 4 0 4 8 (Wakahara R esculenta, and R. temporaria) and neural crest cells (Gipou1982). The cells remain mitoticaiiy quiescent between loux and Girard 1986, B. bufo). Diffusion of cAMP from the Nieuwkoop/Faber 40-48 afier they enter the genital ridge; chordamesoderm is distributed along a concentration gradia second proliferative phase starts at about stage 48-52 (Ijiri ent that attracts pPGCs on their migratory pathway (Delbos and Egami 1975; Zust and Dixon 1977; Wakahara 1982). et al. 1980,1981, R. esculenta andX. favis; Gipouloux et al. The pPGCs, averaging 30 Fm in diameter, display polymor- 1978, X. laeais; Ratiba et al. 1981, R. dalmatina and X. phsm during migration from their endodermal position to laevis). The cells move actively in a sequence involving elonthe genital crest (Kamimura et al. 1980, X. W s ) . The ultra- gation coupled with the extrusion of dopodia and broad structure is similar in aii anuran species examined thus far blunt cell processes, waves of contraction, and ultimately by (Delbos et al. 1982, Bufo bufo and R. dalmatina; Ikenishi retraction of the trailuig end of the cell (Heasman et al. 1980 and Wylie and Heasman 1976a, b, X. laeais). Fre- 1977; Heasman and Wylie 1978; and Wylie et al. 1979,

144

BRUNO VIERTEL AND SUSANNE RICHTER

X. h s ) . They can adhere to mesentery celis via fibronectin and actin Haments produced by mesentery ceiis (Heasman et al. 1981 and Wylie et ai. 1979, X. M s ) . Concomitantiy with migration of the pPGCs, movement of peritoneai celis lateral to the mesentery to the genninai epithelium leads to the formation of paired, longitudinal, germinal ridges (figs. 5.30A and 5.31A; Lopez 1989, Bombina urientdis, Gosner 24; Ogielska and Wagner 1990, R. ridibundu, Gosner 25; Wylie and Heasman 1976a, b; Wylie et al. 1979, X. IaeviS, Nieuwkoop/Faber 45-50). The ridge can be divided into the anterior progonium, which gives rise to the fat body; the mesogonium, which forms the definitive gonads; and the posterior epigonium. The genital ridge soon acquires a gonadal medulla (figs. 5.30B-C and 5.31A-B) by the proliferation and displacement of genninai epitheliai ceiis within the primordial gonad (Franchi et al. 1962; Iwasawa and Yamaguchi 1984, X. h s ; MerchantLanos 1978 and Merchant-Larios and Vdlalpando 1981, R. pipiens andX. L&; Tanimura and Iwasawa 1988,1991, R. nigromaculata, 1989, Rbacophms arbmeus). The onset and rate of development of the gonads is uneven and depends on the mode of differentiation (Iwasawa and Yamaguchi 1984,Xenupuskaevis; Iwasawa and Kobayashi 1976; Kobayashi 1975; and Tanimura and Iwasawa 1988, 1991, Rana nigromaculuta; Lamotte and Xavier 1973 and Lamotte et ai. 1973, Nimbaphlynuides o~identalis; Lopez 1989, Bombina urientdis; Nodzenski et ai. 1989, Leptobrachium montanum, L e p t o k ~racilis, and Meguphlys nasuta; Ogielska and Wagner 1990,R. ridibundu; Rengel and Pisan 1981, Phyllomedusa sau~agii; Takahashi 1971, R. moosa; Tanimura and Iwasawa 1986, 1987, Bufo j a p o n i c a ~ s u s , 1989, Rbacupboms arbmeus). Sex differentiation essentidy depends on a comcomeddaty interaaion mediated by elaboration and release of sex induaor substances (Burns 1955; Lofts 1974; Merchant-Larios 1978). Neither the genetic constitution nor the inductive role of germ celis duences sex determination (Di Grande 1968, 1987, and Di Grande and Marescaichi 1987, B. bufo; Shirane 1970, 1972, 1982, R. japonica and R. pmosa). Other determining faaors probably include somatic sex genes (Blackler 1970, X. h s ) , sex hormones (Burns 1955; Lofts 1974; Merchant-Larios 1978; Shirane 1986, R. japonica and R. n~rmaculuta), and H-Y antigens (Engel and Schmid 1981, B. buji and X. M s ; Wachtel et ai. 1975, R. pipiens and X. hmis). InZnupus starting at Nieuwkoop/Faber 50 (Lopez 1989, B&im mientalis, Gosner 31; Ogielska and Wagner 1990, R. ridhnda, Gosner 27), ovarian development (figs. 5.30-F) obviously is controiied by estrogen and depends on age and not on metamorphic processes (Chang and Hsu 1987, R. catesbeiana).Oogonia, presumably the result of rnitotic divisions of PGCs, located within the corticai area are arranged in oogonial nests surrounded by foilicular celis (figs. 5.31H-I). The PGCs in the meddary region degenerate. The comcal epithelium and the central medulla are separated by an aceliular coiiagenous layer or basai lamina (fig. 5.31G; Merchant-Larios 1978 and Tanimura and Iwasawa 1988, R. n&romacuiata). Reduction of meddaty tissue in Xenupus begins at Nieuwkoop/Faber 52 (Iwasawa and Yama-

guchi 1984) and Gosner 28 in R. ridibunda (Ogielska and Wagner 1990). A series of cavities fuse to form the ovarian sac (figs. 5.31H-I), which remains incompletely developed at metamorphic clirnax. The medullaty tissue becomes restricted to the lining of the cavities and to the rudunentary rete ovarii (K. L. Duke 1978). In Xenupus, oogonia enter meiotic prophase at NieuwkoopFaber 55 (fig. 5.311; AlMukthar and Webb 1971,Xenopw kaevis; Lopez 1989, Bombtna urientdis; Ogielska and Wagner 1990, R. ridibunda). At metamorphosis most anuran ovaries contain oogonia and primary diplotene oocytes (Al-Mukthar and Webb 1971, X. kaevis; Lopez 1989, B. orientalis; Ogielska and Wagner 1990, R. ridibundu). After metamorphosis, oogenesis and viteiiogenesis (R. A. Wdace 1978; Wassermann and Smith 1978), and foiliculogenesis (Guraya 1978) continue in a speciestypical manner throughout adult reproductive life (Echeverra 1988, Bufo arenamm; Wagner and Ogielska 1990, R. ridibundu; Van Gansen 1986). In most temperate species the reproductive period of an individual lasts severai years. Testis formation (figs. 5.30G-I) begins with the incorporation of germ ceiis into the proliferating meddaty tissue (Iwasawa and Yamaguchi 1984, X. MS Nieuwkoop/Faber 50-58; Lopez 1989, Bombina orientalis, Gosner 31). Spermatogenesis occurs in clusters enclosed by foiiicular celis. In most anuran larvae, spermatogenesis begins late in metamorphosis. The duration of spermatogenesis in frogiets depends on the duration of sexual maturation. Seminiferuus tubules surrounding the genninai cysts are usudy formed from the meddaty tissue (Iwasawa and Yamaguchi 1984, X. W; Lofts 1974; Lopez 1989, B. orientaltr). In R. nigomacuiata (figs. 5.31E-F; Tanimura and Iwasawa 1988) and Rbacophoms arboreus (Takasu and Iwasawa 1983; Tanimura and Iwasawa 1989), seminiferous tubules develop from the inner region of the cortex. In Znopls, formation of se&erous tubules starts at Nieuwkoop/Faber 58 and in B. mientalis at Gosner 44 (Lopez 1989). It is completed after metamorphosis. Meddary cords near the gonadai hilus Merentiate into the rete testis (fig. 5.31F). More anteriorly the cords give rise to the vas efferens which, in most anurans, connea with the anterior opisthonephric tubules to acquire open connections with the Woifiian ducts (figs. 5.31E-F). In some anurans, such as Dism~lo~~uspictus, efferent tuthe bules join the Woffian ducts directly (Beigl 1989). The follide ceiis differentiate postmetamorphicdy into steroidproducing Sertoli ceiis (Bernardini et al. 1990, X. hmis and Guraya 1972, R. pipiens; review by Lofts 1974); interstitiai cek develop into steroid-producing Leydig celis. The Wolfian d u m initidy develop as primary duas draining the pro- and opisthonephros and, after connecting with the efferent tubules, aiso serve as urogenitai duas in males. Their posterior portions are either modified as serninal vesicles postmetamorphicdy or remain simple and undifferentiated (Amer 1972; Beigl 1989; Bhaduri 1953; Bhaduri and Basu 1957). The Mderian d u m develop independentiy of the Woffian duas. Drfferentiation starts with the formation of the ostium abdominaie developed by the closing of a peritoneai groove aiong an area of thickened epithelium at the site of the anterior pronephric fnnel (Richter

ANATOMY

145

1992, R. esculenta, Gosner 43). The lips of the groove close neurai crest origin (Accordi and Grassi Milano 1990),while and the tube then grows posteriorly &til it jo& the cloaca. the gonadai primordia are of mesodermal origin. The Mderian duas are present in both males and femaies, although they typicaiiy degenerate in males (Bhaduri 1953; Pituitary Giund Bhaduri and Basu 1957: M. H. Wake 19791. Embrvonic The stomodeal-hypophysed anlage is visible at NieuwkoopJ differentiation of both genital duas is independent of hor- Faber 20 in Xenupus luevis. The adenohypophyseai anlage monai influences, but as development proceeds they become segregates at NieuwkoopJFaber 28, and by stage 41 the sensitive to steroids secreted by the gonadal primordium. largest mass of the compaa tissue lies anterior to the noEstroeenic steroids are resoonsib~e for-the diierentiation of tochord. Differentiation of the pars intermedia begins at " the oviduct and urogenital papdla and for jelly secretion NieuwkoopJFaber47. Tuberai parts extend longitudindy in (Gunasingh et ai. 1982, Euphyctis h e x ~ l u s ) . NieuwkoopJFaber 5 1, and the entire hypophysis or pituitary gland (neurohypophysis and adenohypophysis) is well deBidder's Ogan and Fat Body veloped at stage 53 (Atwell 1918-1919; Eakin and Bush Bidder's organ develops in maie and female bufonids from 1957; Nieuwkoop and Faber 1956). The dislocation of the pars intermedia, pars nervosa, and the mesogoniai seaion anterior to the definitive gonad. Differentiation starts soon d e r hatching with the prolifera- mediai eminence from late I orometamorohosis to late metaI tion of germ cells scattered among a few medda-llke cells morphic climax is similar in Bufo americanus, Ranap@iens, (figs. 5.30J-N; Tanirnura and Iwasawa 1986, B. j. fomsus; and R. sylvatica. The horizontal axis from the mediai emisee Petrini and Zaccanti 1998 for effects of various hor- nence to the pars nervosa assumes a perpendicular or transmones and inhibitors).The germ cells eniarge and differen- verse orientation. and the ~ a r s intermedia and adenohvtiate into goniai and auxocyte-like ceh both enveloped by a pophysis are drawn dorsdy. By early metamorphic climax, layer of foiiidar cells. During larvai development no sexual the adenohypophysis loses direa contact with the media1 Merentiation in the number of Eerm cells is found. Post- eminence and pars intermedia (Etkin 1968; Hanke 1976). metamorphicdy the number of oocytes and secondary Three types-of pituitary cells occur in larvae (Batten and gonia are significantly higher in females (Tanimura and Iwa- Ingleton 1987; Hanke 1976; Van Oordt 1974; figs. sawa 1987, B. j fmzosus). . 5.32A-B) : thyrotrops (NieuwkoopJFaber 48), a smaiier The fat bo$ composed oniy of somatic cells is derived Type I basophii (stage 52), and adrenocorticotrops or from the progoniai, germ cell-freepart of the germinai ridge. ACTH cells. The A1 acidouhiis secrete prolactin. and A2 In Xenopus, fat deposition starts at NieuwkoopJFaber 51 acidophils that secrete somatotropin are deteaable d e r (Ogielska and Wagner 1990, R. ridibunda, Gosner 29). Fat metamorphic climax. During metamorphic climax (Nieuwbodies are largely composed of adipose cells at the beginning koop/~aber they are the-most numrous cell type in the 57) of hind limb development (Lopez 1989,B. orientalis, Gosner ventral half of the pars anterior. 30; Nieuwkoop and Faber 1967, X. luevis; Tanimura and nyroid Giund Iwasawa 1986, B. j.fomsus). The thyroid begins to differentiate at SedraJMichael 27-28 (Gosner 17-18) as a ventral divertidum near the second Endocrine System visceral arch (Etkin 1936).At SedraJMichael35-36 (Gosner Regulation of the development, growth, and metamorpho- 20) the groove aiong the anterior anlage has disappeared. At sis of tadpoles by the endocrine glands of endodermai or me- SedraJMichael42(Gosner 24) the anlage separates from the sodermai origin is a fascinating aspea of larvai biology. Ex- buccopharynx, contaas the basibranchial, and consists of cept for the iodmated amino acid derivatives of the thyroid, two lobes bounded by a fibrous capsule and conneaed by endodermal glands produce polypeptides, whiie mesoder- an anterior, saddle-shaped structure. At SedraJMichael 46 mal glands synthesize steroid hormones. Endocrine glands (Gosner 26-27) the lobes are separated, colloid-containing of neurai or neural crest origin consist of photoreceptor cells fohcles are differentiated, and prefohcular cells are present. (e.g., dorsai pineai complex, ventral infundibulum, neurohy- At SedraJMichael 50 (Gosner 29) each lobe divides and pophysis anlage of pituitary gland) that evaginate from the comes into close contaa with the hypobranchialplate. Vacudiencephalon and produce oligopeptides, polypeptides, or, oles appear in the foiiicular cells, and a homogeneous basoin the case of the pineai complex, a derivative of amino acid philic colloid Ms the foiiicular lurnen. Beginning with SedraJMichae1 58 (Gosner 41) the lobes are about 1.5 times and polypeptides (chap. 6). The dorsomediai adenohypophysisanlage, the ventrome- that at SedraJMichael 57, the colurnnar cells are tali, and diai thyroid anlage, and the endocrine glands derived from there are large chromatophobe droplets (Michael and A1 Adthe branchial arches d evaginate from the buccopharyngeal hami 1974). area. The endocrine pancreas originates from the midgut anIn Xenupus luavi (NieuwkoopJFaber 33-34), the thyroid lage. It is not clear whether the endocrine pancreas origi- anlage is situated medidy near the first viscerai pouch. At nates from an ancestrdy difise gastrointestinai endocrine NieuwkoopJFaber 35-36 it grows posteroventrdy as a thin system, d&se and already specialized endocrine cells of the strand of cells under the skin. The anlage thickens, spiits into intestinal w d , or the neurai crest (Epple and Lewis 1973; two lobes (NieuwkoopJFaber 37-38), comes to iie near the Wessels 1968). The adrenai glands are of mesodermal and branching point of the truncus arteriosus at stage 39, and is
V

146

BRUNO VIERTEL AND SUSANNE RICHTER

Fig. 5.32. (A-B) Adenohypophysis and (C) interrenal tissue and adrena1 cortex o f X k p u latmc larva. Abbreviations: ACTH = ceiis that ye produce adrenocomcompic hormone, T p I11 basophils, CHC = meduiiary chromaffin ceii, STC = comcular steroidogenicceii, S H = somatotrophic ceii (produces growth hormone), and TSH = ceiis that produce thyroid stimulating hormone, T p I basophils. Courtesy of ye Hanke.

connected with the epithelium of the buccopharyngeal floor by a soiid ceii strand (ductus thyreoglossus). This connection disappears by Nieuwkoop/Faber 40, and the thyroid forms two longitudinal lobes. At Nieuwkoop/Faber 41 the definitive thyroid gland appears as a paired organ lying ventral to each ventral velum and, by stage 43, on both sides of the hyoid arch. The ductus thyreoglossus disappears by Nieuwkoop/Faber 41. Foiiicles with colioid vesicles appear at Nieuwkoop/Faber 46-47 and are weli developed and lined by mucous and cuboidal celis by stage 48-50; foilicle mass and number increase. and the interfollicular connective tissue expands at about the same time. During earbj prometamorphosis, the flat foiiicle epithelium becomes cuboidal and then colurnnar (Jayatilaka 1978; Nieuwkoop and Faber 1956). Nanba (1972) o u t h e d the development of the thyroid in Ranujaponku, and Hayes (1995) and Hayes and Wu (1995) discussed the interdependence of corticosterone and thyroid hormones in the larvae of Bufi bmem.

EndomCrCne Pancrem The islets of h g e r h a n s develop from celis lining branching ducts or capiiiaries that extend through the exocriie pancreas tissue. The precursors of a-celis (= glucagon or A celis) are not visible before Gaien/Houiiion 27-28 in D s ~ o s u imloss pia% (Beaumont 1970) but are visible at Shumway 24 in Ranappiens (Kaung 1981). The irst a-granules are synthesized by the Golgi apparatus. At Gaiiien/Houllion 31 the a-ceiis are differentiated (Pouyet and Beaumont 1975,Alyies obst&m). The p-celis (= insuiin or B ceiis) are differentiated during h b - b u d stage. At Taylor/Koliros I1 in R.ppiensJ p-celis and insuiin are already present and increase until Taylor/Koiiros XVIII and XIX, respectively (Kaung 1983). In Xenqiw W the endocrine tissue starts to develop at Nieuwkoop/Faber 42 (Fox 1984), although islet ceiis are not differentiated before stage 48. Fox (1984) suggested that pcelis at Nieuwkoop/Faber 42 in X. laepis may be undifferentiated A ceiis or centroacinar celis of the adult pancreas. The endocrine pancreas of Xenqiw metamorphs (after Nieuwkoop/Faber 61) consists of acini surrounded by a- and Pcells. After metarnorphosis the endocrine celis occupy about 3% of the total pancreatic tissue, and there is a drastic increase in insulin production in prometamorphic and metamorphic ciimax larvae (from Taylor/Koilros XI; Buchan 1985; Frye 1964). After an initial decrease of p-cells during metamorphic climax (Taylor/KollrosXM-XX), stasis in Taylor/Koiiros XXI-XXV is foiiowed by another decrease during later stages (Kaung 1983, Ranupijkm). During prometamorphosis and metamorphic climax, pancreatic weight and insuliniike imrnunoreactivity decreases (Hulsebus and Farrar 1985, R. catesbeianu).

ANATOMY

147

middle of metamorphic clunax. Aidosterone concentrations Adrenal Gland peak at the start and end of metamorphosis (Jolivet-Jaudet Steroidogenic cells of the adrenai glands are of nlesodermai and Leloup-Hatey 1984; Leloup-Hatey et ai. 1990). origin, and the chromafJin or catecholamine cells are modified neurons derived from the neurai crest (Accordi and Grassi Milano 1990). Mesodermai cells proliferating from Summary the dorsai root of the dorsai mesentery at the same leve1 as Compared with other amphibians, larvai and adult anurans the coeliacomesenteric artery extend to the opisthonephric are quite different, and these diflerences provide a unique anlage (Nieuwkoop/Faber 4 2 4 3 , X. k s ) . The interrenal opportunity to study a free-living and actively feedmg develprimordi~un cells migrate dorsdy to the ventrai side of the opmentai stage of a tetrapod. They develop larvai organs in aorta and f the space posterior to the mesenteric artery be- the sense of precursors or transitory stages of dserent gradU tween the two posterior carduiai veins. A niediai peduncle uations. These precursor organs (e.g., heart, artery, vein sysforms ventrav in contact with the mesenteric dorsai root. tem, and lymphatic system) are not simply organs in waitWhen the two posterior cardinal veins unite to form the in- ing, they are highly adapted to aquatic life and are significant terrenal vein, the mediai peduncle is split and the interrenal for larvai survival. Anatomicai characters of the heart and aranlage is compressed between the interrenal vein and the teries are essential for a water and air breather. The lymphatic ventral side of the aorta (Nieuwkoop/Faber 44-45). Be- system, concentrated in the buccopharyngeai region, is the tween Nieuwkoop/Faber 46-50, two cell masses lie ventrai buTer of the filter-feeding larva against microorganisms to the aorta and dorsai to the interrenai vein aiong the me- ingested aiong with the irrigation current. Some of these diai face of the opisthonephric anlage. At ~ieuwkoo~/Faberorgans and the endocrine glands are resorbed, rebuilt, and 50 the precursors of chromath cells migrating from the changed at metamorphosis. Branchiogene endocrine glands sympathetic gangiia region penetrate the interrenal anlage. do not reach their definite position before metamorphosis. During Nieuwkoop/Faber 57-66 the adrenai tissue reaches The skin is a respiratory, sensory, and mucus-producing its definitive position on the medioventral face of the organ, and few characters other than the lateral line are opisthonephric kidneys (Accordi and Grassi Milano 1990; changed during metamorphosis. As is typical for precursor Grassi Milano and Accordi 1986; Nieuwkoop and Faber organs, only parts or certain anatomical features are espe1956). The position of the adrenai glands of Bufo bufo at cidy significant for larvai life and thus are changed during Rossi XI changes sirnilarly during metamorphic clima (Ac- metamorphosis. cordi and Grassi Milano 1977). In contrast, the buccai structures, filter apparatus, and larPrecursors of chromatKn cells can be detected in low den- vai stomach (manicotto glandulare, an adaptation for sussities in Bufo bufo at the end of embryonic development pension feeding) are alrnost totdy resorbed at metamorphic (Rossi XXV). The number of chromatKn and corticai (ste- ciimax. These organs are significant for speciai adaptations roidogenic) cells increases at the beginning of premetanlor- to aquatic life (e.g., suspension feeding) and are larvai organs phosis (Rossi 1-111) (fig. 5.32C). At the beginning of pre- senso stricto even if a clear distinction between precursor metamorphosis, chromafJin cells are still in an embryonic organs and typicai larvai organs is not aiways possible. Aortic state (Rossi N) are diflerentiated at the end of premeta- arches are typicai adaptations to aquatic larvai life that are but morphosis (Rossi VI; Accordi and Grassi Milano 1977). In reduced during metamorphosis, and some parts become the premetanlorphic larvae, polygonal steroidogenic cells form primary arteries of the lungs and heart. The larvai excretory the interrenal cords and enclose blood sinuses, and chro- system consists of transitory and persistent parts that d o w ma& cells form sma disparate groups (Accordi and Grassi storage and transport of the gonadai products of the adult. Milano 1990; Chester-Jones 1987; Grassi Milano and Ac- The development of the kidneys begins with the diflerentiacordi 1986). Two types of c h r o m f i cells have developed tion of the pronephros (larvai organ, sensu stricto), which is at the end of premetamorphosis. At the beginning of meta- a transitionai larvai excretory organ. During larvai life, the morphic climax (Rossi XII) the steroidogenic ceiis form thin urinary function s M s to the opisthonephric kidney of the cords that lie pardel to the surface of the gland and sur- adult. The ducts of the opisthonephric kidneys are associated rounded by connective tissue. intimately with the reproductive system, and gonads deAcidophiiic Stilling's or surnmer cells (Stilhg 1898), velop from adjacent tissues; after the excretory ducts have probably related to steroidogenesis (Accordi and Grassi Mi- developed, the reproductive system usudy taps into them. lano 1990; Chester-Jones 1987), are found only in ranids and rhacophorids after metamorphosis and vary Gith season (Accordi and Cianfoni 1981; Grassi Milano et ai. 1979; ACKNOWLEDGMENTS Grassi Milano and Accordi 1983, Rana escubnta complex, The parts of this chapter written by B. Viertel were supR. montezumae, and Polypedates bucomystax). ported by the Ernst-Kalhof Foundation. He thanks C. C. Corticosteroids are synthesized by the steroidogenic cells Bangnara (Utrecht), J. Bengassat (Jerusalem), H. Fox ( h n present at Nieuwkoop/Faber 54 in Xenopus laevis (Leloup- don), G. Frank (Wien), W. Hanke (Karlsruhe),J. D. Horton Hatev et al. 1990). Corticosterone concentration varies from (New Orleans), J. Meseguer (Murica), G. Pekny (Wieti), abou; 15 ng rn-i during premetamorphosis and at the end J. B. Turpen (Omaha), and M. Ueck (Giessen) for h d l y of metamorphic ciimax with a peak of 50 ng r n - I at the supplying iilustrations. He aiso is indebted to B. Grilhtsch

148

BRUNO VIERTEL AND SUSANNE RICHTER

and S. Richter for vaiuable bibliographcai advice. He wishes to thank I. Galbraith (Wiesbaden) for h s untiring work as a translator. He is indebted to K. Rehbinder for her drawings and M. U h a n n (Mainz) for her work in the photographc laboratory. H e is gratefui to the following publishers and copyright holders for permission to reproduce p i a r e s and tables: Faber and Faber Ltd. Publishers (London); Taylor and Francis Ltd. (London);McGraw-Hdl, Inc. (New York); Royai Society of South Africa (Cape Town); Academic Press (Orlando); Plenum Publishing Corporation (New York); Editrice Compositori Bologna (Bologna); Anne Orcet Livres Scientifiques (Poudy Sur Loire); Cambridge University Press (Cambridge); American Society of Zoologists

(Lawrence); Oxford University Press (Oxford); SpringerVerlag (Heidelberg); Alan R. Liss, Inc. (New York); Vaillant-Carmanne S. A. (Lige); Gustav Fischer Verlag (Jena); E. J. Briii (Leiden); Akademische Verlagsgesellschaft Geest und Portig K. G. (Leipzig); and the Zoological Society of London (London). S. Richter is gratefui to B. Viertel for his keen and continuou~ support. She is aiso gratefui to A. Tanimura, H . Iwasawa, and I. Naito for their valuable and spontaneous help in providing iiiustrations. She is indebted to M. Stachowitsch for assistance in English translations and to S. Neulinger for preparing the iiiustrations.

INTEGRATION
Nervous and Sensory Systems
Michael J. Lannoo

Introduction
The nervous system, consisting of the brain and its craniai nerves and the spinai cord and its nerves (fig. 6.1), underlies aii vertebrate behaviors. Vertebrates detect environmentai cues iitered through their variou senses and send these signais centraily via peripherai nerves. This information is processed by either relatively simple spinai reflexes or more complex brain circuitry, and commands are sent back through peripheral nerves to drive muscles and glands in some presumably adaptive manner. An adaptation is deined here as a feature that contributes to the survivai of an individual or to the survivai of its offspring (Liem and Wake 1985). For most anurans, the tadpole n e d circuitry is remodeled abruptiy at metamorphosis. Sensory systems, muscle groups, and glands are reworked during the transition from aquatic, essentiaiiy limbless, swimming larvae to, in most cases terrestrial, quadrupedal, saltatory adults. During this remarkable transition. the nervous svstem must shift its ability to process and int&rate the infoimation received and the systems being afFected between two very different types of ornanisms. This transition means that in addition to the usual and important questions of neurai structure and function, the tadpole nervous system requires a further leve1 of anaiysis; unlike that of most other vertebrates, the nervous system of the aquatic tadpole must be considered within the context of its rapid metamorphosis into the nervous system of a terrestriai frog. In this light, three questions concerning the neurobiology of tadpoles are addressed. (1) What portion of a tadpole's nervous system reflects what a tadpole actually uses? (2) Which systems must be reworked at metamorphosis? (3) What tadpole neurai tissues are developed for later use by the adult?Also, interspecific variations in the behavior and ecology of tadpoles reflected in their neuronal organization and function must be considered. The general neurobiology of anuran tadpoles is less weii known than that of either anuran embryos or adults. For example, Spemann discovered the mesodermic induction of ectoderm to neuroectoderm in anuran embrvos (Ham, burger 1988). Adult anurans were used in the discoveries of the workings of the neuromuscular junction (Birks et ai. 1960) and the d e s underlying neuronal speciicity within the visual system (e.g., Constantine-Paton et ai. 1983; Reh et ai. 1983; R. W. Sperry 1963). An excellent review of the neurobiology of amphibians, including adult structures, is provided by Wilczynski (1992). More recentiy, studies of tadpoles have provided an understanding of the growth of the cerebeiium (e.g., Hauser et ai. 1986a, b, and Uray
U
\

The most refined methods of anatornicai analysis cannot reveai the things that are of greatest significance for understanding the nervous system.
Our primar- interest is in the behavior of the living

body, and we study brains because these organs are the chief instruments which regulate behavior.
Herrick 1948:5

MICHAEL J. LANNOO

and summarize the organization of the tadpole nervous system with a brief consideration of the changes that it undergoes during metarnorphosis.

An Overview
The nervous system can be divided into (1) peripheral (PNS) versus central (CNS) systems, with the CNS divided further into the brain and spinal cord (figs. 6.2-6.3); (2) somatic (external or voluntary) and visceral (intemal or involuntary) systems; (3) sensory, integration, and motor systems; and (4) neurons and their more numerou supportive glial cells. Neural processing begins with information acquired through the sense organs. External sensory systems detect the types and intensities of energy present in the environment and translate this information into neuronal signals. Sensory systems act as flters for the reception of types of stimuli with adaptive value while presumably eliminating those that are less important. Internal sensory systems monitor the visceral environment and insure that the animal's physiological mechanisms are operating within their limits of tolerante. The brain. which receives sensorv information and transmits motor commands either through cranial (CN) or spinal (SN) nerves (figs. 6.2-6.3), is divided into a forebrain (prosencephalon), midbrain (mesencephalon), and hindbrain (rhombencephalon; fig. 6.4; also see D. Black 1917); The forebrain is divided into an anterior telencephalon and a posterior diencephalon (fig. 6.4), which function in receiving oifactory and terminal nerve information and integrating sensory inputs and motor outputs. One region of the diencephalon, the hypothalamus, regulates hormone produaion and involuntary autonomic responses. The midbrain consists of a dorsd, sensory tectum and a ventral, motor tegmentum (fig. 6.4). The midbrain integrates first-order (primary) visual inputs with higher order inputs from the lateral line, auditory, and vestibular systems. The hindbrain includes the cerebellum and meduiia oblongata (fig. 6.4), and the combination of the mid- and hndbrains forms the brain Fig. 6.1. Dorsal view of the brain and spinal cord i a generalized tad- stem. The area receives primary inputs from aii of the senn sory systems except the terminal nerve and oifaaion, includpole. Cranial and spinal nerves are shown as if they have been severed. ing vision, acousticolateral, gustatory, and general sensations, and sends motor commands to the muscles and glands 1985), the formation of and variation in the spinal cord (K. of the head, cervical region, and viscera. The reticular frmaNishkawa and Wassersug 1988, 1989; Nordlander 1984, tion of the brain stem is involved with integrating sensory 1986), and sexual differences w i t h the nervous system inputs and with overaii arousal. The cerebelium mediates motor learning and coordination. (Gorlick and Kellev 1987). The spinal cord is both a relay center and controller of The information contained in this chapter largely results fiom research on species of Rama and Xinopus with most reflexive behaviors and receives sensory information from other information based on species of Bufo and Hyh. The specialized touch, temperature, pain, and postural receptor detaiis of the organization of higher brain centers ar; known organs through its dorsal nerve roots. The spinal cord either only for adults and are presented as a starting point for un- diiectly provides the motor output to the muscles of the derstanding the brain of tadpoles. To avoid the pitfaiis of trunk, taii, and developing limbs (reflex behavior; see Shirigenerahing about "the tadpole:' I point out known phylo- aev and Shupliakov 1986 for the anatomy of these connecgenetic and ecological variations. I introduce the nervous tions) through its ventral roots or first relais these sensations system and the organization of the tadpole brain and then to the brain and then transmits the brain cornrnands back to examine the organization and function of the spinal cord and the muscles. W~thin brain stem and gray matter of the the brain, discuss the variou sensory systems used by tadpoles, spinal cord, sensory centers tend to be aoisal, autonomic
i

INTEGRATION

151

thetic trunks, each a series of gangiia that extend from anterior to posterior and paraiiel to the spinai cord (fig. 6.6; Taxi 1976). sympathetic ganglia receive pregangiionic fibers that trave1 through cornmunicating rami from the spinai nerves (fig. 6.7; W. M. Davis and Numemacher 1974 and Schlosser and Roth 1995). About one quarter of the fibers in a spinal nerve are pregangiionic parasympathetics (fig. 6.8; Vance et ai. 1975). Pregangiionic fibers enter the sympathetic trunk and may course either anteriorly or posteriorly

IV

VI1 nALL VI11

nPLL
Fig. 6.2. Lateral (i@) and dorsai (nhht)views of a generaiized tadpole brain showing the positions of the craniai, lateral line, and selected spinai nerves. Craniai nerves (CN) : I = oifactory (CN O is induded with CN I), I1 = optic, I11 = oculornotor, IV = ti-ochlear, V = mgeminai, VII = facial, W I = acousticowstibular, M = glossopharyngeai, X = vagus, XI = accessory, and XII = hypoglossai. Lateral line (LL) nerves: nALL = anterior division and nPLL = posterior division. Spinai nerves (SN): 2 and 3 are the anteriormost spinai nerves. From Stuesse et ai. (1983) and Lowe (1987); reprinted by permission of Wdey-Liss, a subsidiary of John Wdey & Sons, Inc., and of Stuesse and Lowe.
IX X, M XII

centers are lateral and intermediate, and motor centers are ventral (fig. 6.5B). In addi2on to neurons. the vertebrate brain contains eiiai ceiis of three basic types: itrocytes, ependymai ceiis, anduoligodendrocytes. Giiai ceils are thought to provide the skeleton of the brain, to provide nutrition to neurons, and, in some specialized cases, to guide developing neurons as they migrate to their inal position. Oiigodendrocytes form the mvelin sheaths that encircle neuronai axons. The autonomic nervous system regulates involuntary body functions and insures that an organism is operating within its physiologicai iimits. The sympathetic and parasympathetic components typicaiiy act antagonisticaiiy. The sympathetic system is energy expending and considered the "fight or flight" component; it haits gut peristalsis, tightens gut sphincters, dilates pupils, and increases heart rate (Pick 1970). The parasyrnpathetic system is energy conserving and effects the opposite results. The most structuraiiy visible features of the itonomic nervous system are the p&red syrnpa-

Fig. 6.3. Dorsai view of a generaiized brain and spinai cord of a tadpole showing craniai and spinai nerves. Craniai nerves I11 and Vi and the posteriormost 10-18 pairs of spinai nerves whch innervate the tail segrnents have been omitted for clariq. Abbrniations as in fig. 6.2.

152

MICHAEL J. LANNOO

TEC

6.8). Cell bodies (gray matter) are organized centraiiy into an H-shaped pattern with bilateral dorsal and ventral horns. Dorsal horns receive inputs from the sensory dorsal roots; ventral horns contain the motor neuron pools that supply the muscles of the trunk,taii, and limbs. W i t h the ventral horns, ceiis in both the medial and lateral motor pools tend to innervate the trunk and taii muscles, whde additional cells from the lateral motor pool innervate the lirnb muscles (Ebbesson 1976; A. Roberts and Clarke 1982; Silver 1942). An intermediate gray area that contains preganglionic autonomic fibers is poorly understood in amphibians (Ebbes-

Fig. 6.4. Lateral (I$)and dorsal (right) views of the tadpole brain. Abbreviations: Forebrain: TEL = telencephalon and DI = diencephalon; rnidbrain: TEC = tectum and TEG = tegmentum; and hindbrain: CB = cerebelium and MO = m e d d a oblongata.

through two or three gangha prior to synapsing on postsynaptic neurons. Postsynaptic neurons leave the sympathetic t u k and follow the major arteries to visceral organs. Prern synaptic neurons may also synapse distal to the sympathetic truck in viscerai plexuses (figs. 6.6-6.7). For example, the solar plexus in anurans is formed by branches emanating from sympathetic ganglia 3,4, and 5 that supply the stomach and adjacent parts of the alimentary canal (see C. A. Brown et al. 1992). The sympathetic ganglia and nerves of aii vertebrates develop from neurai crest cells. Neurons of the parasympathetic system leave the central nervous system either via cranial nerves or through the sacral parasympathetic system. Parasympathetic fibers from the laryngeus ventralis branch craniai nerve CN X provide parasympathetic innervation of the smooth muscles of the lungs, heart, and stomach. O h et ai. (1989) observed parasympathetic neurons by backiiiing CNs V, VII, IX, and X branches with cobalt lysine. Parasympathetic ceii bodies form an almost continuous column through the brain stem dorsai to the corresponding motor neurons of the same craniai nerves. Heathcote and Chen (1991) detailed the development of the parasympathetic cardiac ganghon in Xenopus hetis. The sacral parasympathetic system functions in reproduction and for this reason may not be well developed in tadpoles.

SSP

Spinal Cord
The spinal cord occupies the n e d canal within the vertebral column and is round or oval in cross section (figs. 6.7-

Fig. 6.5. Cross sections of the brain stem. (A) The brain stem has an inside (ventricular) and outside (pial) surface, and the dorsal, vertical alar plates are separated from the ventral, horizontal basal plates by the suicus Limitans i the lateral ventricular w d . (B) Cranial nerve and n brain stem functional components are arranged into four longitudinal columns. (C) The ceiiuiar organization of the brain stem. Asterisk indicates position of suicus limitans, and dotred lities separate longitudinal columns. Abbreviations: AP = alar plate, BP = basal plate, P = pial, SA = somatic atferents (sensory), SAI, = stratum aibumen (major area of fiber tracts), SE = somatic efferents, SEP = stratum ependymale ceiis, SGR = Stratum griseum (major ceii body layer), SSP = stratum subependymaie ceiis, V = ventricuiar, VA = visceral atferents, and V E = viscerai efferents (motor).

INTEGRATION

153

Fig. 6.6. The sympathetic trunk showing ganglia 2-10 and the coeliac (gut) plexus of an adult &na esculenta. Ganglion 8 is split in the specimen drawn, and the ventral rami of spinal nerves 3-6 are indicated. Abbreviations: CP = coeliac plexus, Gan = ganghon, and SN = spinal nerve. Redrawn from Taxi (1976) with modifications of the original figure suggested by Dr. Taxi; reprinted by permission of SpringerVerlag q d Taxi.

from the cord progressively more anterior than they exit the neurai canal (fig. 6.9). Nerves exit the neurai canal between the vertebrae, except for SN 10 and 11 (if present), which exit via forarninae in the coccyx. The morphology of the spinai cord and spinal nerves and their relationships to tail myotomes through metamorphosis have been studied by K. Nishkawa and Wassersug (1988) for both Rana and Xenopus. Spinal nerves range from 23 to 29 pairs prirnitively (e.g., ascaphds and discoglossids), and reductions have occurred independentiy at least seven times during anuran evolution. The posterior end of the spinai cord, termed the filum terminde, in R. catesbeiana and R. pipiens (Chesler and Nicholson 1985) contains a functional neuropile supported by a large number of gliai ceiis (H. Sasaki and Mannen 1981). Tadpoles have a ventral root in the position of SN 1 that is apparentiy homologous to CN XII in other vertebrates and is designated here as such. All other spinai nerves have a dorsai root with a large ganglion (M. R Davis and Constantine-Paton 1983 and Willielm and Coggeshaii 1981) and a ventral root that fuses with the dorsai root distai to the ganglion (fig. 6.7). Peripherai to this junction, a smaii branch of the spinai nerve passes dorsaiiy to innervate the skin and muscles of the dorsai trunk. The large ventral branch of each spinai nerve innervates ventral and lateral skm muscles of the trunk and tad. The ultrastructurai development of ventral roots has been studied by Nordlander et ai. (1981). A ramus cornrnunicans containing preganglionic
Dorsal Gray Dorsal Root

son 1976). Ascending and descending axons, organized in fiber tracts (white matter), surround the gray matter. Forehand and Farel (1982a, b), Holder et ai. (1987), Nordlander (1987), Nordlander and Singer (1982a, b), Nordlander et ai. (1991), A. Roberts (1976), A. Roberts and Patton (1985), A. Roberts and Taylor (1983), and J. S. H. Taylor and Roberts (1983) described the development of spinai neurons and fiber pathways in anuran larvae. Variation in the morphology of the cord aiong its length depends on the type and number of its connections. The spi- Fig. 6.7. A schematic representation of the organization of the spinal nai cord, especiaiiy its white matter, is wider nearer the brain cord. The left side of the illustration shows dorsal and ventral homs than at thoracic or posterior levels so that the cord tapers composed of dorsal, internediate, and ventral gray matter, and the anteriorly to posteriorly (fig. 6.3). Imposed upon t h s taper- white matter (fiber tracts) is peripheral. The right side shows the types ing, the spinal cord is wider at the cervicai and lumbar levels of spinal fibers and their locations, including the sympathetic trunk to accommodate the sensory and motor neurons that form anda sympathetic plexus. Symbols (dorsal to ventral): gray dotted = the brachiai and lumbar plexuses in the limb regions (K. Nis- somatic afferents bringing general pain, pressure, temperature, and hkawa and Wassersug 1988 and Sutherland and Nunnem- touch sensation from cutaneous receptors as weil as aflerents from acher 1981, Hyh and Eleuthero~Ius). Lateral motor pool muscle spindles through the dorsal ramus; gray solid = afferent fibers neurons increase in number with development of the limbs bringing visceral feedback; black solid = preganghonic sympathetic fibers to gangha in the sympathetic trunk and visceral plexes; black in the tadpole. dashed = postganglionic fibers to visceral organs to control autonomic The spinai cord supplies each body segment with a pair of funmons; and gray dashed = somatic efferents to voluntary muscles spinai nerves (figs. 6.7 and 6.9; K. Nishikawa and Wassersug through the ventral ramus. Redrawn from Noden and De Lahunta 1988) numbered according to the vertebral leve1 at which (1985); reprinted by permission of Wdiams & W h s Co., Noden, the pair exits. Proceeding down the cord, spinai nerves arise and De Lahunta.

MICHAEL J. LANNOO

Fig. 6.8. Fiber types in the anuran spinai cord. The left side of the iustration shows dorsal sensory fibers and a dendritic tree from an axiai motor neuron from the right side; on the right there are three motor neurons that send axons ventrdy to the periphery. The ventromedial celi represents an axiai motor neuron of the mediai celi group (M), and the more laterai celis represent limb motor neurons from the lateral celi group (L). The transition between white matter and gray matter is shown by a dashed iine. From Szkely and Czh (1976) after Szkely (1976); reprinted by permission of Springer-Verlag and Szkely.

fiber that appears to terminate in the dorsai horn. The mediai division consists of a large-diameter fiber that enters the dorsal h d u s and spiits into ascending and descending coiiaterals. The ascending coiiaterai courses craniaily into the brain stem, where it continues into the straturn aibum (fig. 6.5C). Axons in this traa terminate in the hindbrain vestibular nuclei (spinovestibularfibers), the retidar formation (spinoreticdar fibers; figs. 6.11-6.12), and in the diencephaion. In anurans, ceiis from the dorsal root gangiion aiso projea diredy into the cerebeiium (Szkely et ai. 1980). In the spinai cord these fibers ascend in the ipsilaterai dorsal funiculus (Gonzalez et al. 1984; Joseph and Whitlock 1968; Van Der L i d e n and Ten Donkelaar 1987; Van Der Linden et ai. 1988), enter the cerebelium, and synapse diredy on Purlunje cells (Ebbesson 1976). In the cerebeiium these dorsai spinocerebeiiar fibers rningle with fibers of the ventrai spinocerebeliar traa (a component of the ventrai white matter). The anuran spinai cord differs from that of ail other vertebrates except turtles (Kunzle 1983) in having this direa projection from the dorsai root gangiia into the cerebeiium. The function of this projection is not known. In the ventrai spinai cord, the dendrites of motor neurons
Xenopus Rana

sympathetic fibers extends from each ventrai branch to the sympathetic nerve trunk. Most anuran tadpoles have more than 20 pairs of spinai nerves (K. Nishikawa and Wassersug 1989), which reduce to about 10 pairs after metamorphosis. In Ranu, SNs 2 and 3 join a co&eaing ramus to form the brachiai plexus (e.g., Oka et ai. 1989). Spinai nerves 7 and 8 form the lumbar plexus, and SN 7 gives off a branch (the diohypogastric) to muscles of the laterai and ventral body w d . Spinai nerves 7 and 8 may fuse with SNs 9 and 10. The organization of neurons into nerve trunks from SNs 8 and 9 to the giuteus muscle in Bufo marinus addts was studied by D. R. Brown et ai. (1989). The ventrai branches of the spinai nerves between the brachiai and lumbar plexuses give off branches to the laterai and ventrai body w d musdature and to the skin. In anurans branches of SNs 9 and 10 form a ~lexus inthat nervates the bladder, cloaca, oviduas (in addt femaies), and lymph hearts. Developmentdy, prior to the formation of the spinai ganglia, peripherai sensation is provided by large, dorsai Rohon-Beard celis (fig. 6.10; Bixby and Spitzer 1982; Hughes 1957; Spitzer and Spitzer 1975).These ceiis are first preient in the spinal cord at gastruiation (Lamborghini 1980), increase in numbers that peak in tadpoles, and are then reduced and eventuaiiy lost at metamorphosis (Decker 1976: Eichler and Porter 1981: Lambor~hini 1987). In the sensory spinai cord each nerve fiom the dorsai root courses centraiiy into the dorsai horn of the gray matter and divides into a mediai and laterai division (e.g., Antai et ai. 1980; see fig. 6.8 for general appearance of &coming sensory fibers). The laterai division consists of a small-diameter
U

Fig. 6.9. Schematic view of the posterior spinai cords inXenopus h & and Rana mtesbeiana tadpoles. The soiid bars indicate the position of the celi bodies of motor neurons. Note that the spinai nerves (SN) exit the cord progressively more posteriorly than the position of their celi bodies and that the spinai cord extends more posteriorly in &nopus. Redrawn from K. Nishikawa and Wassersug (1988); reprinted by permission of Wdey-Liss, a subsidiary of John Wdey & Sons, Inc., and of Nishikawa.

INTEGRATION

155

described the reticulospinal, rubrospinal, tectospinal, and trigeminospinal tracts; and characterized two smder projections, one from the posterior ventral nucleus of the midbrain and one from the interstitial nucleus of the medial longitudinal fasciculus. A. Roberts and Aiford (1986) described the role of descending neurons in producing fiaive swirnming in Xenopus. In the ventral spinal cord, the number of motor neuron cell bodies is matched to the number of muscle motor units (McLennan 1988; Prestige 1967,1973; Rubin and Mendell 1980; D.C. Speny and Grobstein 1983). There is a developmental overproduction of motor neurons, and cells that do not establish connections or the proper type of connections with motor units eventudy die (Ferns and Lamb 1987; Hughes 1961; Kett and Poilack 1985; Lamb 1981; McLennan 1988). The result appears to be an appropriate matching of motor neuron number and size with motor unit numbers and sizes (D. C. Sperry 1987). The organization of cervical spinal cord motor nuclei differs between anurans and caudates (D. B. Wake et al. 1988). The development of spinal motor neurons and the mechanisms these cells use to match motor unit numbers is an Fig. 6.10. The spinal cord in a young tadpole. Abbreviations and symactive area of research (e.g., Farel and Bemelmans 1985, bols: I (no stipple) = developing interneurons, M (dark stipple) = devel1986; Lamb 1981; Lamb et al. 1989; McLennan 1988; oping motor neurons, and R-B (hrsomedial b k k ) = Rohon-Beard celis. The developing central canal (centralopenspace) is delimited by D.C. Sperry and Grobstein 1983). Nordlander (1986) degerminal zones (lightstipple).Developing ascending and descending scribed the normal development of motor neurons in the tail fiber tracts (iwe~ularstipple) located lateraiiy. Redrawn from A. Rob- ofX;ompus tadpoles, and D.C. Sperry and Grobstein (1985) are erts and Clarke (1982); reprinted by permission of The Royal Society examined the effects of hormonal manipulation on lumbar of London and Roberts. motor neuron numbers and cell sizes inX;ompus. Van Mier et ai. (1985) studied the development of motor neuron denproject lateraily into the w h t e matter, where they are con- drites in first- and second-order cells within the ventral horn. tacted by axons descending from the brain stem (see fig. C. L. Smith and Frank (1988a, b) and Van Mier and Ten 6.8). The morphology of motor neuron dendrites has been Donkelaar (1988) examined the peripheral speciicity of senstudied extensively (Antal et al. 1986; Bregman and Cmce sory connections in the developing spinal cord of R. cates1980; Rosenthal and Cmce 1985). Babalian and Shapovalov beiana. (1984) studied the synaptic interactions between single ventrolateral tract fibers on motor neuron dendrites. Brain The anuran spinal cord receives inputs from a variety of Hindbrain (rhombencephalon) sources within the brain (Ten Donkelaar and De Boer-Van Huizen 1982). Afferents from the forebrain diencephalon The hindbrain consists of the medulia (caudal ~ o r t i o n the of trave1 in the ipsilateral thalamospinal tract (Ebbesson 1976). brain stem), isthmus (connection between the m e d d a and The bulk of this projection is to the two anterior spinal seg- the midbrain), and cerebellum. The spinal cord grades into ments where neurons effect neck movements involved with the m e d d a so that in gross appearance it is difKcult to deterixing gaze in adult frogs. The presence and function of these mine exactly where the spinal cord ends and the h d b r a i n projections in tadpoles is unknown. Vestibular neurons in begins (figs. 6.3-6.4). the medulia also project to the spinal cord via the vestibuloThe organization of the brain stem (hindbrain + midspinal tract, where they terminate within the dendritic fields brain) is &nerdy the same as the spinal cord, with sensory of both medial and lateral motor neuron pools. Vestibular elements located dorsdy and motor elements ventrally (fig. inputs iduence posture and orientation (Stensaas and Sten- 6.5). The verticdy oriented sensory portion of the brain saas 1971). stem is termed th alar plate, and &e horizontaily oriented Reticulospinal pathways are poorly understood in anur- motor portion is termed the basal plate (fig. 6.5A). The divians. Medial reticular cells project to the cervical spinal cord, sion between these two areas is termed the sulcus l i t a n s and lateral reticular cells project to the lumbar spinal cord. and is visible along the wail of the IVth ventricle (fig. 6.5A). Accurnulating evidence suggests that reticulospinal neurons The brain stem can be divided further into longitudinal colform part of the pattern generating circuits of a variety of umns that represent funaional similarity (fig. 6.5B). The sobehavioral patterns including swimming (Van Mier and Ten matic motor column lies along the midline, somatic sensory Donkelaar 1988). Corvaja and d'Ascanio (1981) investi- portions lie along the lateral border, the visceral motor colgated the spinal projections from the midbrain in Bufa, umn is in a mediocentral position, and the visceral sensory

MICHAEL J. LANNOO

TBS

Fig. 6.11. A series of transverse sections through the tadpole brain. In each section, key nuclei and tracts are shown on the left, and gross histological appearance is shown on the right. Some structures may be located slightly anterior or posterior to the sections shown. Abbreviaaons: APL = lateral pomon of the amygdala, APM = mediai portion of the amygdala, CB = cerebeiium, DP = dorsai pailium, DS = dorsal striaturn, FLM = medial longitudmal fasciculus, H D = dorsal habenula, HV = ventral habenula, HY = hypothalamus, LL = lateral lemniscus, L Line = terminal area for prirnary lateral h e derents, LPD = lateral dorsai paliium, LPV = lateral ventral pallium, LS = lateral sepnun, M = Mauthner cell, MP = medial pallium, MS = mediai

sepnun, nuc CB = nucleus of the cerebeilum, OC = optic chiasm of CN 1 , POA = preoptic area, PM = profundus mesencephali, RF = 1 reticular formation, SPC = spinocerebeilar tract, ST = solitary tract, TBSP = tectobulbar and tectospinai tract, TE = tectum, TO = torus, VS = ventral striatum, VTN = ventral thalamic nucleus, 1 1 = nucleus 1 of CN 1 1 VI = nucleus of CN VI, Vm = motor nucleus of CN V, 1, VIII = terminal area for auditory and vestibular prirnary derents, IXm = motor nucleus for CN IX,and Xm = motor nucleus for CN X. Redrawn from Nieuwenhuys and Opdam (1976);reprinted by permission of Springer-Verlagand Nieuwenhuys.

INTEGRATION

157

Fig. 6.12. Horizontal schematic section showing key brain stem and ventral (right) brain nuclei through portions of the dorsal stem. Somatic motor nuclei are indicated by dark stipple, visceral motor neurons by hatching, and octaval region by light stipple. Abbreviations: CB (hea~y dashed line) = cerebellum, CN TV = trochiear nerve, C N V = trigeminai nerve, CN Vi1 = faciai nerve, C N Vm = auditory/vestibular nerve, C N M and X = common accessory and vagai nerve root, I = nudeus isthm, LLi = lateral line region, nALL = anterior lateral iine nerve, nPLL = posterior lateral line nerve, PRM = nudeus profundus, RU = nudeus ruber, SO = superior olive, ST = solit&tract, TO = t o m , I11 = oculomotor nucleus, TV = trochiear nudeus, Vm = trigeminai motor nucleus, VI = abducens nucleus, VIIrn = faciai motor nucleus, Vi11 = octavai region (lkht &@h),M and Xm = motor nudei of the accessory and vagai spinai nerves, and XII = hypoglossal nucleus. Reprinted from Nikundiwe and Nieuwenhuys (1983);reprinted by perrnission of Wiiey-Liss, a subsidiary of John Wiiey & Sons, Inc.

(v)

column is laterocentral (fig. 6.5B). T h s gude can be used to determine the approximate locations of first- and secondorder nuclei and their tracts ( N h n d i w e and Nieuwenhuys 1983 and Opdam et 4 . 1976). Histologicaiiy, four laminae withm the brain stem are superimposed upon these organizational columns (fig. 6.5C). A medial lamina of cells termed the ependymal layer iines the fout-th ventricle (figs. 6.5A, C). These smaii cells send a single neurite covered with- numerou short appendages radiaiiy into more lateral laminae. Lateraiiy adjacent to the

ependymal layer, the subependymal layer is completely free of cells but contains the radial neurites of the ependymal cells in the brain stem. The stratum griseum contains the majority of the cell bodes. These cell bodes can be either large or smaii with large ceiis frequently grouped into clusters. The stratum albumen is the most peripheral and is a continuation of the peripheral fiber tracts of the spinal cord. Axons within the stratum albumen send collaterals or terminal axons into the stratum griseum. The stratum albumen also contains the terminations of ependymal cells and some ceii groups embedded within the fiber layers. Arnong amphbians, the brain stem of anurans is considered to be more Merentiated than that of caudates. N h n d i w e and Nieuwenhuys (1983) used cell staining techniques to characterize the nuclei of the brain stem in Xenopus, and these authors describe seven primary efferent (motor) nuclei, 13 primary afferent (sensory) nuclei, 7 reticular formation nuclei, and 15 "relay" nuclei (see Opdam et ai. 1976 for an inter~retation the brain stem orof ganization in Rana). The brain stem receives or sends fibers from aii of the cranial nerves except CN O (= terminal nerve) and CN I (= olfactory nerve) which project into the telencephalon. In the brain stem the division between the mid- and hindbrain occurs posterior to the leve1 of CN l (figs. 6.3). The soV matic sensory column (SA, somatic afferents; fig. 6.5B; see longitudial organization above) is located dorsolaterally and receives inputs from -proceeding back to front -the posterior division of the lateral line nerve (nPLL), CN VIII, the anterior division of the lateral line nerve (nALL), and the sensory component of CN V (fig. 6.12). Within the acoustico-vestibulo-lateral line complex, the vestibular nucleus is most ventral. the two acousuc nuclei are central. and the lateral iine nucllus is dorsal. The nuclei for the general and special visceral afferents (VA; fig. 6.5B) are located within the midlateral brain stem and consist of components of CNs V, VII, E, and X. In general, these nuclei contain diffuse assemblages of ceiis and are not easily recognized (Nikundiwe and Nieuwenhuys 1983; Opdam et al. 1976). The nuclei for the general and special visceral efferents (VE; fig. 6.5B) occupy the mediocentral brain stem. The general visceral efferent is the parasympathetic innervation (carried by CNs 111, a, X). The special visceral efferent comand ponent consists of the neutrons projecting to muscles derived from branchiomeres (carried by CNs V, VII, E , X, and XI). These visceral motor nuclei form a continuous column through the brain stem (Ebbesson 1976; Nikundiwe and Nieuwenhuvs 1983; Oka et al. 1989; Opdam et al. 1976). The most Aedial biain stem column contains the somatimotor cell bodes (SE, somatic efferents; fig. 6.5B) that innervate the tongue (CN XII) and the extraocular eye muscles (CNs 111, n and VI; fig. 6.12). Stuesse et al. (1983) re) ported that the hypoglossal nerve (CN XII) in Ranapipiens contains a dorsomedial and a ventrolateral subnucleus. The reticular formation is located in the ventromedial portion of the brain stem (fig. 6.11G) and, although not shown completely here, projects from the midbrain through the hindbrain and into the cervical spinal cord. Its cells and fibers are difise. The reticular formation receives collaterals
,
L

158

MICHAEL J. 1,ANNOO

from brain stem nuclei and tracts involved in reflex arcs. The superior olive (fig. 6.12) is located within the basal plate and receives neurons from the dorsal octaval nucleus (CN VIII; F d e r and Ebbesson 1973) and the midbrain tegmentum (Wilczynski and Northcutt 1977). The olive sends fibers to &e midbrain toms semicircularis (~ikundiwe Nieuwenand huys 1983 and Opdam et al. 1976). The nucleus isthmi (fig. 6.12) may be involved in processing binocular inputs in adults and in tadpoles with overlapping visual fields. Most tadpoles do not have binocular vision and the function of this nucleus in tadpoles with lateral eyes is not clear. The nucleus isthmi is reciprocdy connected to the midbrain tectum (Grobstein and Comer 1983; Udin et ai. 1992; Wdczynski and Northcutt 1977). Among fiber tracts the medial longitudiiai fasciculus (figs. 6.11E, F, G) is a phylogenetically old feature of the vertebrate brain stem. This tract contains reticulospinal, vestibulomesencephalic, and vestibulospinal axons and mediates the vestibulo-ocular reflex. ~ h lateral lemniscus (fig. e 6.11F) carries second-order auditory and lateral line nerves from their primary nuclei to the midbrain toms semicircuiaris. (Wilczynski 1988; Will 1988). S e m (1972) detailed the develo~ment these and other brain stem structures in of Rana temporaria. Mauthner neurons (fig. 6.11G) are the largest axons in the brain stem and appear to initiate the startle response of tadpoles, although other ceiis may also contribute (R K. K. Lee and Eaton 1989). Mauthner neuron connections have been described in Bombina, Bufo tewestris, Hyla cinerea, I&.loulapulchra, Potypedates bucomystax, Rana escubnta, Scaphwpus holbrookii, and Xenopus laevis (Wdi 1986). Mauthner ceil dendrites are contacted by fibers from the CNs VI11 and V, lateral line neurons, and midbrain neutrons (Cioni et ai. 1989; Wdi 1986). The spinal motor neurons that receive input from the Mauthner ceils in larval amphibians are the earliest to develop (Blight 1978; Nordlander et al. 1985). Stefaneiii (1951) reported that Mauthner ceil degeneration is related to taii resorption at metamorphosis even in species with aquatic adults, while Moulton et al. (1968) and Will (1986) reveaied the persistence of Mauthner cells in adult anurans. The sensory and motor connections of CN XII (fig. 6.11H; Stuesse et ai. 1983) of Ranapipiens originate from two nuclei located mediaiiy and laterdy withm the cauda medda. Motor fibers from the medial nucleus project to the extrinsic tongue muscles in adults. Motor fibers from the lateral hypoglossal nucleus innervate the sternohyoid muscle. Sensory fibers arise mainly from the tongue, and in the brain stem, they trave1 posteriorly in the dorsolateral fiber tract to thoracic levels. The development of these motor and sensory connections in tadpoles probably depends on the extent of tongue development. The central projections of the CN IX-X complex (fig. 6.12) were mapped by Stuesse et ai. (1984; adult Rana catesbeiana) and Simpson et ai. (1986; adult Xenupus laepis).Motor fibers of CN IX arise in the brain stem from a s m d ventrolateral nucleus, and those of CN X originate from a slightly more posterior nucleus (Stuesse et al. 1984). AfFer-

ents from CNs IX and X enter the dorsal roots of the respective nerves and descend in two tracts, either the solitary tract (fig. 6.12) or the spinal tract of CN V and the dorsolateral tract of the spinal cord. In Xenupus, CNs IX and X fibers are within the three roots of the H-X complex and withm the nPLL (Simpson et al. 1986). The most anterior of these IX-X roots (root 1) contains sensory fibers that terminate in the solitary tract and on lateral line efferents. Root 2 contains somatosensory fibers that terminate in the posterior meduiia and anterior spinal cord and motor fibers from CNs IX and X. Root 3 contains motor fibers to the laryngeal musdes and efferents to the viscera. The results of these studies confirmed the same general conclusions of Rubinson and Friedman (1977) on Rana catesbeiana, R. pipiens, and Xenupus muelbri. Lateral line, acoustic, and vestibular afferents terminate in different areas of the somatic sensory column (figs. 6.116.12; Fritzsch 1988; McCorrnick 1988; Wiil1988; Wdi and Fritzsch 1988). Herrick (1948) and Larsell (1934) proposed that lateral line target neurons receive auditory fibers at metamorphosis. Fritzsch et ai. (1984) and Jacoby and Rubinson (1983) did not support this hypothesis based on modern tract-tracing techniques. Lateral line fibers project into an intermediate nucleus, whle auditory fibers project to a distinct nucleus located more laterdy and ventrdy (Fritzsch et al. 1984). According to Aitman and Dawes (1983), some anterior fibers in the posterior division of the CN LL (nPLL) may be cutaneous afferents that project to the spinal cord in a manner similar to that of CN V. Within the lateral line system, second-order neurons project to the contralateral midbrain torus semicircularis (figs. 6.11-6.12) via the lateral lernniscus (fig. 6.11F). Efferent neurons (not shown) project from the brain stem back to receptors and prevent hair ceiis from firing. These fibers originate fiom a CN VI11 nucleus located vent r d y and medially withm the medulla (Fritzsch 1981a; Russell 1976). Primary afferents from the amphibian and basilar papillae (fig. 6.13, see auditory system) project into one or two nuclei at the dorsolateral edge of the aiar plate (figs. 6.11-6.12; Aitman and Dawes 1983; Fritzsch et ai. 1988a, b; Jacoby and Rubinson 1983). Rubinson and Skiies (1975) suggested that primary auditory afferents also project directly to the superior olivary nucleus. Second-order projections from the primary auditory nucleus project to the superior olivary nucleus (Rubinson and Skiies 1975) and the midbrain torus semicircularis (fig. 6.12; Nikundiwe and Nieuwenhuys 1983; Opdam et al. 1976, Rana; Wilczynski 1988). Sensory fibers from neurons innervating the semicircular canals, utricle, saccule, and lagena project into the meddary vestibular complex(fig. 6.13). Sensory vestibular fibers aiso project directly to the cerebeiium (Altman and Dawes 1983). Efferents from the brain stem to the vestibular end organs arise from the same nucleus as lateral line efferents, and single efferent neurons may send axons to both the lateral line and vestibular systems (Claas et al. 1981). Based on second-order projections, the vestibular complex can be divided into anterior and posterior divisions (not shown). In Rana

INTEGRATION

159

Fig. 6.13. The organization of the anuran ear in adults. In tadpoles, a bronchiai columella pierces the dorsai aorta, and the plecmim and tympamc columella have not yet formed. Abbreviations: AP = amphibian papilla, BP = basilar papiia, H = hindbrain, L = lagena, P = plecmun, S = saccule, SC = semicircular canals, T = tympanum, TC = tynpanic columella, and U = utride. Redrawn fiom Capranica (1976);reprinted by permission of Spnnger-Verlagand Capranica.

pi@ns the anterior portion projects to the nuclei of CNs 111, IV, and VI and contributes to the vestibulo-ocular reflex; posterior efferents form the vestibulospinai pathway (Montgomery 1988). Th trigeminai nerve is composed of a motor nucleus that innervates the muscles of mastication (fig. 6.12), a midbrain (mesencephalic) sensory nucleus that mediates masticatory forces, a principie sensory nucieus, and a spinai sensory tract that is an anterior continuation of the spinai dorsolaterai column (Nikundiwe and Nieuwenhuys 1983, XenUpzls; Opdam et ai. 1976, Rana). These last two trigeminai nuclei mediate somatosensation from the head. The mesencephalic and motor trigeminai nuclei are involved in a reflex arc. Resistance to iaw closure in both tadpoles and adults is relayed through receptors to ceiis in the mesencephaiic trigeminai nucleus (see Koilros and McMur- Midbrain (msencqhalon) rav 1956 for cellular details of thts nucleus) that project to The midbrain is the anterior continuation of the brain stem motor trigeminai neurons and prevent further jak closure. and is divided into a ventral tegmentum, which continues S. Lewis and Straznicb (1979) examined the development the ventral motor columns, and the dorsai tectum, which of these neurons. Motor trigeminai cells formed embryoni- continues as the meddary sensory columns (figs. 6.4 and cally are responsible for buccai movements in both the tad- 6.11). The tegmental nuclei include CNs I11 and IV and nupole and the adult. Aey and Barnes (1983) and Barnes and cleus ruber (fig. 6.12; Nikundiwe and Nieuwenhuys 1983). Wey (1983) found that the same cells that innervate the tad- The nuclei of CNs 1 1and IV (aiong with the hindbrain CN 1 mle muscles innervate the adult muscles. demite the facts VI and perhaps CN V) send motor fibers to the extraocular , that adult muscles first arise at metamorphosis and that the muscles (Straka and Dieringer 1991). In the tegmentum tadpole and adult muscle groups subserve different feeding these nuclei are located dorsomediaiiy. Craniai nerve I11 behaviors. These authors indicated that trigeminai motor emerges from the brain stem ventraiiy, and CN IV emerges neurons are recycled and respecified during metam~r~hosis. dorsaiiy. Both nuclei receive vestibular fibers from the medOmerza and Aey (1992) showed that about 80% of motor d a r y vestibular nucleus. The t o m semicircularis receives axons supplying adult muscle fibers originate from tadpole second-order lateral line (Plassman 1980; Will et al. 1985) neuromuscular junctions. and second- and hrd-order auditory inputs (Feng and Lin 1991; Potter 1965), which it processes and sends to the tectum. The t o m semicircularis also receives ascending inputs The cerebeiium is perhaps the easiest neuronai structure to from the spinai cord (Ebbesson 1976), reticular formation, recognize because of its location (figs. 6.11F-6.12), highly hypothaiamus (Neary and Wiczynski 1977), and reciproca1
I

regular cellular organization, and stereotyped inputs. The cerebellar cortex is composed of an external molecular layer, a thin layer of large Purkinje cells, and an internal granular ceil layer (e.g., Herrick 1948; Larseii 1967; Sotelo 1976; Szkely et ai. 1980). Deep to the cortex, cerebellar nuclei (fig. 6.11F) serve as relay centers for afferent and efferent cerebdar fibers (Gonzalez et ai. 1984; Montgomery 1988). The cerebellum receives i n ~ u tfrom the s ~ i n a i s cord.-vestibular nuclei, lateral line system, and other hindbrain nuclei (Grover and Grsser-Cornehls 1984; Larsell 1925; Sotelo 1976).Neurons in the dorsal root ganglia of the cervicai and lumbar regions project directly to the cerebellum (Joseph and Whitlock 1968; Szkely et ai. 1980).These fibers terminate in a somatotopic pattern. A second type of spinocerebellar projection, arising from secondary cells in the spinai cord, is present in both dorsal and ventral spinocerebellar tracts (Van Der Linden et ai. 1988) and terminates as mossy fibers in the cerebellar granular cell layer. Climbing fibers originate in the medda at the location of the inferior olive. Efferents from the cerebellar nuclei project to the contralaterai cerebellar nucleus and to the s ~ & d cord. Cerebellar efferents aiso project biiateraiiy to the'basal plate of the med d a and to the midbrain (Montgomery 1988). The cerebellum functions in motor learning: and coordi" nation (Freeman 1965) and in anurans consists of the corpus, paired auricles, and a nodulus. The cerebelium does not fuiiy mature untii metamorphosis (e.g., Gona et al. 1982; Koilros 1981). The appearance of the corpus cerebellum in tadpoles is associated with the development of the tail musculature and the spinocerebellar tracts. The cerebellum aurieles, which may receive and process lateral line inputs, are more fdly developed in tadpoles than adults (Larseii 1967). The protraaed development of the anuran cerebellum has made this structure a usem model for researchers interested in pattern formation withtn the nervous system (e.g., Hauser et ai. 1986a, b; Uray 1985; Uray and Gona 1979, 1982; Uray et ai. 1987, 1988).

160

MICHAEL J. LANNOO

connections from the contralateral t o m (Nikundiwe and ery et ai. (1982) examined the rnesencephaiic nuclei responNicuwenhuys 1983; review by Northcutt 1980). The nu- sible for the optokinetic response in R.pipiens aduits. deus mber forms reciprocal connections with the cerebellum (Larson-Prior and Cruce 1992). Three additional tegmental nuclei have been identiied. The telencephalon and diencephaion (fig. 6.4) integrate senThe nucleus profundus (fig. 6.12) receives fibers from the sory inputs and initiate motor outputs. The telencephaion superior olive (Rubinson and Skdes 1975), but its funaion appears to integrate sensory information beyond that found is unknown. The optic nucleus receives bilateral connections in the thaiamic nuclei (Ebbesson 1980: Northcutt 1981). projecting from each retina (Levine 1980). The mesence- The telencephalon receives anterior inputs from the oifacphalic nucleus of CN V contains large sensory neurons that tory buibs (CN I) and terminal nerve (CN 0) and posterior systems mediate masticatory forces. More cells occur in the trigemi- inputs from the visual, auditory, and nai rnesencephalic nucleus of tadpoles than in adults and by way of thalamic (diencephaiic) relays (Kicliter 1979; during metamorphosis a large number of these celis die Northcutt 1974: Wilczvnski and Northcutt 1983a). The di(Kollros 1984, Ranap$iens; Koiiros and Thesse 1985, Xn- encephalon includes the epithaiamus, consisting of the haopus l m i ) . The midbrain tectum is located dorsal t o the teg- benula and the pineal gland; the hypothalamus, which conmentum (figs. 6.4 and 6.11E) and receives first-order inputs trols homeostasis and drives such as hunger and thirst (fig. from retinai gangiion ceiis and third-order inputs from the 6.11); and the thaiamus, which functions primarily as a relay laterai line and aiiditory systems (Ebbesson 1970; Keating nucleus. Stehouwer (1987) described the effects of forebrain and Gaze 1970; Lowe 1986). The tectum integrates sensory ablations on the behavior bf tadpoles. inputs with motor commands (Elepfandt 1988a, b, lateral Thc terminal nerve (CN O), of iinknown function in anline system,X. laeyt?).Tectai inputs were considered by Wilc- urans, is associated with the olfaaory nerve (Herrick 1909; zynski and Northcutt (1977), and Lowe (1987) showed that Hofmann and Meyer 1989; aiso Szabo et al. 1990). Hofsingle cells within the tectum respond to both lateral line and mann and Meyer (1989) reported that the terminal nerve in visual inputs. Vestibular and auditory signais also reacli the Bufi, &nus a n d x m p s h s projects to the forebrain memidbrain tectunl. Auditory and laterai iine information ter- diai septum, preoptic nucleus, and the hypothalamus. The minates in the toms sernicircularis before being relayed to olfactory bulbs (CN I) located distal to the anterior telenthe tectum (Feng and Lin 1991; Pettigrew 1981; Wilczynski ce~haloh corhected to the brain via olfactorv tracts. The are 1981). The toms is homologous with the mammaiian infe- t&gets of these traas are the main olfactory b d b and the rior colliculus, and the tecturn is a homolog of the superior accessory oifaaory bulb (fig. 6.14; Scalia et ai. 1991a, b). colliculils. The ma& olfacto& bdb. a~cessorv olfactorv buib, and the The tectum is the best-studied structure in the anuran anterior olfaaory nucieus are interconnected ipsilaterdy. brain. Retinal ganglion celi axons of tadpoles projcct to the The main oifactorv bulb and anterior oifactorv nucleus moicontralateral tecnun in an ordered fashion, an arrangement ect to the contralaieral mediai waii of the telen&phalon.ipithat faciiitates the study of neuronal connections (e.g., Con- lateral main olfaaory b d b fibers aiso terminate near the oristantine-Paton and Capranica 1976; Debski and Con- " of the brain stem mediai loneitudinai fasciculus. The zin stantine-Paton 1990; Reh et 4 . 1983). The formation of accessory olfactory bulb projects to the lateral cortex of the order within the nervous system is itself an important ques- contraiaterai telencephalon via the lateral forebrain bundie. tion. Recent papers on this subject involving anurans include The amygdaia, a nucleus associated with the olfactory systein Constantine-Paton (1987), Constantine-Paton et ai. (1983), (fig. 6.14), projects to two nuclei within the ipsilaterai hypoFujisawa (1987), Grobstein et al. (1980); Levine (1980), thalamus and a nucleus posterior to the ipsilateral nucleus Montgomery and Fite (1989), Reh et al. (1983), and Straz- isthmi (fig. 6.12; Kemali and Guglielrnotti 1977). Scalia nicky and Tay (1982). Detads of the salamander tectum in et ai. (1991a, b) provided detaiied descriptions of the inain the context of visual behavior were presented by G. Roth and accessory olfaaory bulb projections in Ranapipiens. Ji(1987), and Lzr et ai. (1991) deinonstrated differences in ang and Holley (1992) described the olfactory bulb outputs the pattern of tectal lamination benveen Rana escunta and in R. ridibunda, and Burd (1992) and Byrd and Burd (1991) Xemps &S. Anterior tectal projections in the tectotha- described olfactorv develo~rnent Xx&us ue9is. in lamic tract and the postoptic commissure (Lzr et ai. 1983; The posterior telencephaion is divided into four areas Vessekn et al. 1971) wili be considcred with the forebrain. around each lateral ventricle (figs. 6.11 and 6.14). The dorPosterior tectai efferents project to the nucleus isthmi (fig. sal part of the posterior te~ence~halondivided into medial is 6.12) and in turn comect bilateraiiy back to the tectum and lateral pallial areas. The septal area fies ventral to the (Grobstein and Comer 1983; Gruberg and Udin 1978; medial I ~aliium. the striatal area iies ventral to the lateral and Udin and Fisher 1985). Other posterior tectal projections pallium and lateral to the septum. The amygdala lies in the include tectobulbar (bulb refers to the meddia) and tectos- posterior striatum. The septum and striatum are synaptic pinai fibers which rnay be responsible for generating or stations where olfactom fibers ioin with fibers from the ditransmitting motor commands (Lzr et ai. 1983). The pre- emephaiic thaiamus &d the hidbrain. The telencephaion tectal nucleus lentifornus (not shown) receives contraiaterai connects with posterior brain areas through the mediai and retinal inputs as well as inputs from the tecnun and tegmen- laterai forebrain bundies (fiz. 6.14). The mediai forebrain tum in Ranapipiens (Montgomery et ai. 1985). Montgom- bundle contains ascending and descending conneaions that

soma tose riso^

INTEGRATION

161

Fig. 6.14. Horizontal section of key forebrairi nuclei. Locations of nuclei are shown on the left and gross histological appearance of the brain on the right. Abbreviations: AMY = amygdala, AOR = accessory olfactory bulb, AON = accessory olfaaory nucleus, CB = cerebelIum, CN I = cranial nerve I (olfactory, s h o m larger than life), HIP = hippocampus, MOB = main olfactory bulb, LFB = lateral forebrain bundie, MFB = media1 forebrain bundie, PRE OPT = preoptic area, and TEC = tectum. Redrawn from Kemali and Braitenburg (1969) and Kemali and Gughelmotti (1987); reprinted by permission of Springer-Verlag.

join septal olfaaory and limbic centers with the midbrain tegmentum and diencephaiic preoptic and hypothaiamic regions. Specific connections here include descending fibers from the sepnim that project to the ventrai thaiamus, preoptic area, and anterior hypothaiamus. The lateral forebrain bundle connects the striatum with the posterior brain and contains afferents from thaiamic nuclei, which in turn receive inputs from the midbrain t o m and tectum (Wilczynski and Northcutt 1983a). Other laterai forebrain bundie fi-

bers include inputs from the ipsilaterai amygdala, the ventrai thalamus, and adjacent preoptic areas. The striatum projeas t h r o u ~ h laterai forebrain bundie most heavilv to the con" the traiaterai anterior endopeduncular nucleus, ventrai thaiamus, and preoptic area (Wilczynski and Northcutt 1983b). Neary and Northcutt (1983) recognized two habenular nuclei within the epithaiamus, two hypothdamic regions (preoptic area and infundibulum), three thaiamic regions (dorsai, ventrai, and tuberculum), and a transitional preteaai area in the diencephaion of adult Rana catesbeiana. ~ h epie thaiamic habenula (fig. 6.11D, E) connects to the septum, hippocampus, hypothaiamus, and thaiamus (Kemaii 1974; Kemaii and Gugiielmotti 1977) and is proportionaiiy larger in animais with a weii-developed forebrain. M. J. Morgan et ai. (1973) described the left-right asyrnmetry in the habenulae of adult R. temporaria. The habenula plus the interpeduncular nucleus form the inner ring of the limbic system, which may be the highest correlative center in the anuran brain. ~ e r r i c k (1948)suggested that this structure controls feeding. The epithaiamus aiso contains the pineai organ (epiphysis), which has a smaii projection (best developed in anurans within amphibians) to the dorsomediai surface of the epithaiamus (fig. 6.15). These fibers contain the frontal and pineal tracts (parietal nerve) and enter the pretectai region in the posterior part of the epithaiamus. I n the laterai and ventral waiis of the diencephaion, the thaiamus forms a web of connecting fibers with ai1 contiguous parts of the brain and serves as & important center fo; sensoh correlation. For example, the optic nerves primarily project to the optic lobes of the midbrain, but aiong their course collaterai fibers are given off to the thaiamus (Fite et ai. 1977) where they synapse with fibers from other sensory systems (Neary and Northcutt 1983). J. C. Haii and Feng (1987) considered the anatomy and physiology of the thaiamic auditory region in adult Ranapipiens and presented evidence for paraiiel circuits. Posteroventraiiy within the diencephaion, the bilobate hypothaiamus (fig. 6.11E) is divided into preoptic and tuberoinfundibular regions. The preoptic nucleus is conneaed to the ventrai thaiamus and the ventrai lobe of the hypophysis (pituitary). Neurons within the preoptic area and ventral hypothaiamus respond to conspecific mating caiis (Aiiison 1992, Hyla cznerea). The tuberoinndibular nucleus is connected to the hypophyseai portal vessels. The general functions of the hypothaiamus include activating the brain stem reticular formation; controlling the autonomic nervous system, metamorphosis, and pituitary gland; correlating gustatory, visceral, and olfactory inputs; and regulating feeding rates and water balance (Sarnat and Netskv 1981). The hypothaiamus and pre-optic area receive fibers from most regions of the forebrain, but telencephalic olfactory inputs and medullary gustatory inputs appear to predominate. Efferent fibers project to motor centers, the brain stem reticular formation, thaiamus, forebrain, and olfactory iimbic centers. The hvoothaiamus does not receive fibers &om the major somatosensory pathways. In nonrnarnrnahan vertebrates, fibers of the optic tract that terminate directly in the
~ ~

il

MICHAEL J . LANNOO

the lateral line system, olfaction, taste, vision, and rnisceiianeous senses (,.i., photodetection through the pineai complex and somatosensation-pain and posturai sensation, temperature, and touch). For each system, the material wiN be presented in the sequence: receptor structure and function, peripheral innervation, first-order connections within the brain or spinal cord, development, and patterns of phylogenetic and ecological variation.

Fig. 6.15. The projection of pineal axons into the ventral diencephalon in embryonic anurans. Low-power diagram (upper) for orientation of brain regions within the embryo, and higher magnification ( h m ) to show features of the projections. Abbreviations: F = forebrain, H = h d b r a i n , M = midbrain, O P = optic placode, OT = otic placode, P = pineal gland, P R OPT = preoptic commissure, V = CN V, and VC = ventral commissure. Redrawn from R. G. Foster and Roberts (1982);reprinted by permission of Springer-Verlag and Foster.

preoptic area and hypothaiamus may mediate color changes in the skin. The specific conneaions of the preoptic area of the hypothaiamus (Neary and Northcutt 1983; fig. 6.14) that project to the pituitary are associated with the medial forebrain bundle, mediai pallium, laterai amygdala, and infundibuium; see Fite (1985) for data on the pretectai area. The laterai forebrain bundle is the main afferent pathway for visual signais into the forebrain (Kicliter and Northcutt 1975). Information about the visual world, especially moving objeas and changes in general iliumination, is processed by telencephalic neurons. The pituitary gland lacks neurons and is composed of gliai atrocytes and pituicytes (Gerschenfeld et ai. 1960). Within the pituitary, the neurohypophysis is located ventral to the hypothalamus and connected to it via a fiber ttact called the infundibulum, and the adenohypophysis is located in a diverticulum in the roof of the mouth.

Sensory Systems
The sensory capabilities of generalized tadpoles include: audition and vestibular functions, mechanoreception through

Auditoly System The anatomy, development, evolution, and physiology of the anuran auditory system have been weii studied (e.g., Fritzsch et al. 1988a; Viiiy 1890; Wever 1985). The role of audition in spacing, courtship, and isolating mechanisms contributing to speciation is weii known for aduits (Fritzsch et ai. 1988a; McCormick 1988; Ryan 1986, 1988; J. J. Schwartz and Weiis 1983). It is important to realize that the mode of transfer of pressure waves (sounds) to the inner ear (Jaslow et al. 1988) is different in tadpoles than in aduits. In aduits, sound impinges on the tympanic membrane where pressure differentials are transformed into mechanical movements. The middle ear transmits these vibrations to the inner ear via (fig. 6.13) the plearum (attaches to the tympanum), columeiia (the connecting bone), and opercuium (links columeiia to the oval window, which in turn connects to the inner ear). In some anurans the columeiia has been lost (Trueb 1973, 1979). Motions of the oval window create movements of the perilymphatic fluid that generate movements of the endolymphatic fluid. The endolymphatic fluid stimulates the hair of the sensorv neuroe~i;heliaand the basilar and amphibian papillae. Tadpoles hear despite lacking middle ear bones and a tvmDanum. Sound oressures are transmitted to the inner ear via a bronchiai columella. which " grows from the round window (a connection betwien the middle and inner ears) toward the ipsilaterai bronchus and lung at about the time of opercular closure (Witschi 1949, 1956). In the process of growing, the columeiia pierces the dorsal aorta (see Boatwright-Horowitz and Sirnmons 1997). According to Capranica (1976) and Witschi (1949), this unusual arrangement mav isolate the columella from other tissues and permit it a greater degree of movement. The columeiia attaches to the lung and bronchus, which vibrate in response to changes in water pressure. The inner ear of tadpoles and aduits contains two auditory receptive areas, the amphibian papiiia and basilar papiiia (fig. 6.13; E. R. Lewis and Lombard 1988). Amphibians are unique in this regard because most other vertebrates have one auditory receptive area epithelium (e.g., saccuie in fish, bailar papia in reptiles, and cochlea in birds and mammals; Wever 1973). The surface morphology of the amphibian papdia of Rana catesbeiana is described by E. R. Lewis (1976) and E. R. Lewis and Lombard (1988). The basilar and amphibian papiiiae are tuned to different frequencies which tend to be s~ecialized an animal's acoustic environment. to The number of hair ceii receptors within these papiiiae varies interspecificallyand with size intraspecifically. Typically, large animais have large papiliae that contain numerous hair celis
i

INTEGRATION
i Capranica 1976). The basilar papilla is the generalized auditory organ. Its hair ceiis receive only afferent innervation and respond to frequencies above 500 Hz. The amphbian papilla is sensitive to both low- and high-frequency stimuli. During metamorphosis of Bufi reguhrts, the walls of the dorsal aorta close; the bronchal columeiia is resorbed; and the oval window, operculum, and tympanic columella form. -1t metamorphosis the cartilage for the plectrum is visible and the tympanic membrane is visible as a crescent-shaped smicture (Capranica 1976). Auditory fibers from the basilar and amphibian papiiiae trave1 in CN VI11 (acousticovestidar nerve) and terminate in the auditory nucleus in the brain stem (Boord et al. 1970; Capranica and Moffat 1974; Larsell 1934; Matesz 1979; McCormick 1982). Acoustic fibers within CN VI11 are organized by the frequency of the stimulus they carry. The auditory system in anurans has been considered in detail by Fritzsch et al. (1988a). Shofner and Feng (1984) exarnined the developing basilar and amphbian papillae with both light and scanning electron microscopy and described an increase in the size of the sensory epithelium and the addition of hair cells as the animal grows. Capranica (1976), Fritzsch et al. (1988a, b), Sedra and Michael (1959), and Witsch (1949, 1956) detailed the changes in the auditory apparatus through larval development and metamorphosis in Bu$ and Rana. Mudry and Capranica (1980, 1987), Mudry et al. (1977), Neary (1988), and Neary and Wilczynski (1986) described auditory pathways in the forebrain of adult Rana. See Fritzsch (1988) for a description of the evolution of the auditory system in anurans.

163

striolar curve and, taken together, are responsive to movements through 360". Type I hair cells with long stereocilia that extend to the tip of the kinociiium are found along the striola. Type I1 hair cells with a long kinociiium and short stereociiia are found along the sides of the striola. The saccular mac~ila oriented nearly vertically and faces is posterolaterally. The striola is curved and the hair cells, conSidered toge&er, are sensitive through 360". Outside the striola, kinocilia are directed outward and inside they face inward. Most saccular hair cells resemble the utricular Type I1 cells. The posterior protrusion from the saccule is termed the lagena and has a ciliary pattern resembling the utricle. Afferent neurons are bipolar cells. Axons terminate in the vestibular nucleus (e.g., R. F. Dunn 1978). Hair cells also receive efferent fibers. Caston and Bricout-Berthout (1985) described conneaions of vestibular nerve to the horizontal semicircular canal in Rana esculenta and showed that the efferents both inhibit and facilitate the afferent response. These authors reported that neurons within the vestbular nucleus respond both to motion through the vestibular nerve and light pulses from the midbrain tectum. Van Bergeijk (1959) described a reflexive vestibular behavior that allows Xenopus tadpoles to maintain a stable position in the water colurnn.

VestibularSystem The vestibular system consists of three orthogonally oriented semicircular canals plus a utricle. a saccde. and a lagena that function together in orientation and maintenance of body position in space (fig. 6.13; E. R. Lewis and Lombard 1988). As with the lateral line and auditorv svstems. , , the sensory mechanism is fluid (endolymphatic fluid in the vestibular system) motion across an epithelium of mechanorece~tive ceiis. The sensorv e~ithelia the three senlihair of circular canals are contained in ampullary bulges w i t h the canais, while utricular and s a c d a r hair cells project into an otolithic membrane. The semicircular canals detect a n d a r acceleration or rate of body turning and tilting and are oriented in orthogonal planes with respect to each other (e.g., E. R. Lewis and Lombard 1988; Precht 1976). Each of the semicircular canals emerges from the elongate utricle. The anterior vertical semicircular canal detects motion in an anterolateral-posteromedial plane, the posterior vertical canal detects motions in an anteromedial-posterolateralplane, and the horizontal canal detects horizontal movements. The anterior vertical canal of one side of the tadpole is oriented in the same plane as the contralateral posterior vertical canal. The utricle and saccule monitor linear acceleration and gravitational puli. The utricular macula is oriented horizontally and is crescent-shaped around a central structure termed the striola. The hair cells are polarized in a direction radial to the
i

' 3

Lateral Lzne System The lateral line system of aquatic anamniotes consists primitively of mechanoreceptive and electroreceptive end-organs (Buliock et ai. 1983). Anurans retain mechanoreceptors (= neuromasts; Fritzsch 1981b) but have lost their electroreceptive subsystem. Mechanoreceptive neuromasts are sensitive to local water displacements (i.e., the large-scale movements of water molecules) rather than changes in water pressure (small local movements of water molecules that vibrate around a relatively fixed point) to which the auditory system is most sensitive. The lateral line system of tadpoles consists of three main lines each on the trunk and head (fig. 6.16A; Escher 1925; Kingsbury 1895; Lannoo 1987; Shelton 1970, 1971; M. Uchiyama et al. 1991). Tadpoles with midventral spiracles have a symmetrical arrangement of the neuromasts. but those with sinistral spiracles often have asymmetrical arrangements perhaps because of developmental interference caused by the spiracle (Lannoo 1987). Hair cells are the acnial displacement detectors of neuromasts (fig. 6.16B; Flock 1965; Flock and Wersdl 1962; Jande 1966; I. J. Russell 1976; M. Uchiyama et al. 1990, 1991). Hair cells contain a ciliary bundle (the "hairs") at their external suface that consists of a single kinocilium (fig. 6.16C) located at an edge of the bundle and numerous stereocilial T h s anatomical-polarization corresponds to a directional sensitivity. Displacements of the cilia in the direction of the kinocilium depolarize the ceii while displacements in the opposite directin hyperpolarize the cef Depolarizations increase the rate of firing of the ceii, and hyperpolarizations prevent the ceii from firing (Flock 1965; review by I. J. Russell 1976). Typically, adjacent hair cells are oriented oppositely so that a stimulus that depolarizes one hair ceii wdl

164

M I C H A E L J. L A N N O O

PSO

Fig. 6.16. The lateral h e system of a generalized tadpole. (A) Named components and orientation of neuromasts i tlie lateral line system. n Abbreviations: AN = angular, AOR = anterior oral, D = dorsal body, I 0 = infraorbital, LOR = longitudinal oral, M = middie body, PIO = posterior infraorbital, PSO = posterior supraorbital, SO = supraorbital, and V = ventral body line. (B) One p r i m q neuromast (a) and tliree secondary neurornasts (b) forming a receptor unit (= stitch) witli tlie polarization of hair ceUs as tliey project into the gelatinous copula. (C) A single hair cell cut tlirough the kmoc~lium(k) and stereociha bundle and showing tlie afferent (aff) and efferent (eff) imervation. Redrawn from Lannoo and Smitli (1989); reprinted by Oxford University Press.

hyperpolarize the other. Neuromasts are innervated by muitiple neurons, and hair ceiis that are oriented simdarly are innervated bv fibers of the same neuron. Neuromasts are usuaiiy elongated, and their direction of maximum sensitivity corresponds to the long axis of the oval. Sensitivitv in directions other than the main axis is Droportional to the cosine of the angle from the main axis (Flock 1965). Unlike other tadpoles, pipoid larvae have round neuromasts; such structure makes their axis of sensitivity difficult to determine (Lannoo 1987). In contrast, aquatic pipoid adults have linear neuromasts (Shelton 1970, 1971). Typicaiiy, neuromasts w i t h a line are oriented in the same direction (fig. 6.16A; Gorner 1963; Lannoo 1987; Shelton 1970). Dorsal trunk neuromasts are most sensitive in to dis~lacements a dorsoventral direction. and middle and ventrai neuromasts are most sensitive to anterior-posterior displacements. On the head, supraorbital and infraorbital lines around the eve are most sensitive in directions tanaen" tial to the margin of the eye. As these lines run onto the snout, neuromasts change orientation to become anteriorposteriorly sensitive. The angular l h e is most sensitive to dorsoventral displacements while the two hyoid lines combine to be sensitive to both dorsoventrai and anteriorposterior displacements (Lannoo 1987). As generaiized, pond-dwelling tadpoles grow, prirnary neuromasts formed embryonicaiiy divide to form secondary neuromasts (fig. 6.16B; Lannoo 1987; Russeii 1976). A secondary neuromasts derived from a single primary neuromast are aiigned in the same direction as the primary neuro-

mast and compose a receptor unit (after Zakon 1984; previously termed stitch or plaque, Lannoo 1987, 1988; I. J. Russeii 1976). Hair celis within neuromasts are innervated by bipolar prirnary a'ierent neurons whose ceii bodies are 1catd in ganglia (Shelton 1970; Van Der Horst 1934). Neuromast afferents course through branches of either the anterior (nALL, head) or posterior (nPLL, trunk) lateral line nerve. Once in the brain, aii lateral line a'ierents bifurcate to send one coiiateral anteriorly and one posteriorly within the meduliary nucleus mediaiis (Altman and Dawes 1983; Boord and Eisworth 1972; Fritzsch et al. 1984; Jacoby and Rubinson 1983; Lowe and Russeli 1982; Matesz and Szkely 1978). Neuromasts ais0 are innervated by efferent fibers whose ceii bodies are located in the ventral medulia (WiU 1982). Efferent axons course through the lateral line nerves. Single efferent ceiis may send axon; to the lateral line, auditory, and vestibular systems (Claas et al. 1981). The efferent system appears to hyperpolarize hair ceiis, which prevents neuromast function and precludes overstimuiation (e.g., during swimrning; I. J. Russeii 1976). The laterai line system may be responsible for maintaining the spacing between individuais in schools of Bufo tadpoles. Diminishing the sensitivity of neuromasts through exposure to streptomycin, which is toxic to hair celis, alters the orientation of individuals and the distance between tadpoles of Xenupus laevis (Lum et ai. 1982). In adult Xenopus, Elepfandt (1988a, b) .found that the oriented response to water displacements diminishes with the suraical ablation of neuromasts. " Neuromast receptor units vary in number of organs, number of hair celis per organ, and their position relative to the epidermai surface. These morphologicai features correlate weli with the particular habitat of a tadpole (Lannoo 1987) and presumably are adaptive. Generalized pond tadpoles have a large number of units composed of three to five secondary neuromast organs that are positioned flush with the epidermis. In contrast, stream and arboreal tadpoles have fewer receptor units composed solely of primary neuromasts that are positioned below the epidermai surface. In general, reductions in the number of receptor units occur within lines rather than through the elunination of whole lines, aithough Acaphus tadpoles appear to have lost their oral line (Lannoo 1987). Considerable variation in neuromast features is found within the genus Osteopilus (Hylidae). Older tadpoles of the bromeliad-dweiiing 0 . bmnneus (see Thompson 1996 for mformation on the habitat and ecology of these tadpoles) have receptor units composed of single (primary) neuromasts, while congeneric, pond-dweliing tadpoles (0. hminuensis and 0 . septentrionalis) have receptor units containing multi~le secondarv neuromasts (Lannoo et al. 1987). The lateral line system develops embryonicaiiy from preand postoptic placodes (R. G. Harrison 1904; Lannoo and Smith 1989; S. C. Smith et al. 1988, 1990; Stone 1922, 1933; see J. F. Webb and Noden 1993). These placodes give rise to primordia that migrate in anterior and posterior directions to form the neuromast organs and their innervation (review bv Lannoo and Smith 1989'). Primordia from vreotic placodes migrate anteriorly to form the cranial laterai

INTEGRATION

165

line component, and postotic primordia migrate posteriorly r t o m trunk and tail components. At least three preotic o and three postotic placodes give rise to individual nerve branches and their ganglia (Northcutt 1989; S. C. Smith A al. 1988). Neuromasts derived from preotic placodes are innervated by branches of the nALL, while neuromasts from pstotic placodes are innervated by the nPLL. Winkibauer and Hausen (1983a, b, 1985a, b) developed and tested a model for the formation of the supraorbital neuromast line in A h p u s . They proposed that stem cells are aiiocated randornly to neuromasts as they are dividing. Each stem ceii divides seven times before becoming terminal; terminal ceiis kcome hair cells. Therefore, the number of hair cells per neuromast depends on the numbers of stem and terminal cells aiiocated per neuromast. The generality of t h s model to other species, and in fact to other lines of neuromasts within ,Xytms, has not been determined. The olfaaory system of anurans is divided into two paraiiel subsystems termed the primary or main olfactory system <AIOS) and the secondary or accessory olfaaory system <-\OS).In the MOS, the olfactory afFerents innervate receptors located in the sensory epithelium of the nasal capsule and terminate on target neurons in the main oifactory bulb (AiOB; fig. 6.14). The uitrastructure of the oifactory buib and its neurons is described in K. H . Andres (1970) and Burton (1985). Neurons from the MOB projea into the lateral cortex of the telencephalon (Kemaii and Guglieimotti 1987; Northcutt and Royce 1975; Scalia 1976; Scalia et al. 1968, 1991a, b). Neurons from the AOS afferents form the

vomeronasal nerve, which innervates olfaaory receptors in the sensory epithelium of the vomeronasal organ and proje a s to the accessory olfaaory buib (AOB; fig. 6.14). Neurons in the AOB projea to the amygdaloid nucleus on the lower surface of the telencephalon (Kemaii and Guglieimotti 1987; Scaiia 1976). The AOS is well developed in anurans and caecilians but poorly developed in most salamanders. The degree of development of the AOS is positively correlated with the development of Jacobson's organ, which is absent in neotenic salamanders. The amygdaloid nucleus is correspondingly well developed in anurans and caecilians, as is the striatum (see Forebrain). The sensory epithelia of the MOS and AOS are similar to each other and to the nasal epithelium of other vertebrates (fig. 6.17). Cilia on the external surface of the olfaaory epithelium originating from the olfactory neurons are interposed by an alrnost continuous sheet of microvilli (fig. 6.17). Below the surface of the epithelium, the olfaaory neurons become grouped into olfactory iiae containing hundreds of unmyelinated axons that form a compaa olfactory nerve (Burton 1985). The main olfactory and accessory olfaaory nerves course separately into the anterior portion of the olfaaory bulb where their synapses form olfaaory g l o m e d . Each glomerulus is a spherical group of terminais from numerou primary olfactory neurons that contacts the dendrites of mitral ceiis, the principie postsynaptic neurons. Each mitral cell appears to contaa more than one glomerular dendrite. The axons of the mitral cells form the efferent projeaions of the olfaaory buibs. The olfaaory system develops from an eaodermal placode in X. laevis (Klein and Graziadei 1983). Whether the

Fig. 6.17. Developmental series (I@ to niht) and organization of the olfactos. epithelium. Note the ciliated ces separating th dark oifactory neurois. From ~ l e m Graziadei (1983); reprinted by permission of and Wiley-Liss, a subsidiaty of John Wdey & Sons, Inc., and of Kiein.

166

MICHAEL J. L A N N O O

olfactory system of pipoids, with their aquatic tadpole and adult stages, is representative of ali tadpoles remains to be determined. Likewise, the embryonic differentiation of olfactory placode tissue into principal and accessory systems remains to be described, although there is likely phylogenetic and ecological variation in the olfactory system of tadpoles. There are known variations in the structure of the internal nares and associated skin folds (Wassersug 1976a). Pond, stream, and arboreal tadpoles probably are sampling fndamentaily different olfactory environments and using their olfactory systems to mediate different behaviors. For example, pond tadpoles appear to recognize kin via olfaction (chap. 9). In adult anurans, taste buds are located on papillae or protrude only slightly from the mucosal surface and are scattered over the floor and the roof of the buccal cavity and the dorsal surface of the tongue. Taste buds in the roof and floor of the buccai cavity are nonpapillary, while those on the tongue occur on fngiform papiilae. Each taste bud papillary disc, which can contain up to 700 receptor celis depending on the species, is innervated by 5-10 myelinated nerve fibers (C. B. Jaeger and H h a n 1976). Taste buds are innervated by CNs VI1 and IX. The facial nerve innervates the roof and floor of the buccai cavity, while the glossopharyngeai nerve innervates the tongue. To my knowledge, taste buds have not been described in tadpoles but are assumed to be present. Visual System The visual system of anurans is well known (Fite 1976). What is not known is the extent that tadpoles, some of which are nocturnaily active, rely on visual cues. The extraocular eye muscles of tadpoles appear to be the same as those in adult frogs but less developed (Nowogrodzka-Zagrska 1974). Adult frogs aiso have levator bulbi muscles innervated by C N V which lift the eyes back into position after they have assisted in pushing food into the esophagus. Arnniotes focus images on the retina by changing the shape of the lens. Arnphibians and fishes focus by changing the distance between the lens and retina -proximaily for far vision and distally for near vision. Because the refractive index of water and the vitreous humor are similar, tadpoles have less need for accornmodation than do adults. The cornea is divided into an inner and an outer layer that fuse at metamorphosis (reviewed by Koliros 1981). The retina receives and processes light energy with two basic receptor types, three kinds of interneurons, and ganglion celis that project into the brain (figs. 6.18-6.19). The vertebrate retina has a number of features that are unusual from the perspectives of mechanical design and neural processing. First, the sensory surfaces of the photoreceptors face backward so that light must pass through the ganglion celis and their axons and the interneuron layer before reaching the receptors. Second, photoreceptors are stimulated by darkness; light hyperpolarizes and inhibits rods and cones. Third, the neurotransmitter released from the photoreceptors hyperpolarizes one

Fig. 6.18. Rods and cones in the anuran retina. Abbreviations: C = single cone, DC = double cone, GR = green rod, and RR = red rod. Redrawn from Donner and Reuter (1976);reprinted by permissioii of Springer-Verlag and Donner.

bipolar cell type while depolarizing the second bipolar cell type. Fourth, photoreceptors carry and process information without generating action potentials. Rods and cones are composed of three segments (fig. 6.18). An outer distal segment consists of a series of stacked disks containing the photoreceptor pigment (Corless and Fetter 1987 and Corless et al. 1987a, b, Ranapipiens). The inner segment contains many mitochondria, the nucleus, and ribosomes. The synaptic terminal occurs proximaliy and transmits the receptor signal to the interneurons, whch is the first step in sending vTsual information to the brain. The

anuran retina contains green (sensitivity peak at 433 nm) and red rods (502 nm; fig. 6.18; Donner and Reuter 1976; J. Gordon and Hood 1976; W. R. A. Muntz 1962a, b; \V. R. A. Muntz and Reuter 1966). Green rods are sensitive to violet and blue light and are known to occur only in anurans and caudates. Green rods are irnportant in light adaptation and provide inputs that conuast with those from cones. Red rods appear irnportant in dark adaptation. Anurans possess a dichromatic cone system of color vision. Single cones have a peak frequency sensitivity benvecri 575 and 580 nm, and double cones are most sensitive at 502 nm (Hailman 1976). As photons excite these visual pigrnents they become bleached; a chemicai change in the permeability of the receptor membranes affects the transmission of neural signals. In tadpoles, adult retinai pigmerits are prcceded by their corresponding 3-dehydroretinai pigments (i.e., porphyropsin rather than rhodopsin; Donner and Reuter 1976). The developmental shifi in the spectrai preference in

rods

OPL

I PL

on cent

Fig. 6.19. Conneaions of anuran photoreceptors within the retina. Abhreviations: AC = amacrine celi, BC = bipolar ceii, H C = horizontal ceU, GC = ganglion ceU, IPL = inner plexiform layer, off cent = off-center ganglion cell, on cent = on-center ganglion ceU, and OPL = outer plexiform layer. Redrawn from Badey and Gouras (1985). Copyright 1985 by Elsevier Science Publishing Co., Inc.; reprinted by permission of Elsevier, Bailey, and Gouras.

tadpoles was studied by R. G. Jaeger and Hailman (1976). Arnphibians usually lack a rnacula, an area of high visual acuity. For a discussion of the intricate biochernistry underlying phototransduaion, see Donner and Reuter (1976) and J. Gordon and Hood (1976). Rods and cones M e r in their sensitivities. In general, rods are able to detect lower light levels than cones but are poor at spatiaiiy and temporaiiy resolving this light. This is because the large photoreceptor outer pigment is more likely to be hit by photons, but this size compromises spatiai resolution. Also, rod photoreceptor pigment stacks are less polarized and therefore more sensitive to reflected linht. and rods converge onto bipolar neurons. The slower response of rods to photostirnuiation compard with cones limits their temporal resolution. In the retina the majority of synapses occur in the inner (IPL) and outer plexiforrn layers (OPL; fig. 6.19). The thin IPL is wllere rods and cones synapse, and the thick OPL is where the amacrine, bipolar, and ganglion ceiis synapse (Dowling 1976; J. Gordon and Hood 1976). Rods and cones send information to the gangiion cells either directiy through bipolar interneurons or indirectly via aniacruie, bipolar, and horizontal interneurons. Information from the photoreceptors is sent to either oncenter or off-center bkolar cells-(fig. 6.19). Bipolar ceiis are connected by inhibitory horizontal cells. When one ceii is excited it inhibits the surrounding cells of the same polaritv ( e . ~ on-center ceiis inhibit suriounding on-centei cellsj. , This arrangement increases contrast and sharpens edges. Bipolar ceiis connect to ganglion ceiis directly or indirectly through amacrine cells (fig. 6.19), whch mediate interactions between the on-center and off-center systems and have severai other speciaiized functions (Donner and Reuter 1976). Ganglion ceiis transmit information from the retinal cells centrallj into the brain through CN I1-the optic nerve and u a a . The retina (techmcally part of the brain) in Ranapipiens contains about 450,000 gangiion ceiis, 5-7 times as rnany bipolar cells and 2-3 times as many photoreceptors. Severai types of ganglion ceiis are classified on the basis of the types of information they carry; B. D. Frank and Hollyfield (1987a, b) described seven classes in R. pipiens. Stirling and Merrill (1987) detded the functionai morphology of a ganglion cell type that transmits the off-center response. Gazr and Grant (1992) described the patterns of retinal cell death during ontogeny of Xenupus hai~. retina projects The principaiiy to the contraiaterai midbrain tectum, although a small ipsiiaterai projection that appears at metamorphosis rnay aiso exist in aduits (M. Schutte and Hoskins 1993). Other retinai fibers projea bilaterdy to the diencephaiic thaiamus and pretectai visual centers (these nuclei in turn projea to the tectum). The brain also sends neurons centrifto the retina. This smaii projection (30-40 neurons, Rana) originates from the telencephalic lamina terminalis near the olfactory nuclei (fig. 6.14) and courses through the preoptic area and into the optic u a a to the retina. While the number of afferent fibers ui the optic tract increases during ontogeny, the number of efferents remains constant (H. Uchiyama et ai. 1988).
V ,

kU ay

168

MICHAEL J. L A N N O O

The iiterature on the visual orientation and feedmg behavior of anurans has been dominated by studies on adult frogs (Ingle 1976; Maturana et ai. 1960). From a phylogenetic perspective (review by Grsser and Grsser-Cornehls 1976), the retinas of adult ranids respond to cues from stationary objects, those of bufonids respond to cues from moving objects, and hylids respond to intermediate stimuli. These visudy guided behaviors correspond to ecological activity patterns (sit-and-wait vs. mobile predators) and result from differing proportions of retinal neuronal classes. Interestingly, the tadples of the species reviewed by Grsser et al. (1976; Bufo bufo, Osteupilusscptentrionalis, Rana esculenta, and R. pipicns) were d generalized pond dwellers (Orton 1944). This begs the question of whether the retinal morphology of these tadpoles reflects their common lifestyle or the more variable iifestyles of their respective adults. Retinal ceii number increases with growth in tadpoles. Beach and Tacobson (1979a., b) and P. Grant et al. (1980) , examined the patterns of ceii proliferation in the retina of Xenupus. S. Grant and Keating (1986) studied the metamorphic and postmetamorphic maturation of the retina and the midbrain optic t e m m . Dunlop and Beazley (1981) described the increasing number of ganglion cells in tadpoles through adults in Helewpoms eyrez, where a visual streak is Dresent in adults but not tadooles: Bork et al. (1987) deScribed the growth of optic fibers in the retina f X2Loptcs, and C d e n and Webster (1979) described the metamorphic changes that occur in the myelin sheath of optic nerve axons
L ,

guson 1970) and tadpoles (Justis and Taylor 1976) can orient relative to a shoreline when dowed only pineal input. In most anuran tadpoles and adults, the peripheral portion of the pineal organ is termed the frontal organ (= stirnorgan; reviewed by Adler 1976). The skin with reduced pigmentation overlying the frontal organ is c d e d the brow spot. Eldred et al. (1980) showed that the central projections of the frontal organ are similar to the ordinary pineal projections and are widespread, including the amygdala, pretectal region, and the central gray of the mesencephalon and dience~halon. Muscle Spindles are stretch receptors that provide information about the location and position of the limbs, trunk, and tail. S~indles reflexivelvconnectedto motor neurons are through a pathway termed the stretch reflex, which serves to maintain muscles at their functiondy appropriate length (e.g., E. G. Gray 1957; reviewed in Ottoson 1976); Gans and De Gueldre (1992) presented some data on the physiology of larval muscles.

Metamorphosis
The changes that occur in the nervous system of anurans during metamorphosis and the effects of thyroxine in driving these changes are summarized in Hughes (1976) and Koiiros (1981). Some conclusions can be drawn about how the nervous system accommodates the tadpole and adult stages. First, the tadpole nervous system comprises features (e.g., lateral line system and posterior spinal cord) that are exclusively larval traits lost at metamorphosis. Second, the tadpole has nervous elements (e.g., limb motor neurons and the corpus of the cerebelium) that apparentiy are not important in tadpoles but will become functional in adults. Third, the tadpole nervous system includes cells (e.g., motor trigeminal ceiis) that perform one function in tadpoles and are reworked at metamorphosis to perform a new function in adults. Findy and perhaps most remarkably, a majority of the ceiis, especidy in higher order nuclei, are not known to change during metamorphosis. These cells may function appropriately within the demands of both the tadpole and adult worlds. The brains of tadpoles are not d alike. Variation appears to be common even within the limited number of tadpoles that have been studied. Orton's (1944) tadpole types offer a starting point for comparative neurobiologists interested in the ecological and phylogenetic bases (but see chap. 4) of neuronal variation in these animais, and the diversity described in chapter 12 provides the mileposts for such investigations. Despite enormous efforts by many researchers to provide basic information about the organization and functioning of the anuran nervous system, we are only now beginning to understand how tadpoles integrate nervous system development and differentiation, and how it is altered to fit the demands of different environments. Learning by tadpoles and how this is modified at metarnorphosis is poorly studied (e.g., Munn 1940a, b; Punzo 1991; Strickler-Shaw and Taylor 1991; Youngstrom 1938).

Miscellaneous Senses Adult frogs, and apparentiy tadpoles, are sensitive to pain and touch through free nerve endings in the epidermis and dermis (reviews by Catton 1976 and Spray 1976). Cutaneous nerves, among the 60-80 myelinated neurons arising from each spinal nirve, can be g-6uped into two functiond categories: those less than 3 pm, which are also the slowest conducting fibers (0.8-5 m/s) that transmit temperature and pain impulses, and those greater than 5 pm with conduction belocities greater than 7 m/s that transmit phasic and tonic pressure information. In addition to s m d fibers with slow conduction velocities, pain and temperature receptors are characterized by a long latency from-stimulus to &scharge and by an action potential of long duration. The pineal organ (epiphysis) is a s m d projection attached to the dorsomedial surface of the epithalamus (fig. 6.15) which, within amphibians, is best developed in anurans. These fibers enter the pretectal region in the posterior portion of the epithalamus. The cytology of the pineal is described by Van de Kamer (1965). This organ functions in detecting features of ambient light, such as seasonal variation in photoperiod and perhaps the strength and inclination of sunlight, that mediate physiological functions such as breeding and ovenvintering (Bagnara 1965; Binkley et al. 1988). Locomotor activity is altered by a sudden drop in light intensity detected by the pineal organ (R. G. Foster and Roberts 1982). Hendrickson and Keiiy (1971) described the development of this structure. Frogs (D. H . Taylor and Fer-

INTEGRATION

169

Summary
There are two significant generalizations that can be drawn hom a review of the nervous and sensory systems of anuran tadpoles. Few taxa have been examined, but there are many variations exhibited even among this small number. Future studes of a more ecologically and phylogenetically &verse assemblage will most certainly turn up other interesting patterns. The second generality arises from the profound and abrupt remodeling of the neural circuitry at metamorphosis (also chap. 5). In a Me cycle that includes a profound metamorphosis from an aquatic larva to a terrestrial adult, questions about adult influences on tadpole nervous system morphology must be considered in addition to the more obvious questions of neural structure and hnction in tadpoles. Several conclusions can be drawn. First, tadpole nervous systems have elements (e.g., lateral line system) that typically function only in tadpoles. Second, tadpole nervous systems have other elements (e.g, limb motor neurons) that are not important to tadpoles but are important in adults. Third, other elements (e.g, trigeminal motor neurons) are reworked at metamorphosis to perform new functions in adults. Fourth, the majority of neurons are not known to change during metamorphosis. This latter uniformity is re-

markable when one considers the differences between a swimming, herbivorous tadpole and a hopping, carnivorous adult frog. Finally, the brains of tadpoles are not al alike; l there is no such thmg as "the tadpole brain" even within the limited number of species that have been studied. Despite enormous efforts by many researchers, we are now only beginning to understand tadpole nervous systems.

ACKNOWLEDGMENTS
I express my greatest respect and appreciation to those whose work I have cited here and apologize to those whose work I have inadvertently slighted either through oversimplhcation or omission. I thank D. Pratt, who produced the original artwork, and S. Lannoo, who redrew the figures borrowed from other sources. Acknowledgments to R. J. Wassersug, K. Nishikawa, K. Hoff, and D. Townsend (my ex-labmates in Richard Wassersug's lab at Dalhousie University) and to R. Altig, R. McDiarmid, J. Eastman, and B. Bock for their comments on earlier drafts of this chapter. This chapter is contribution number 452 from the Iowa Lakeside Laboratory and was supported by NIH Grant NS30702-01 and a grant from the Ball State University Office of Research and Sponsored Programs.

ENDOTROPHIC ANURANS
Development and Evolution
Giseile Thibaudeau and Ronald Altig

Introduction
The quote from Raff (1987) concerns the development of sea urchins, but the conspicuous parallels during the evolution of direa development in sea urchins and endotrophic anurans are significant. Endotrophic anurans obtain their entire developrnentai energy from parentai sources, most commonly vitellogenic yoik (Altig and Johnston 1989), and the developmentai patterns of these anurans deviate in many ways from those of typicai anurans. The evolutionary history of the various forms of endotrophic development and the pertinente of understanding these developmentai modes relative to anuran systematics still are debated (see chap. 11and Bogart 1981). Lutz (1948:29) made a fascinating projection: "As long as anuran taxonomy continues to be based exclusively on adult morphology they will be left aside in establishing phylogeneticai relationships." Many papers pertinent to understanding the evolution of endotrophy in anurans involve ancillary subjeas or deai with other organisrns (table 7.1) Aitig and Johnston (1989) deined the following six developmentai guiids of endotrophs: (1) viviparous (after exhaustion of viteiiogenic yoik, fetus in the ovidua feeds on oviducal materiais to complete a modified development before birth as a froglet), (2) ovoviviparous (embryo completes a modified development in the oviduct via only oogenic energy sources and is birthed as a froglet; see Blackburn 1994 for opinions on terminology), (3) direct developer (embryo completes highly modified development via oogenic energy sources in an oviposited egg that is not intimately associated with a parent's body and hatches as a froglet), (4) paraviviparous (ernbryo completes a modified development via oogenic energy sources in a site other than the reproductive traa of the mother and is "born" as a froglet), (5) exoviviparous (embryo develops via oogenic energy sources in a terrestriai egg before the hatchhg moves to a site usually in or on the maie parent's body and a froglet eventually is birthed from that site), and (6) nidicolous (eggs oviposited terrestrially, and embryo develops from oogenic energy sources to produce various sons of free-living, nonfeeding larvae). The nidicolous and direa development modes are more common than other endotrophic guiids. Diversity is highest among the nidicolous group where various developmentai patterns provide a range of morphotypes; at the unnlodified end of this continuum an entire tadpole morphotype is produced. Near the other end of the continuum, there are various forms of embryos with progressive losses of typicai tadpole features. If endotrophy occurs in a presently recognized genus, usually all species involved are in one guild; exceptions

While it might seem strange to focus on a kind of evolution that fails to &ect the morphology of the adult, it is precisely because of this faa that direct developing forrns offer such superb experimental systems to examine the modifications possible withm ontogenies that acheve the same end by dXerent trajectories.
Raff 1987:6

E N D O T R O P H I C ANURANS

171

include Eleutherodactylus (teirnatobine leptodactylid, 2 endotrophic guilds), Gastrotheca (hemiphractine hylid, 1 endotrophic and 1 exotrophic guild; Matthews 1957), Mantidadylus (manteiiine rhacophorid, 1 endotrophic and severai exotrophic guilds), and Mgophys (megophryid, 1 endotrophic and 1exotrophic guild). Arnong microhylids, endotrophv occurs in 6 of 10 subfamilies and is ~ervasive 4 of in thiseiall brevicipitine (Africa) and asterophbnine and genyophrynine (Australopapua) frogs have direct development. All cophylines (Madagascar) are nidicolous with similar morphologies, aithough embryos of Stump* seem to have a different developmentai trajectory. These observations should not be used to imply anything about the evolution of endotro~hv within these lineages. " The characterization of a taxon as endotrophic often is based solely on the presence of a small clutch of large, usually nonpigmented ovarian eggs. Based on known relationships benveen egg size, pigmentation, and the occurrence of tadpoles, this criterion should be employed cautiously. Even though more than 1,000 species in more than 90 genera and 11 famrlies are involved and data have appeared over a span of more than 225 years (e.g., Bavay 1873; Bello y Espinosa 1871; Fermin 1765; Spallanzani 1785), developmental data on endotrophic anurani generally are sparse to absent. There has been little standardization in terrninology, staging systems, and information presented. The majority of this information is derived from~hutherodactylus (telmtobiine leptodactylid), Gastrotheca (hemiphractine hylid -even though
1 ,

the most cornmonly studied species in fact produces tadpoles) and Nectophrynoies (bufonid). The only complete staging table of an endotroph is based onE. coqui (Townsend and Stewart 1985). and considerable data on other Eleutherodadylw make th developmental patterns of this genus the best understood. Salthe and Mecham (1974) suggested that direct development of Eleutherodactylus may represent the most advanced type of endotrophy, although this idea may only reflect the paucity of data for other taxa and our understanding of the evolution of the various forms ofendotrophy. Primarily through the work of Eugenia De1 Pino (e.g., 1975, 1989a, 1996; De1 Pino et ai. 1992) and her colleagues, the unique development of hemiphractine hylids, based primarily on the genus Gastrotheca, is aiso well documented. The first part of this chapter is a collation of mformation on the development of endotrophs (tables 7.2-7.3; figs. 7.1-7.2) based on a common terminology and the staging systems of Townsend and Stewart (1985; direct developers) and Gosner (1960; nidicolous larvae). In table 7.4 we standardize stages of Townsend and Stewart (1985) with three other systems involving Eleutherodactylus. Townsend and Stewart (1985:432, table 3) compared stages and developmental times of eight other species of Eleutherodactylus with those of E. coqui. Stages of Gastrotheca hbambae (De1 Pino and Escobar 1981) can be standardized against Gosner stages (table 7.5). In species of Gastrotheca that do not have tadpoles, a modified staging scheme is needed at least for

Table 7.1 Summary of anciary subjects, often pertaiiung to organisms other than amphbians, that couid have a notable impact on various studies of endotrophic anurans Subject Developmental regulation Evolution of viviparity Evolution of viviparity and egg retention Hybridization of ovo- and viviparous lizards Developmental vs. evoluuonary rates Fertilization mode and parental care Parenta1 care in Eleutheroda~lz~s Froglet transport in Eleutheradaqlus Reproductive energy docation/viviparous Lizard Multigene families regulate development Ontogeny/phylogeny relationships Reproductive strategies in heterogeneous habitats Regulatory geries and development Evolution of viviparity in fishes Evolutionaq loss of feeding larvae in an echinoderm Evolution of larval types in invertebrates Developmental energy use in echinoids Maternal-fetal oxygen transport in lower vertebrates Evolution of viviparity Immunological/endocrinological problems in viviparity Maternal body volume constrains viviparity Costs of viviparity in scorpions New hypothesis for evolution of viviparity Loss of larval types in invertebrates Evolution of direct developmeiit in salamanders Evolutionaq significance of fetal maintenance Phylogenesis of direct development and viviparity Evolutionaq scenario of viviparity Evolution of oviducd viviparity and egg retention

Referente
Alberch 1982,1985,1987 Amboroso 1968 Andrews and Rose 1994 Arrayago et al. 1996 Arthur 1982 Beck 1998 Bourne 1997, 1998 Diesel et al. 1995 Doughty and Shtne 1998 Dressler and Grusse 1988 Fink 1982 Giesel 1976 Gluecksohn-Waelsch 1987 Guiiiette 1987 Hart 1996 Havenhand 1993 Hoegh-Guidberg and Emlet 1997 Ingerman 1992 Mathies and Andrews 1996 Medewar 1953 Q u d s and Shine, 1995, 1996 L. R. Shaffer and Formanowicz 1966 Shtne 1995 Strathmann 1978 D. B. W&e and Hanken 1996 M. H . Wake 1977 M. H. Wake 1989 M. H . Wake 1993b M. H . Wake 1993c

172

GISELLE THIBAUDEAU AND RONALD ALTIG

Table 7.2 Alphabetical list of f d e s and genera including species (number of species in parentheses) known or suspected of being endotrophs. n More detailed data for some taxa appear i table 7.3. Arthroleptidae Arthroleptinae: Avthroleptis ( 12) Bombinatoridae: Barbounrla (2) Brachycephalidae: Brachycephalw (2) and Pyllaphyne (1) (part, 3 out of 217 species may be endotrophs, at least under some circumstances)CrepiBufonidae:Altiphtynoides ( l ) ,Andinuphtyne (3), BU$J +h~yne ( 1), Didynamipus (1), Frostiw (l), Laurentophtyne ( l ) , Metaphtyniscus (1), Nectophtyne (2), Nectophynoides (5), Nimbaphynoidez (2), Oreophymlla (6), Osmnophtyne (6), Pelophlyne (8), Rhamphophph-e (8), Tmebella (2), and WokentoIffinu(2) Dendrobatidae: Colostethus (part, 3 out of 90) Hylidae Hemiphractinae: Ctyptobatrachus (3), Fectonotus (5, feeding occurs under some circumstances in some species), Gar~votheca (part, 27 out of 43 species),Hemiphracbus (5), and Stejania (7) Leiopelmatidae:Lehelma (3) , . Lep~odactylidae Leotodactvlinae:Adenomem (oart. unknown number out of 6) and Vanwlinius (1) ~eimatobihae: Adelophtyne ~tycholos Cycloramphus (1 out of 25 species):~ischidodactvlzbs(2), Ekutherodactylus (517), EuparkereUu (4), (2), Eupsophus (8), Geobatrachus (I), Holoaden (2), Ischnomema (4), Phlynopus (20),Phyllonartes (4), Phyzelaphtyne (I), and Z d m n w (3) Megophqidae: Mgophtys (part, 1 out of 24 species) Microhylidae Asterophryinae:Asterophtys (I), Batyoenys (7), Callulops (13), Hylophorbus (I),Mantophtyne (3), Pherohapsr ( l), Xenobatrachus (17), and &mrhina (6) Brevicipitinae: Baebreviceps (I), Breviceps (13), Callulina ( l ) , ProbrePiceps (3), Spelmophryne (1) Cophyiinae:Anohnthyla (4), Cophyla ( l ) ,Madeccmophtyne (I), Platypelis (9), Pbthohntohyla (14), Rhombophyne (I), and Stump@ (6) Genyophqmnae: Aphantophlyne (3), Copiada (5), Choerophtyne (I), Cophixalus (28), Genyophtyne (l), Oveophtyne (24), and Sphenophlyne (24) Melanobatrachinae:Melanobatrdus (1) and Parhoplophtyne (1) Microhyhae: Adehes ( l ) ,Alberictls (3), Arcmomer ( l ) , Gastrophynoidez ( l ) , Hyophtyne (l), kzluphtynus (lO), Myemiella (I), PhyneUu (I), Synaptumnus (3), and Syncope (2) Myobatrachidae Limnodynastinae: Kyaranus (3) and Philmia (1) Myobatrachinae:Arenophtyne (10),Rrsa (I), Btyobatrachus (I), Geocvinia (7), Myobatrachus (I), and Rheobatrachus (2) Pipidae Pipinae: Pipcr (part, 1 out of 7 species) Ranidae Petropedetinae: Anhydroph~e and Avthroleptella (6) (1) Raninae: Batrachyiades ((8, Ceratobatrachus ( l ) , Discodeles (5), Palmatorappia (I), Phtynohn (I), Plntymantis (42), and Tayiurana (2) Rhacophoridae Manteiiinae:Mantzdactylus (part, 6 and 7 i each of two species groups out of 61 species) n Philautinae: Philautus (87, unknown number are endotrophs) Rhinodermatidae: Rhinodma (2) Sooglossidae:Nesomantis (1) and Sooglossus (2)
L

(T),

later stages. We focus on developmental, anatomical, and physiological aspects of endotrophy that have been largely ignored in previous surnniaries of anuran reproductive biology (e.g., W. C. Brown and Alcala 1983; Crump 1974; Dueliman 1985; Salthe and Dueliman 1973; Salthe and Mecham 1974; Tyler 1985). Sufice it to say that endotrophs typicdy produce fewer, larger eggs than similar-size frogs that produce tadpoles. Discussions of breeding moda (e.g., De1 Pino 1980a; De1 Pino and Escobar 1981; Dueliman and Gray 1983; Dueliman and Maness 1980), the evolution and morphology of the brood pouches of egg-brooding hylids (e.g., De1 Pino 1980b; De1 Pino et ai. 1975; R. E. Jones et al. 1973), and parental care (e.g., Taigen et al. 1984; Townsend 1986, 1996; Townsend et ai. 1984) are omitted. Taxonomic names used here are from D. R. Frost (1985) and may differ from those found in the published papers. Citations that pertain to specific species are in table 7.3. The second part of this chapter is a heuristic discussion of developmental trajectories and related topics concerning the evolution of endotrophy; parts of chapter 11 also are pertinent.

Development of Endotrophic Anurans

Because the direct-developing Eleutherohqlns cogui has been studied extensively (e.g., Townsend and Stewart 1985), data for this species are summarized first, and data for other endotrophs are added in comparison. The study by Lynn (1942; Eleutherohqlus nnbicola) provided the most anatomical details of an endotroph. Ehson et al. (1990) summarized considerable information on endotrophc larvae including culturing methods. The incomplete information on endotrophs is obvious in the foiiowing text. Gametes and Fertilization Eggs of endotrophs usudy are larger than those of similarsize frogs that produce tadpoles and are laid in various sites with sufficient moisture (e.g., Estrada 1987; Prez-Rivera and Nadal 1993). Parental care is common. Much of the research on oogenesis and egg components (i.e., yolk densis; meiotic mechanisms, RNA components) of endotrophc eggs was based on hemiphractine hylids. The 3-mm eggs of Gastrotheca riobambae (produces tadpoles) have lower ribosomal gene ampiification, fewer nucleoli (about 300), less

ENDOTROPHIC ANURANS

173

Table 7.3 Summary of breeduig biology of endotrophic anurans. Enig = ecomorphological guild (Aitig and Johnston 1989): DD = direct developer, EV = exoviviparous, NI = nidicolous, OV = ovoviviparous, PV = paraviviparous, and VI = viviparous; Mod = mean ovum diameter, mm; Clutch = eggs/dutch; Deve = developmentai time, days; Site = developmentai site: UO = under objects, IB = in burrow; ES = elevated sites; OV = ovidud-ovoviviparous, IS = in stomach of femaie, BS = in back skin of female, 0 B = on back skin of femaie, IP = in partiai or complete dorsai pouch of female, VI = oviducai-viviparous;VS = in maie vocal sac, ID = in dorsal pouch of maie, and OM = on back of maie; Hatch = hatchling SVL, mm; and Reference = a pertinent referente. Taon ..irthroleptidae, Arthroleptinae Arthrokptis msculum poecilonotuc wahlbmgi Bufonidae Altiphrynoides maicomi Didynamipus sjoestedti Laurentophyne parkeri Nectophyne afia Nectophymides tomeiri Nimbaphvides ocddentalis Omoph7yneUa n&va Pelophyne brevtpes Dendrobatidae Colostethus chakopis stephani Hylidae, Hemiphractinae C~toba~dchus jkhmzanni Fk-ctomtus fitzgmaldi goeldii Emg Mod Clutch Dev, Site Hatch

Referente

DD DD DD NI DD DD NI OV
VI

Guib and Lamotte 1958 Lamotte and Perret 1963b Wager 1986

H. H. Wake 1980
Grandison 1981 Grandison 1981

Orton 1949 Lamotte and Xavier 1973 McDiarmid and Gorzula 1989 Aicaia and Brown 1982 Kaiser and Altig 1994 Junc et ai. 1994 Ruthven 1915 D u e h a n and Gray 1983 Weygoldt and De Carvalho e Silva 1991 Heyer et ai. 1990 D u e h a n and Maness 1980 Del Pino 1980b Del Pino 1980b Del Pino 1980b De1 Pino 1980b De1 Pino 1980b De1 Pino 1980b D u e h a n and Maness 1980 D u e h a n and Maness 1980 Archey 1922 Beli 1978 N. G. Stephenson 1955 Heyer and Silverstone 1969 Heyer et al. 1990

DD NI NI NI PV NI

hI T

ohausi NI flgmaeus NI Gastrotheca cwatophryes PV christiani PV ernestoi PV mfira PV testudinae PV weiniundi PV Hemiphractus scutatus PV Stejania scalae PV Leiopelmatidae Leiupelma archeyi EV hamiltoni EV hochstetteri NI Leptodactylidae, Leptodactylinae Adenomera hylaedaqh NI mamorata NI Leptodactylidae,Telmatobiinae Eleutherodaqlus abbotti DD atkinsi DD aua'unti DD augusti DD cooki DD

A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 Valett and Jameson 1961 A. Schwartz and Henderson 1991

174 Table 7.3 contznued Taxon

G I S E L L E T H I B A U D E A U A N D R O N A L D ALTIG

Emg

Mod

Clutch

Dev,

Site

Hatch

Reference Townsend and Stewart 1985 Heatwole 1962 A. Schwartz and Henderson 1991 A. Schwara and Henderson 1991 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 M. H. Wake 1978 Ovaska 1991 L. Adamson et al. 1960 Lynn and Lutz 1946b Lynn 1942 Lutz 1944a A. Schwartz and Henderson 1991 A. Schwartz and Hcnderson 1991 Gitlin 1944 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 A. Schwartz and Henderson 1991 Formas and Vera 1980 Cei and Capurro 1958 Formas and Vera 1980 Lutz 1944b

coguz co~ltutus dimidiatus fmleri gossei hedricki inoptatus jasperi johnstonei martinzcens~ nnrutus nubhla pams pa"zae planirostris pwto&nsic thorectes parians varieyi wetmorez Eupsophus roseus taeniatw vittatus Zachaenus pamltrr Microhylidae, Brevicipitinae Breviqs dpemus Microhylidae, Cophylinae Anodonthyla bouiengeri

DD DD DD DD DD DD DD OV DD DD DD DD DD DD DD DD DD DD DD DD
NI NI NI

DD DD
NI

Blommers-Schlosser 1975 Glaw and Vences 1992 Glaw and Vences 1992 Blommers-Schiosser 1975 Blommers-Schiosser 1975 Glaw and Vences 1992 Glaw and Vences 1992

phylioahqla NI 2Mademsophyne truebae Platypelis grandis NI Piethodontohyla notostim NI Rbombophyne testudo NI Stumpjia grandis NI Microhylidae, Genyophryninae Cophixalus parkeri DD Oreophyne annulata DD Microhylidae, Microhylinae Calaphynus pleurostgma NI Myeniella micvops DD Synapturanus salsmen NI Srncope antenmi NI Myobatrachidae, Limnodynastinae Krarranus sphagnhlus NI Philoria j?osti NI Myobatrachidae, Myobatrachinae Arenophyne rotunda DD

W Y ~

W C. Brown and Alcala 1983 Berry 1972 Izecksohn et al. 1971 Pyburn 1975 Knigel and Richter 1995

De Bavay 1993
Littiejohn 1963

E N D O T R O P H I C ANURANS Table 7.3 continued Taxon Emg Mod Clutch Dev, Site Hatch Reference Ehmann and Swan 1985 J. D. Roberts 1981 Tyler and Davies 1983

175

Assa darlingtoni Myobatrachus gouldii Rheobatnubus silw Pipidae, Pipinae Pipa pipa Ranidae, Petropedetinae Anhydruphqwe ratlayi Arthmleptella hewitti lightfooti Ranidae, Raninae Ceratobatrachus guentheri Platymantis guentheri hamiue meyeri Rhacophoridae, Philautinae Phiuutus llcsobrachius schmackeri Rhinodermatidae Rhivwdemza danvinii Sooglossidae Sooglosus gardim'

F. Schtte and Ehrl1987

Warren 1922 De V i e r s 1929c Wager 1986 specimens Aicaia and Brown 1957 Aicaia 1962 A i d a 1962 Aicala and Brown 1982 A i d a and Brown 1982 Jorquera et ai. 1972 specimens

18 and 28s rRNA, and slower developmental rates (2 weeks to complete gasmlation) than those of Xenopus l m i ( 14 h; De1 Pino et al. 1986). The degree of rRNA amplification is generally low in the slow-developingherniphractine hylids, (De1 Pino and Humphries 1978; but Flectonotuspy~rnaeus Macgregor and De1 Pino 1982) has about 2,000 meiotic oocvte nuclei with low levels of rRNA am~lification and the embryos develop more rapidly. T h s results from about 11 rounds of oogonial mitoses occurring in ovarian chambers or cysts (De1 Pino and Humphries 1978) in which nuclear diviiion occurs without cvtokinesis: such a wocess ~roduces fewer eggs with more nutrients and other components (e.g., high concentrations of rRNA, C = 1.7 x 10-l2g) necessary for endotrophc development. The nuclei eventually segre. gate into a-peripherdband of large nuclei and a central group of smaller ones. The central group and ali but one peripheral nucleus disintegrate, and that peripheral nucleus, ~ e r h a ~ s one with the hizhest DNA content. remains. the " The amount of nuclear DNA present may also affect developmental rates. The mean Au/nucleus for 9 species o f E h t h erodactylus is 10.722.2 (7.9-15.2) versus 13.224.6 (4.526) for a number of other species of frogs (Goin et al. 1968). Mantellines have 8.3-13.3 pg of DNA per diploid nucleus (Blommers-Schlosser 1981), and development in ali members of two species groups is thought to be endotrophic.
1 1

Oocyte maturation (De Albuja et al. 1983) is assumed to be mononucleate (De1 Pino 1989b; Nina and De1 Pino 1977, Eieutherodaqlus; Gall 1968, E. johnstonei; Nina and De1 Pino 1977, E. unistrqgatus (have many large nucleoli in the germinal vesicle). De1 Pino (1989b) suggested that oogenesis of Eleutherod.rylus may be modified to allow accumulation of RNA and proteins to accommodate faster development of the large eggs. Sperm motility is high in low tonicity conditions, and sperm enuy is limited to about 10% of the surface of an Eieutherodcrctylw egg (Elinson 198%). Polyspermy may occur with normal development, and ai nuclear events of the l first ceU cycle occur near the animal pole (Gitlin 1944; Goin 1947; Sampson 1904). In Gastrotheca riobarnbae a translucent spot containing the single nucleus is visible at ovulation (De1 Pino et al. 1986). Cleavage is assumed to be holoblastic and unequal (Elinson 198%), although meroblasty has been suggested (De1 Pino and Escobar 1981 and De1 Pino et al. 1975, G. riobarnbae) or reported (De V&ers 1929b, c, Arthroleptella and Breviceps; Inger 1966, Philautus hosii; Warren 1922, Anhydrophtyne rattrayi). Warren (1922) reported that was the animal pole ofdnhydrc~phtyne syncytial. Subsequent stages through neumlation probably follow the typical anuran pattern (De1 Pino and Elinson 1983; Elinson 198%). Gastdation in Kvawanussphagnicolus, an unmodified nidic-

176

GISELLE THIBAUDEAU A N D RONALD ALTIG

E N D O T R O P H I C ANURANS Fig. 7.1. Embryos and hatchling o f E h t h e m ~ L ~ m q u i . are Stages fiom table of Townsend and Stewart (1985). (A) Embryo, dorsai view, head upward, limbs buds visible, Townsend/Stewart 4. (B) Embryo, lateral view, removed fiom egg capsules, Townsend/Stewart 6. (C) Embryo, ventral view, within egg capsules, Townsend/Stewart 12. (D) Embryo, lateral view, within egg capsules, Townsend/Stewart 12. (E) Embryo, stained with a rnonoclonai antibody for Type I1 coiiagen, most dark areas are cartilage, Townsend/Stewart 9. (F) Hatchling, lateral view of head, stained for cadage.

177

olous form, resembles that in Ranu (J. A. Moore 1961), although the eventual neurai plate appears narrower and shorter.

Early Emlnyology
Hemiphractine hylids foiiow a nove1 pattern of early development. Although Gastrotbeca riubambae eventuaiiy produces a tadpole (De1 Pino and Eiinson 1983; De1 Pino and Loor-Vela 1990; Eiinson and De1 Pino 1985), an embryonic disc forms as a result of gastruiation. The embryo proper is derived from this disc in a pattern similar to that seen in birds and reptiles (see Gatherer and De1 Pino 1992). Involution is limited, and the primitive gut is lunited to a smaii cavity at one end of the embryo. In most genera the yok

Fig. 7.2. Embryos of different endotrophic fiogs. Stages are fiom the table of Townsend and Stewart (1985), except in (D). (A) Ccratubamzchw&u~nthcn' (Ranidae, Raninae), lateral view, Townsend/Stewart 13. (B) Gmtheca sp. (Hylidae, Hemiphractinae), venaal view, beii giiis (Microhylidae, visible, Townsend/Stewart 15. (C) Copi&-m

Genyophymae), removed from egg capsules, lateral view, Townsend/ Stewart 14. (D) Embryos and nidicolous tadpole of Plcthodontotryla ing~inalis (~icroh~lide, Cophylinae), intra&ipsular embryos, Gosner stage 20; nidicolous tadpole, Gosner 27.

178

GISELLE THIBAUDEAU A N D RONALD ALTIG

Table 7 . 4 Staging systems of Lynn (1942, Eleutherodactylus nubbla), Lynn and Lutz (1946b,E. nasutus), and Sampson (1904, E. martinicensir) standardized against that of Townsend and Stewart (1985, E. coqui) Townsend/Stewart
1 2 3 4 5 6 7 8 9

Lynn/Lutz

L~
24 23.5
-

Sampson
-

I
-

Body, Lirnbs, and Tail VI The lateral body wail appears between the limbs at TownVI1 send/Stewart 7 as a small convex, pigmented border. This vIn border grows ventrally to surround the yolk mass through 10 Townsend/Stewart 12 (Goin 1947, Eleutherodactyusplani11 rostrzi; Lynn 1942, E. nubicoh; Lynn and Lutz 1946a, E. 12 IX pentheri). The four limb buds appear in Eleutherodactylus X embryos soon after the neural groove closes in Townsend/ XI Stewart 4 and before gdi buds appear. Derived from lateral XII plate mesoderm, the rounded buds lie lateral to and seem13 XIII 14 ingly are not conneaed to the presumptive trunk.By Town15 send/Stewart 5, limb buds have increased in size and visibly hatch XIV Hatch connect with the trunk. Elbow and knee joints appear at Townsend/Stewart 6, and foot paddles have formed by Townsend/Stewart 7. Digits begin to differentiate in TownTable 7.5 Comparison of the staging system of De1 Pino and Escobar (1981) for Gastvothecariobambae with that of Gosner (1960) send/Stewart 8, and during stages 9-13 the limbs and toes elongate and toe pads eventuaily form. The front legs usuaily Event Gosner De1 Pino/Escobar are folded mediailv below the head. while the hind limbs are flexed ventrallykr ventrolaterailyhong the body. Elinson Fertilization Gray crescent (1994) showed that proper leg development of E. coqui may 2 cell be controiied by a systemic factor and an influence of thyroid 4 cell hormone could not be demonstrated (also Lynn and Pea8 cell don 1955). 16 celi Limb buds of other taxa may appear sirnultaneously (e.g., 32 celi Midcleavage Arthroleptis, Eleutherodactylus, Leiopelma, Nectophywides torLate cleavage nieri, Philautus, and Pipa), the front legs may develop earDorsal lip lier than the hind legs (e.g., Breviceps~scus), the hind legs or Midgastruia may appear earlier than the front legs (e.g., Anhydrophlyne Late gastruia Neurai plate rattrayi). When larvae of Flectotonotus are released from the Neurai folds pouch, 'their hind legs are comparable to those of a tadpole Rotation at Gosner 39. Alcala (1962) noted that the advanced embryo Neurai tube of Platymantisguentheri with large front limbs and no hiAd Tail bud limbs or tail reported in Aicala and Brown (1957) was malMuscle response formed. Heart beat Except for hind limb development, the relatively unmodGi circulation ified nidicolous larvae at advanced stages (e.g., Gosner 30Clear cornea 36) closely resemble much younger larvae of exotrophs (e.g., Fin circulation Operculum begins 21-23) in proportions and shape. The hind limb buds are Operculum continues proportionately and absolutely huge for the size of the larva Operculum closes (Junc et al. 1994; Kaiser and Altig 1994). Limb < 0.5 x diameter In Eleutherodactyluscoqui, a taii bud appears at Townsend/ Limb 2 0.5 x diameter Limb 2 1 X diameter Stewart 4, the tail elongates and membranous fins develop Limb 2 x 1.5 diameter during stages 6-10. The fins begin to regress after TownLimb = 2 X diameter send/Stewart 12 and are entirely resorbed within 2 days of Toe development h a t c h g . The taii curls ventraily between the h d legs and

23 22 21 19 18 17 16 14 13 12 11.5 11 10.5 10 9 6

I1 I11

Iv
-

mass remains roughly spherical throughout development, but in Ceratobatracbus (personal observation) and Platymantis (Aicala and Brown 1957; Atoda 1950), the yolk starts to partition along the ventral midline, most pronounced anteriorly at the beginning of the process, at about Townsend/ Stewart 6. The taii extends anteriorly in this sagittal depression, and the invagination increases progressively untii by about Townsend/Stewart 13 the yolk mostly resides as m o ventrolateral masses. Mourv and Hanken (1995) is the sole paper that addresses neural crest migration in a direa developing frog.

ENDOTROPHIC ANURANS

179

along the venter or lateraiiy along the side of the ernbryo, tooth has not been reported frorn rnernbers of any other geusuaiiy to the lefi. In other species, the tail and ins are dis- nus. Advanced embryos of Stefaniagoini have a srnaii, triantorted so as to appear somewhat like a rnushroorn whose cap gular, keratinized tubercle in the middle of the upper jaw covers the posterior hemisphere of the ernbryo. A section of that Dueiiman and Hoogrnoed (1984) thought might be the tail inat about midlength appears firrnly adherent to the a remnant of the upper jaw sheath. Based on the location inner surface of the egg jelly in E. inoptatus (personal obser- of the tubercle and the age of the ernbryos, this structure vation). Other species of Eleutherodactylus (e.g., E. hazelue may be an egg tooth. An unidenthed leptodactylid ernbryo and E. meyeri) and frogs in severai other guilds (e.g., Asa, (R. W. McDiarrnid, personal cornmunication) has an apparCeratobatrachus, several Gastrotheca,Myobatrachus, Nectophry- entiy anaiogous keratinized structure on the symphysis of noides tornieri) have srnaii tails with ins srnaii to absent. the protruding lower jaw. Embryos of Discodeles opisthodun lack a tail, and there appear to be no data to support the staternent of Rostand (1934) that the tail of Nectophrynoides is hollow. The tail fins of di- No vestiges of tadpole rnouthparts are known in E. q u i or rect developers are thin and nonpigmented, and the blood other Eleutherodactylus, aithough a detailed developrnental vessels are obvious. Whether or not there is a suficient in- study of the rnouth area is lacking. Vestiges of oral structures crease in vascularity to suggest a specific respiratory func- are present in some hemiphractine hylids (Wassersug and tion for the tail is debatable. De Vilhers (1929b) found no Dueliman 1984), Arthrolepteliu (Lambiris 1988a, b), and specialized vascularization in the fin of Arthro&ptella and Nimbaphqwoides occidentalis (Lamotte and Xavier 1972); the concluded that gas exchange is enhanced only by increased iatter u&ors sunnested that the rnouth~arts fetuses of in surface area. Warren (1922) stated that embryos of Anhy- Nectuphrynoides were somehow used in feeding in utero. drophryne ratcrayi developing from smaiier eggs had larger Arthrolepteliu has d r a - and suprarostal cartilages (Lambiris tas than embryos developing from larger eggs. Even 1988a, b; Van Dijk 1966), and the occurrence of rnouthpart though Townsend and Stewart (1985) intirnated poten- remnants in this genus agrees with Elinson's ideas (1990) tiai ecological, developrnental, and systernatic correlations that the typical lack of rnouthparts results frorn the absence among the number and size of gilis and tail morphology, of the proper inducing sources. Embryos of Cycloramphus dehtive data are lacking. stejnegeri have an aitnost complete oral apparatus, and those of Rhinoderma have nascent tooth ridges on the lower laIntgument and E Tooth m bium (Cei 1987).In direct-developing species of Gastrotheca, IMost endotrophic eggs and ernbryos are nonpigrnented and pigmented rnouthparts forrn but as development proceeds appear white, yellow, or orange. Those of Pipapipa and some they undergo changes associated with metamorphosis and nidicolous forrns are exceptions, and Noble (1917) noted produce the mouth configuration of a frog (Wassersug and that early ernbryos of Cryptobatrachussfhrmanni are black; D u e h a n 1984). Nectophylnoides tomieri has a ridge at the based on some of our observations, we suggest that ths col- place the lower labium Godd forrn in an exotrophi~tadpole, oration is a preservation artifact. Ernbryos of Ebutherodac- and the sides of the upper jaw overhang the lateral parts of ~Lus mqui and probably rnost other taxa begin to deposit the lower jaw (Orton 1949). Most nidicolous cophyiine larpigment on the head and along the dorsal rnidlme by Town- vae have reduced. soft rnouth~arts resemble those of the that send/Stewart 5. In subsequent stages, chrornatophores exotrophic scaphiophrynine Scaphiophiyne; larvae of Stumpdifferentiate lateraiiy but do not go beyond the lirnits of the fia do not have such rnouthparts and mouth developrnent body waii. By Townsend/Stewart 12 aii but a sliver of yolk proceeds more directiy toward that of a frog. is enclosed by the body waii, and the entire yoik rnass is pigEiinson (1990) suggested that the lack of rnouthparts rernented by stage 13. Pigment is rnost intense dorsaiiy and sults frorn the absence of the inducing cartilages or ectoderlateraiiy, and characteristic patterns on the head and legs are mal cornpetence, or both, but it is obvious that detailed, evident by Townsend/Stewart 14. As a generality, the adult corn~arativedevelo~mentdstudies are needed to underpattern is present by hatching in direct developers but ap- stand the developmental patterns more effeaively. Adhesive pears later in other endotrophic froglets. The tail muscle and glands apparentiy are lacking in aii nonnidicolous endoins are typicaiiy nonpigmented. trophs and often are nonfunctional in nidicolous forms (e.& In species of E&utherodugvLus, a single keratinized egg Cei and Capurro 1958; Formas and Pugn 1978; Formas tooth forms on the skin adjacent to the premaxillary syrn- and Vera 1980, Eupsophus roseus and E. vittatus; Akala and physis late in Townsend/Stewart 12. The conical tooth proj- Brown 1957, Pephryne brepipes).The less-modifiedernbryos ects verticaiiy from its insertion and terrninates in a single or of Philautus (Aicaia and Brown 1982) and Philoria (Littlebiid point (Hardy 1984; Lutz 1944a; Noble 1926a). The john 1963) have funaional adhesive glands. single point of the conical egg tooth of E. urichi from Trinidad projects anteriorly. Differences in the number of points Gills, Operculum, Spiracle, and Respiratoly Systems and the projecting direction suggest the employrnent of Pharyngeal slits apparentiy are absent in ali endotrophs. One different motions during hatching or differences in the char- or two pairs of smali gilis with fimbriae present as not much acteristics of the egg jelly. Except for the ranid Discodeles opzs- more than terminal knobs and bumps are present for less thodun (Alcala and Brown 1957; Boulenger 1886), an egg than one-third of development in E. coqui, and the number
VV

180

GISELLE THIBAUDEAU A N D RONALD ALTIG

of gills (0-2) and arches involved vary among species. Gill buds with circulating blood appear at Townsend/Stewart 5, reach maximum extent by stage 7, and regress by stage 9. Most Nectophpides lack gills, but N. tomievi has two pairs. Atoda (1950) and Boulenger (1886; see DueUman and Tmeb 1986; Noble 1927) thought that a pleated area (= abdominal sacs of Alcala 1962) on the lateral body wall between the legs of Ceratobatrmhw, Discodeles, and Plutylnctntis serve as a respiratory surface. Our observations of the development of Ceratobatrachwfiuentheviagree with those of Noble (1925b, 1927) who suggested that these pleats are simply wrinkles in the lateral body wall that develop as the yolk reserve is depleted. The unique bell gills of hemiphraaine hylids exist in several configurations. Not long after neurulation is complete, two pairs of anlagen are visible immediately behind the presumptive head (Del Pino et al. 1975; Noble 1927). The anlagen do (e.g., Gdrtrotheca) or do not fuse to produce the 1-2 pairs of functional gills attached to the first or second branchial arch, or both ( D u e h a n and Gray 1983). The gills develop into concave, vascularized discs or beil-shaped cups, lie in intimate contact with the egg jelly (Del Pino 1980b; Del Pino et al. 1975), cover 50%-100% of the embryo, and the right and left sides may overlap. Analogies to placental extraembryonic membranes have been suggested (e.g., Noble 1917; VaIetc and Jarneson 1961). One or two gill stalks, each containing efferent and afferent vessels, extend between the top of the bell and the first and second aortic arches of the embryo. Noble (1917) and I. Spannhof and Spannhof (1971) noted that the strands of single-stranded gills armally are composed of two tightly wrapped strands. Striated muscle cells may pull the gill into the gill chamber at birth (Noble 1917), although Del Pino et al. (1975) noted that circulation to the @s ceases soon after birth and that the gills are resorbed w i h 24 hr in G m t h e c a Itiobambae. Embryos cultured in Ringer's solution retain gills. The development of the aortic arches is relatively simple in Eleuthemdactylus coqui, and the vascular network of vessels extending around the yolk mass are based on one or two large vessels that enter near the branchial arches (see Lynn 1940b). Opinions concerning the operculum and spiracle (e.g., presence, configuration, homology among endotrophs and between endotrophs and exotrophs, and terminology) in endotrophs are codksing. By definition, a spiracle cannot form if the operculum is absent or does not form to a similar extent as it does in typical tadpoles. Several authors stated that an operculum is absent in Eleutherodact_rlus, but lateral folds (= dermal folds, branchial or gular fold) continuous with the skin of the head and sides of the body cover the base of the forelimb (Lynn 1942). These folds remain separated and never enclose the entire Limbs or produce a spiracle. Sampson (1904) stated that these folds were not homologous with the operculum of typical tadpoles. Warren (1922) noted that inAnhydmphryne a fold grows forward over each limb from behind, but he did not discount that these folds were homologs of the true operculum even though shifted in position. Leiupelma hochstetteri (Bell 1978, 1985) and Nectophl.tynoides (Orton 1949) have an operculum, and L. arch@

and L. hamiltoni have a small fold at the base of the legs. Embryos of Platymantis (Akala 1962; Alcala and Brown 1957) and CeratobatrmhusfiwnthepZ have a membrane that hides the front legs until rate in develooment. Either the front legs developyater than the hind le& or an operculum covers the front legs o f h h y d " p h r y 9 ~ ratcrayi andArtholeptellu l;Bhtfooti, and the front legs of B&ceps adsperms are hidden "under the skin" (Wager 1986). At least some cophyline larvae have bilateral, slitlike spiracles contrary to statements by Blomrners-Schiijsser (1975). A number of modified nidicolous soecies lack a soiracle. A spiracle is absent in most egg-brooding hylids but opercular folds remain open and then regress and close by the time of birth. Del Pino and Escobar (1981) suggested that these openings represent a partial development-<;f the spiracle. A spiracle forms in Flectmtus, Philoria, Pipa, R t l q h o m s , and fiinoderma. The resoiratorv use of the small eills that onlv ~ersist for a short peiiod i n J ~ h u t h e o 1 u s auestioned. k h e large tail of some direct developers may serve as a respiratory organ (see Townsend and Stewart 1985). The walls of the pouches of egg-brooding hyLids and the integumentaryincubating crypts of Pipapipa may allow gas exchange with the embryos (Del Pino et al. 1975; Noble 1925b; I. Spannhof and Spannhof 1971). Lutz (1944b) suggested that the . -large surface area of the yolk of Zmhwenuspawulus may serve in respiration and afford protection from predators.
V

Sensoy and Newmu Systems Optic placodes are evident by Townsend/Stewart 4 in E. coqui. The lens and iridial pigment appears by Townsend/ Stewart 6, and the iris is black by stage 10. Retinal layers form in Townsend/Stewart 9, and by stage 13 the pupil darkens and the iris attains the adult coloration. Auditory vesicle diverticula form at Townsend/Stewart 3 and soon lose their connections with surface eaoderm. These vesicles exist as single chambers until the saccule and utricule form at Townsend/Stewart 9. Semicircular canals are apparent by Townsend/Stewart 13, and the eustachian tube and tympanic cavity appear somewhat later. The endolymphatic calcium deposits (ECD; otocyst of Chibon 1960; otoliths of Hughes 1962) are present in many direct developing anurans. These deposits, important to calcification of the skeleton, may be homologous with CaCO, deposits in the endolymphatic sacs (Dempster 1930; Etkin 1964) of typical tadpoles. The ECD appears as small spots at Townsend/Stewart 6 and develops into quadrangular patches by stage 8. Extensions from the ECD form during Townsend/Stewart 9-12, join at the midline by late in stage 12, and fade by stage 15. An ECD has been documented in Gratobatrmhus and other E l e u t h e r o ~ l w(antitlensis, fiwntheti, johnstonei, martinicensis, eautus, and porto&ensis) ; in Cmatobatrmhus, similar deposits are arranged along the dorsolateral extent of the spinal column. OIfactory organs open to the pharynx at Townsend/Stewart 6 and olfactory pits appear apparent by stage 9. Neuromasts are absent in Eleuthem+lw nubimla (Lynn 1942) and uncommon in other nonnidicolous endotrophs. Em-

ENDOTROPHIC ANURANS

181

first ossifications to appear i Eleutherodadylus are in the apn pendicular skeleton at Townsend/Stewart 12. The ischium, sternum, and episternum ossi@ during metamorphosis in Rana but show no ossification at hatchmg in Eleutherodactylus. The angular, mentomeckelian, and septomaxilla appear later in development than in exotrophic tadpoles. An embryo of Eleutherod#ylus slightly before hatchmg is very similar to a ranid tadpole at the completion of metamorphosis, although they each arrived at tliis state by different routes. Dbestive, Uro~enital, Reproductive Systems and Cranial elements associated with larval mouthparts are abThe prirnitive gut separates the early embryo from the large sent. Swanepoel(1970,Brevi~ps adspenus) specifically examo k mass, ends b h d l y anteriorly, and is open posteriorly ined the ontogeny of the chondrocranium and nasal sacs, via a persistent blastopore. Rupture of the oropharyngeal and Haas (1996) reported heterochronic changes in the demembrane during Townsend/Stewart 9 aiiows communica- velopment of the head of herniphractine hylids. tion between the buccal cavity and the exterior. Rapid absorption of yok by the endodermal ceils lining the gut Physwlogy tract to become convoluted and filied Burggren (1985) suggested that conditions of O, CO,, and causes the di~estive with yok. O i y the anterior and posterior ends remain hol- pH witliin aquatic egg masses are not as severe ai has been low throughout embryogenesis, and the intestine never suggested (e.g., R. M. Savage 1935). Bradford and Seymour forms a coil mical of tad~oles. Warren (1922) stated that (1985, 1988a, b), Gollman and Gollman (1993a), A. A. the gut ofdk(ydruphiyne Idevelops first ai the kterior and Martin and Cooper (1972), Seymour and Bradford (1995), posterior ends. Jorquera et al. (1982, Rhinudemza spp.) and Seymour et al. (1991a, b, 1995), Seyrnour and Roberts Teran and Cerasuolo (1988, Gastrothecagrmilis) also studied (1991), and Shoemaker and McClanahan (1973) supplied larval gut development. Vilter and Lugand (1959) noted pertinent data on various facets of intracapsular metaboprecocious development of the alimentary tract in Nectophiy- lism and development of terrestrial eggs. Caiiery and Elinson nuides and suggested that the small f e m "drinks" a nutritive (1996) examined the urea-cycle enzymes during ontogeny substance se&ted bv the oviducal e~ithelium. Labial fila- of Eleutherodaa$us wqui, and Alcocer et al. (1992) determents found on these embryos may be specializations to aid mined that ureotelism was the prevailing mode of nitrogen in the stimulation and absorption of oviducal secretions (La- excretion in larvae of Gastrotheca riobambae. motte and Xavier 1973; M. H . Wake 1980,1982). Endotrophs take 13 days to 9 months to hatch, and the Three pairs of convoluted tubules form the pronephros, developmental period of Eleutherodu&us coqui ranges from and a straight tube leading to the Wolffian duct forms the 17.1 days at 24.6"C to 26.3 days at 21.1C (Townsend and mesonephros at Townsend/Stewart 6. By stage 9 the meso- Stewart 1985, 1986). A decrease of 1C increased develnephros has enlarged, and by stage 13 the pronephros is opmental time by 2.5 days. Their Q of 3.92 is higher , nonfunctional (Lynn 1942; also see Meseguer et al. 1996; than any of 15 species of temperate-region frogs except Selenka 1882). The urinary bladder diverticulum from the Rrcaphus. This large value suggests that direct developers cloaca and the liver and pancreas diverticula from the gut are have a greater sensitivity to environmental temperatures and present by stage 6. In the next few stages, the liver becomes reduced tolerance to thermal deviations. Wager (1986) lobed and the gabladder is formed. The gonadal ridge ap- noted that Brm'ceps adspemus takes about 2 months to hatch pears between Townsend/Stewart 6-13. Chbon (1962, E. and Anhydrophiyne rattrayi and Avthroleptis wahlberg-i take 4 martinicensis) and Larnotte and Xavier (1973) and Lamotte weeks. et al. (1973, Nimbaphynuides occidentalis) discussed gonad Eleutheroda~lus coqui can develop in water of low tonicity from fertilization through hatching (Elinson 198%). development. Cei (1980:281) cornmented on the development of Ba- Wager (1986) noted that Anhydrophryne larvae drowned if trachyh taeniata as foliows: "After 15-20 days [when em- placed in water, and Lynn (1948) found that embryos of E. bryos are intracapsular] solid wiutish wastes can be seen in ricordi raised in water grew more slowly than those in air the taill' Cei speculated that tliis storage was similar to that unless the jeily was removed. The development of embryos of a cleidoic egg. He did not state if the observed tadpoles of Gastrotheca riobambae (De1 Pino et al. 1975) in water or lvere alive or preserved, and if this whte material is differ- Ringer's depended on the stage of development. Heart rate ent from the white precipitate that commonly occurs in is about 12 beatslmin at 30C in Platynzantkpehensis at preserved tadpoles, rther investigations certady are war- about Townsend/Stewart 7 (Atoda 1950); Burggren et ai. (1990) presented data on the developmental changes in carranted. diac physiology of E. coqui. Larvae of Batrmhyh start development in what appears to be a nidicolous mode but spend The sequence of postcranial ossifications in Eleutherodactylus most of their development as exotrophic tadpoles. is similar to that of &na, although these bones o s s 4 along The thyroid gland differentiates by Townsend/Stewart with cranial ossifications (Hanken and Summers 1988; 13. Tail degeneration is stimulated by thyroid hormones Lynn 1940a) that are highly modified from the norm. The (Hughes 1966; Lynn and Peadon 1955), while limb ini-

bryos ofArthrolepteLla have neuromasts on the head (De Villiers 1929b; Lambiris 1988a, b). The neural tube is closed by Townsend/Stewart 3 in Eleutherodu~lus coqui, primary brain divisions are present by stage 6, and ali parts of the brain are present in correct proportions by stage 9 (also Hughes 1959; Lynn 1940b). Schlosser and Roth (1997b) studied the modified development of the peripheral nervous system of E. wqui.

182

GISELLE THIBAUDEAU AND RONALD ALTIG

tiation (but not completion) is independent of thyroxine in contrast to the condition in typical tadpoles. The lack of effects of thyroidectomy (Hughes 1966; Lynn and Peadon 1955) suggests that development is thyroid independent, although Elinson (1990) noted that there may be thyroid hormones stored in the egg. Goiterogenic properties of phenylthiourea affected pigmentation in E. ricordi more severely than in other tadpoles (Lynn and Sister Aifred de Marie 1946). Oviducal secretions of Nimbaphlynoides occidentalis are maintained during pregnancy by corpus luteumprogesterone mediation until the last two months of gestation (Xavier 1977). Brink (1939, Arthrokptella) and Lynn (1936, 1948, Eleutherodactylus) addressed growth and function of the thyroid gland, and other references that pertain to endocrinology, development and metanlorphosis include Exbrayat and Hraouibloquet (1994), Hourdry (1993), Hughes and Reier (1972), D. H. Jennings (1994), and B. E. Morgan et al. (1989).

Evolution of Endotrophy
Endotrophy was attained through modifications of the normai amphibian developmental prograni (also see chap. 11). We do not imply that ai lmeages with similar development l necessarily were produced by the same evolutionaq pathway, but certain common denominators are present. The fact that these alterations have occurred repeatedly with similar results suggests provocative ideas about potential evolutionary and genetic mechanisms (see Fang and Elinson 1996). All obsewed "intermediates" would not have had to occur as a progression; those that appear to be intermediates between typicai amphibian development and the more derived conditions should suggest probable sequences of changes and Dromote an understanding: of develo~mentaiconstraints. thers have used this reasckig to illu&ate sequential aiterations of breeding modes of amphibians (e.g., Grandison 1978; Heyer 1969, 1973a; M. H. Wake 1982) and other taxa (e.g., RafF 1987; Strathmann 1978, 1993; Strathmann et ai. 1992). In some circumstances the intermediates seemingly would be at a disadvantage by not expressing the appropriate phenotype for either option. In the case of anuran development, an exotrophic tadpole, a profoundly modified developmental pattern as in Eleutherodactylus, and many cases &at apper as intermediate morphologies along the nidicolous continuum ali produce a froglet. Because of the duai Me-history pattern of anurans, the developmentai exdusion of features uniaue to tad~olescauses elaborate changes in larvai developmentai patterns with no effect on the adult. Although the understanding of development of endotro~hs based on rather few data involvinn a smali asis semblage of taxa, we suggest several ideas that differ from current dogrna. Genetic on-off signals (e.g., morphogen production and distribution. cellular comDetence and induction., Dattern formation) d u e n c e the sequence of events or pattern of development. The period during which an event(s) occurs reflects the rate at which a develo~mentaievent or Drocess occurs: Genetic mechanisms are of prime irnportance in considering
I U
I .I

sequence of developmental events, whiie abiotic factors;particularly temperature, often are more instrumental in immediate considerations of altering developmental period. Aiterations in developmental sequence were instrumental in the evolution of endotrophy, while alterations in developmental period probably had a rninor influence on these unusual developmental patterns. Even cursory comparisons show that endotrophy is not simply a shortened or telescoped sequence of events. Developmental periods of typical anurans from egg to metamorph range from 9 days (e.g., Forge and Barbault 1977, Bufopentoni) to 3 or more years (e.g., Metter 1967, Ascaphus truei) with inconsequential changes in general developmental sequence; both extremes exceed the known developmentai periods of endotrophs, and in the examples given, differences in sequence reflect habitat (Nodzenski and Inger 1990) and not development. A common staging table will work in either case. A reasonably consistent developmentai sequence in early stages of ali exotrophs eventualiy results in a wide diversity of morphologies (e.g., the intricate oral apparatus) because of developmentai modifications later in ontogeny. Because most endotrophs lack the oral apparatus and have less variation in tail morphology compared with exotrophic tadpoles, the morphologicai variation in endotrophs is lower than in exotrophs. Conversely, the multiple times that endotrophy has evolved in a number of lineages suggest that (1) the genetic mechanisms involved are either relatively simple, simply regulated, or likely to occur; (2) similar mechanisms operated repeatedly to produce similar developmentai patterns; and (3) the mechanisms involved alterations in gene regulation rather than mutations of structural genes. It seems highly unhkely that homologous mutations of structurai genes would occur repeatedly in the proper sequence in so many Merent lineages. If there has been a reversal from endotrophy to exotrophy (e.g., Wassersug and Duellman 1984), gene regulation rather than structurai mutations seems a more likely mechanism. We do not imply that the process was not complicated; changes in the (1) development, genetics, and physiology of the egg and embryo and (2) behavior, morphology, and physiology of the adult had to be properly orchestrated (e.g., Xavier 1977). Based on the obsewations that apparent intermediates between exotrophic and endotrophic development occur, that species of Gastrotheca may produce either tadpoles or froglets, and that a frog results from ali such developmentai variations, we reiterate that the examination of extant taxa will provide insight into the probable patterns and mechanisms and aid in illustrating the developmentai aiterations. Depending on the hierarchicai leve1 of developmental regulation that is postulated, ali lineages would not necessarily pass through the same sequence of events. These presumed intermediate species should be placed in a sequence only for heuristic purposes, and they should not be viewed as incomplete attempts (Lutz 1948) at terrestriai development. In hopes of stirnulating thoughts on the subject and coilection of pertinent materiais, we continue this discussion under the following three topics: (1) two general pathways based on the presence or absence of oviposition, (2) characteristics of exotrophs

ENDOTROPHIC ANURANS

183

that seem to imitate conditions found in early stages of nidicolous and direct developing endotrophs, and (3) changes within the various forms of endotrophy. Om-positwn and Endotr-phy Although the changes most certainly occurred in concert, we suggest that various alterations in oviposition were a fundamental factor in the evolution of endotrophy. Without egg retention (= lack of oviposition), the associated changes in both adult reproductive systems that d o w ovovivipary and vivipary could not have arisen. Even so, the ovoviviparous forms (e.g., Ehutherodu&lur ja'peri is more similar to direct developers compared to some species of Nectophrynozdes, which are more similar to nidicolous larvae) do not follow similar embryological patterns. If so, one would predict that these conditions arose Merently; the condition in Ehutherodu&lur rnay have involved the retention of a directdeveloping egg, while in Nectuphrynoides the changes involved the retention of an egg that would develop at some point along the nidicolous continuum. Although these differences do not suggest it, ovovivipary and vivipary also could have evolved in sequence once oviposition was suppressed, or at least a series of ovoviviparous-likeconditions (i.e., vary as to length of time retained) rnay have been intercalated into the evolution of vivipary. The other forms of endotrophy involve a common oviposition pattern and moddied developmental patterns. Hemiphractine hylids have highly modified breeding behaviors (i.e., fertilization foliowed by positioning of eggs) and a unique embryology. Exovivipary, which essentidy involves nidicolous larvae that are intimately tended by a parent, could easiiy be derived from the normal nidicolous types by moddication of either parental (e.g., Rhinudema) or larvai (e.g., Assa, some Leiopelma, and some Sooglussus) behavior. Negative geotaxis, perhaps modulated by a chemotactic response to material from the adult skin, would suKce in the latter case. Even though we do not have enough details of even morphological (much less genetic) development to make astute predictions about the possible evolutionary steps leading to the larval types along the nidicolous continuum, the developmental regulation that produced these forms is surely less complex than that which produced direct development. At least for the present discussion, we will consider nidicoly as a potentiai precursor to direct development. A postulate of where on the nidicolous continuum that direct development rnight have arisen wiil be tenuous at best. In summary, hemiphractine hylids possess highly modified breeding biologies and strangely different developmentai patterns that certainly evolved via mechanisms different from d other endotrophs. Other endotrophic forms rnay have similar developmentai stages, and in the next section we present scenarios whereby particularly nidicoly and then direct development might have arisen from exotrophic development. Environmentdy influenced variations in development are important, but they ordy aiter developmentai period and not

developmentai sequence. Regardless of the embryonic period, exotrophic embryos typicdy hatch w i t h a narrow range of stages centered at about Townsend/Stewart 22. From hatching untii the tadpole stage at Gosner 25 (about 3% of develo~mental time and about four develo~mental stages), the hatchling is particularly vulnerable, locomotes slowly over short distantes by epidermal ciiia, and stabilizes itselfwith a transitory adhesive gland. Yolk reserves support development for about the first 22 stages, and perhaps 80% of the developmental time commordy remains after yolk exhaustion; this period primariiy involves isometric growth with minor developmentai change. A second period of development (i.e., metarnorphosis) comes at the end of larval ontogeny. We first suggest ways in which exotrophic development might be altered without requiring changes in developmental sequence. Initial developmental changes of exotrophs would affect the period, and therefore the stage, to hatching. The embryo could remain intracapsular during part or d of the time that is normdy spent as a hatchling (Lamotte and Lescure 1977). "Late hatching" commordy accompanies oviposition out of water, on the soii surface near water (e.g., Chirixalus idwotocus, Ranagrayi), on leaves (e.g., centroenids, some hyline and phyliomedusine hylids), and underground or under objects (e.g., some L e p t o ~ l u sPseudophyne; see Alcala , 1962). The development that ensues during the extended intracapsular period enhances the probability of survival and therefore the selection for oviposition out of water if egg mortality is low (ordy 8.1% survival in Philuria; Malone 1985).If reduced oxygen concentration is the hatching stimulus (Petranka et ai. 1983), the increased oxygen concentration in an aeriai environment rnay d o w an extended intracapsular period by dowing the ihreshold concentration of oxygen to be reached later in development. Higher oxygen concentrations also rnay cause the hatching glands to develar, and secrete later than normal. Embrvos that hatch "late.' often have erdarged gds (e.g., ~arnohe Lescure and 1977) presumably to accommodate the larger size and advanced development wMe con6ned in the jeliy. Embryos of Ranagrayi my reach at least Gosner 25 $&r to hatching (personal observation), and embryos of Geocrinia vitoriana rnay arrest at stage 26 for 4 months if conditions are not amenable to hatching (A. A. Martin and Cooper 1972). Downie and Weir (1997) found that the response of tadpoles of Leptoh~lus$scur that were artlficidy induced to arrest differed in growth and metamorphic response. The longer the tadpoleS stayed in the foam, &e more~lowly they grew when released and a smder proportion eventudy metamorphosed. The mean metamorphic size of those that arrested for longer periods was greater than those that left the foam earlier.butthe variabili6 was " greater. Based on oviposition sites on land and vegetation out of water, embryos of Dendrophryniscus minutus and Tachycnemis seychelhnsis rnay also have-th capability to remain &tracapsdar for longLr periods than normal. Researchers need to study the feedback interactions that rnay exist among remaining yolk reserves, stimulation of hatching, and developmentai stage at hatching. We argue that no dditional enirgy or alteration of de-

184

GISELLE THIBAUDEAU AND RONALD ALTIG

velopment wouid be needed to make these rninor developmental adjustments common to nonaquatic eggs, and &e individual can reach Gosner 25 (i.e., a tadpole) in this manner. If particuiate food is not present, the tadpoles of Bufoper&enes metamorphose at a smaiier size but at the same time as siblings that feed (Crump 1989b; aiso see Inger 1992b, RhaccIphol"lts~auni; McDiarmid and Altig 1990, Bufo haematiticus; Weygoldt 1989, Flectonotus~oeldii).Seemingly there are no data on the possibility of amphibian larvae absorbing dissolved organic material (DOM; see Manahan 1990), and bacteriai popuiations were not monitored in Crump's experiments. The ovum of B. peribenes is large and lightly pigmented and the clutch size is smaii for a Bufo of about 50.0 mm snout-vent length. Aitig and Johnston (1989) classified B. penaienes as an exotroph because they considered exotrophy to be the primitive condition and feeding to be the likely scenario foilowed in most natural cases. In the present context, one aiso couid consider B. pen3enes as a facuitative endotrovh residine: at the absolute teiminus of the nonmodified ind of t h l developmentai continuum of nidicolous anurans (Altig and Johnston 1989). Therefore endotrophic develovment is vossible without develovmentai modifications in species whose eggs are weil within the size range and presumed energy content of exotrophs. B. penig-lenes has sufficient energy in the egg, a function of maternal reproductive physiology, so that the resuiting offspring can survive to metamorphosis without feeding, if that scenario is encountered in the environment. The benefits of this facuitative form of develovment are obvious. and one has to wonder if other species that develop rapidly in habitats with apparently low productivity can respond similarly. Also, we suggest caution in subjectively equating the absence of visible aigal growths with limited food availabhty. At least in dystrophic, tropical pools, the detritai/fungal/bacterial component is likely more important as a food source than photosynthetic aigae. Similar cases occur in highiy ephemerai desert pools.
I I

Chan~es within Endotrophs The next step toward endotrophy wouid be to change the facultative nonfeeding type (e.g., B. pen@nes) to an obligatory nonfeeding type (e.g.,Anodonthyla, Phibloria, Phynodon, and Pethodontohyla).That is, an entire tadpole morphotype that does not feed is produced, but one must be aware of the danger of generaliiing based on incomplete knowledge of the developmeiit of frogs with large eggs. A case in point: Dueiiman and Gray (1983) showed that the tadpole of Fectonotusfitzgerali did not feed and metamorphosed in 5-20 days; they predicted that the tadpole of other Fbctonotzts wouid respond similarly. Weygoldt (1989) reported that the tadpoles of E~oeldii feed readily and take 23 days to metamorphose. &o, the fetus of Nimbaphlynoides occidentalis feeds from uterine secretions (Vilter and Lugand 1959), and the embryos of Rhinuhmza darwini may feed from the lining of the vocal pouch of the male (Goicoechea et ai. 1986).We need to know more about comvarative visceral develovment among members of various developmentai guilds. For ex-

ample, does the development of the intestine of Nectophlynoides spp. differ from that of Eeutheroductylus spp.? The next step might involve the first changes in developmental sequence. Because structures lost by individuais in the intermediate and modified parts of the nidicolous continuum occur oniy in embryos or tadpoles, the aduit phenotype is not dected. Considerably less than an entire tadpole morphotype can produce a frog via metamorphosis, and development can proceed toward metamorphosis from an organism that resembles various stages of a typicai exotrophic hatchling. A tadpole morphotype is mandatory oniy if feeding is mandatory because of developmentai energetics but not for the production of a froglet. Whether eggs that develop into nidicolous larvae have more yoik or a higher yoik density than eggs of exotrophs is not known, but some increased amount c e r t d y is expected. Patterns of phylogenetic and ontogenetic gain or loss (e.g., Aiberch and Gaie 1983; Thibaudeau and Aitig 1988) suggest that modifications of development that involve losses may occur in the reverse order of ontogeny of typical species. We have reliable information on the ontogeny of typicai embryos but lack these details on a diversity of nidicolous forms. If these data were avadable, we shouid be able to discern common mechanisms among lineages and make hypotheses about the development of various features or groups of features controiled by a single inductive cascade (e.g., Elinson 1990). Early interruption of the cascade wouid prohibit the formation of aii subsequent structures because of their developmental interdependente. Within the confines of little data, we specuiate that a sequence of expected loss may be adhesive glands, externai giiis, oral apparatus, coiled gut, lateral line, and spiracle. Conversely, Junc et ai. (1994) and Kaiser and Altig (1994) couid not find concordance among character losses of nidicolous lamae w i h the same genus of dendrobatids (see E. N. Arnold 1994). The nidicolous embryos of Arthroleptellahewitti provide another iilustration. Probable vestiges of marginal papiliae at the corners of the mouth and neuromasts are present, and infralabiai and supraiabiai cartilages, which are probably instrumental in the inductive cascade of mouthparts (Elinson 1990), are present. Detailed studies of the developmentaipattern and developmentalreguiation of features of representative types of embryos are needed. As indicated above, the early appearance of large limb buds and the usual lack of a number of larval features in both nidicolous and direct developers support a very tenuous assumption that nidicolous-like patterns occurred during the evolution of direct developers. For the evolution of direct development, (1)an increase in energy, by some combination of increases in egg size or yoik density, or both, beyond that which was posuiated as being needed to enter the nidicolous grade of endotrophy from exotrophy wouid be required (see Digression b e l o ~ )and (2) the first profound , changes, something more than developmentaitruncation, in developmentai timing couid have occurred. The most obvious alteration concerns precociai limb and head formation, and the early appearance of these structures signals a highiy

ENDOTROPHIC ANURANS

185

modified cascade of developmental events. Whiie remembering that stage designations have no relevante to size or real time per stage, limb buds appear at Gosner 26 in typical tadpoles -shortly after the tadpole becomes free-swimming. In direct developers, limb buds appear immediately after neurulation. Comparisons of timing among developmental modes may be frivolous other than as an indication of degree, but limbs begin to develop perhaps 10%-20% (much more for head development) and 9-10 Gosner stages earlier in relative developmental time than in typical exotrophs. Large, heavily vascularized tail fins presumably serve as a respiratory surface, but no data are available to support this idea; embryos of many species, often ones larger than others with large tails, have s r n d tails. External giiis of direct developers never approach the size or complexity of those of many exotrophic hatchlings. Retention of eggs by a direa developing species may result in ovoviviparity and seems like a relatively minor shift in development, but the similarity of development and morphology between embryos of ovoviviparous forms and the modified end of the nidicolous guild (not direct-developing forms) suggests that the evolution of ovoviviparity and direa-development are independent events or that the constraints of ovoviviparitv are different from those for direct development. Changes in embryonic physiology, especiay respiration, adjustments in breeding behavior of the adults to ccomplish.internal fertilization; and alterations in maternal endocrinology to repress ovipositional stimuh would be involved. Even so, the smail number of ovoviviparous fiogs signals either strong, negative selective pressures or constraints. In most circumstances the energetic cost of carrying embryos may be too extreme over too long a period, especially if being gravid retards preparation for subsequent broods. The transition from ovoviviparity to nonplacental viviparity (see M. H . Wake 1978) is perhaps a smder step, aithough some sort of dometry of the feeding and digestive structures would have to occur to aow intrauterine feeding.
L ,

Dgresswn on E Size and Composition m The lack of information concerning the relationships betnreen egg composition and size is a strong deterrent to a thorough understanding of reproductive biologies. Egg size is commonly invoked as an important modher of anuran development, and the largest anuran eggs are laid by endotrophs. Numerous studies show that larger eggs provide greater fitness, even for up to eight years later (Semlitsch et ai. 1988), but it seems that no one knows what it is about a big egg that is good. A big egg is assumed to contain more energy, but (1) variations in yolk density (yolk weightlvolume) and (2) other important components influenced by viteiiogenic physiology (e.g., mRNA, polypeptides translated by the mother) have not been taken into account. Egg sue alone is a poor predictor of endotrophy of any kind; there is a 25% overlap in the range ofendo- and exotrophic eggs sizes. Exotrophic eggs range from about 0.8 (e.g., Dendrophyniscus) to at least 4.0 mm (e.g., h a p h w tnsei), and endotrophic eggs range from about 1.8 (Sooglosus~ardinen) to

10.0 rnm (Gastrotheca watuph~es,G. weinlandi, and Hemiphractus scutatus). Geocnnia rosea, with an egg diameter of about 2.4 mm, develops directly in about 6-8 weeks, while G. victonana has a typical tadpole that develops normdy from an egg 3.1 mm in diameter (A. A. Martin and Cooper 1972). Based on these observations and the case ofBufoped~enes,we contend that the increase in available energy does not increase linearly with increases in egg size, and an increase of energy beyond that contained in some average egg is not the prime prerequisite of endotrophic development. Even through Salthe and Duellman ( 1973) suggested that seleaion favors increased ovum size in s r n d frogs and that only s r n d and medium-size frogs are able to enter fdly terrestrial reproductive zones, there must also be some correlation of egg composition and the likelihood for the evolution of endotrophy within that sma size group. Even if direa development and sma adult sue are correlated, one must consider these two factors on development and eventual morphology (e.g., Hanken 1983, 1982; Hanken and Wake 1993). Based on the geometry of a sphere, does a 5-mm egg contain an increase of exady 27% of d components relative to a 4.5-mm egg? Doubling egg diameter equals an 8-fold volume increase, whereas doubling the diameter of a large egg is a much larger absolute amount but a smaer percentage increase than for a sma egg. N o one seems to know how the composition of energy stores change with egg size and what patterns might be most advantageous under any given circumstance. How does metabolic regulation of yolk consumption differ among developmental modes (e.g., A. A. Martin and Cooper 1972)?Answers to these m e s of , questions likely would demonstrate alterations of maternal oogenic physiology. The amount and distribution of volk affects cleavage. The " membrane-bound, protein-matrix yolk platelets of various sizes and shapes distributed throughout the egg persist at least through neurulation. The transcription of the zygotic genome and the degradation of the platelets as an energy source do not start immediately after fertilization. Eggs vary intra- and inters~ecificav volk densitv. and there are more in and larger platelets toward the vegetal pole compared with the animal pole (= moderately telolecithal). This dispersion produces an increasing yolk gradient from animal to vegetal ; bole. and an increase i iolk aensitv would increase the dope Lf &e gradient. The poskon of the t h r d cleavage plane (fiist horizontal plane) above the equator is determined by this increase in the amo& of volk " nradiekt. We assume that yolk would shft the curve of the yolk gradient toward the animal pole. We suggest that the latitudinal position of the third cleavatze furrow above the eauator reflects the shaoe of u the yolk gradient, and that t h s cleavage plane probably occurs at a point of similar yolk concentration (at least in eggs of about the same size). If so, species that have a high yolk density, whether exotrophic or ndotrophic, compa&dto a less dense egg would have the t h r d cleavage plane at a higher latitude. This circumstance would cause the formation of smaer micromeres and larger macromeres than in a

i'

186

GISELLE THIBAUDEAU AND RONALD ALTIG

companion species with lower yolk density. Aiso, ali anurans, exo- and endotrophs, rnay not practice holoblasty of ali cleavage frrows (e.g., Alcala 1962; De Villiers 1929b; Elinson and De1 Pino 1985; Warren 1922), perhaps because of yolk amount or disttibution. If so, this further increases the size and decreases the number of macromeres. The embryonic disc of Geotheca that resembles the meroblastic blastodisc of amniotes is a paramount example (Elinson and De1 Pino 1985). Muiticeliuiarity is a primary resuit of cleavage, and differences in cleavage patterns (not including holoblastic vs. meroblastic conditions), and therefore blastomere number and size. ~robablvhave no develo~mentdconseauences. ' The resuits of these patterns can be used to determine egg composition. Embryos of many direct developers and certain exotrophs are elevated above the yolk mass. Eggs of exotrophs with extraordinarily smali rnicromeres (i.e., highlatitude third cleavage) and elevated embryos (e.g., centrolenids, Helewporus) rnay share other characteristics with endotrophs. Are there other factors in a large egg that might make a lar~er embrvo?The rather smali intraclutch and certain intraspecific variation in egg size rnay reflect something as simple as water content rather than energy content, and measures of weight/volume (and not simply diameter) or chemical analyses are needed to clarify this idea. The length of the neural groove is the first constraint on eventual size by influencing the starting length of the presumptive body. Eggs with larger diameters provide a longer distance between latitudes across the animal pole to the opposite side, although the amount of increase per unit change in size decreases as egg size increases. If the factor(s) that results in a larger egg is coupled with a faster growth rate, then the benefits attributed to this larger embryo shouid persist throughout life. We do not know how much energy it takes to make a direct developing hatchling of a given size, but we question if the generaliy larger eggs of direct developers are in fact necessary for production of a froglet. We suggest that a large part of the yoik in direct developers is for posthatching growth to an appropriate size to feed -not for expenditure on the formation of a froglet per se. This assumes that endotrophy is no more energeticaliy expensive than normal development. When an endotrophic froglet hatches, there is always considerable yolk that remains. Similar yolk reserves in hatchlings of exotrophic tadpoles support what is essentiaiiy a hatched embryo unta it grows to a feeding stage. Some part(s) of the feeding or digestive system of direct developers, structural or enzymatic, rnay not be capable of handling extraneous food at the time of hatching (e.g., Toloza and Diarnond 1990a, b). We know of no data pertinent to this hypothesis (see Twombly 1996; Wassersug 1986, 1996). Also, the froglet rnay simply be too smali to find suitable-size food items. Vestigial-winged Drosuphila are too large for hatchling E l e u t h e r o ~ l u s mqui (Elinson et al. 1990) that are about 6.0 mm SVL (mouth gape = 1.5 rnm; distance between jaw articulations 3.3 rnm). Aduit Sooglosus gardineri (Mitcheil and Aitig 1983) commonly eat mites,
I
<7

and hatchlings (about 1.6 mm SVL) rnay also be able to consume rnites. We ask if one codd observe normal development afier the removal of the amount of yoik that normaliy remains after hatching of a direct developer and replace it with an inert material to maintain egg conformation? We know of no data that suggest the kinds of relationships among egg size (or other egg qualities), yolk remaining at hatching, and time and growth before first feeding. The absolutely smdest egg that can produce a froglet directly rnay be dictated by embryological mechanics and scaling funaions associated with locomotion, feeding, and dehydration, and the 1.6-mm egg of S.gardzneri is probably near the lower lunit.

Changes in EmJelly Although not a direct modifier of endotrophic development, certain qualities of the egg jeliies had to be modified in association with the various routes out of the aquatic environment. Suggested fnctions for the layers of mucopolysaccarhide jeUy around amphibian eggs include: protection against predators, bacteria, fungi, dehydration, and thermal shock; enhancing solar radiation; attachment to a substrate; and restriction of sperm entry to a specific zone and inhbition of polyspermy. Few correlations among phylogenetic relationships of frogs, their reproductive ecologies (e.g., J. A. Moore 1949b), breeding modes, and egg jeliy configurations have been suggested. The eggs of phyliomedusine hylids and Chikalus hatch explosively, and touching one rnay cause numerou others to hatch in a crackling fry. This explosive hatching suggests quite a dfierent osmotic regime in these eggs compared with aquatic eggs, which become increasingly flaccid as development proceeds. Exiting the aquatic environment wouid produce additional demands on the functions of egg jeliy (e.g., maintaining turgidity to protect the embryos from trampling by parents and avoid the loss of surface area for respiratos. exchange by collapse of the jeily, retaining spherical egg conformation for proper development in the absence ofwater buoyanq protecting the egg from desiccation without inhibiting gas exchange, and aiding attachment to the damp surface of the mother's back). The outer jeliy layer of direct developing eggs ofien is tough and elastic (eggs bounce when dropped), while the ovum wiU rupture under its own weight if the jelly is removed in air. The thick, proximal layer is a stiff gel. Especiaiiy in hemiphractine hylids that carry exposed eggs, the outer jeliy becomes brown and parchmentlike. Exovivipaq and Paravivipaq The site of oviposition, the sex and behavior of the attendant parent, and activities of newly hatched larvae were important parameters used to delirnit exoviviparous and paraviviparous g d d s (Aitig and Johnston 1989). While accepting that a discussion of such behavioral modifications of both embryos and aduits is entirely speculative, we suggest that initial stages in the evolution of exovivipary and paravivipary were driven primarily by selection on male behavior. If tactile or visual stimuli are important in breeding, the male probably

ENDOTROPHIC ANURANS

187

has an increased iikelihood of perceiving these clues based on the amplectic positions of rnost anurans. Experimental ~rotocols shodd &volve the resDonses of blinded and anosrnic frogs to real and durnmy egg rnasses. In the evolution of paravivipary, a change in ovipositionai behavior of the rnaie would position the eggs upon the back of the fernaie (e.g., Hemiphracius) with little or no aiterations of the back rnorphology or behavior of the fernaie. A photograph (collection of R. W. McDiarrnid) of an Atelopus cmciger with two eggs stuck to the lower left quadrant of its back suggests that no great modification of the egg rnernbranes or maternal skin rnust occur to d o w at least initiai cohesion of the eggs and skin (see Bourquin 1985 and Prez et ai. 1992a, b). The evolution of brood pouches codd then proceed. Perhaps the ovipositional loop of Pipa (Rabb and Rabb 1960) is an aborted atternpt to go to the surface to breath; seerningly the fernaie would have to initiate the triw to the surface. but the rnale could interrupt the atternpt via steering. Apparently behaviorai changes prirnarily in males accornpiish internai fertilization in E. cogui and egg positioning in Gastrotheca. Understanding how horrnonal profiles are rnodified in these cornplex forrns of breeding biologies needs additionai research. Horrnone-rnediated behaviors of males, fernaies, and ernbrvos rnust be orchesuated. and studies of similar behaviors & exotrophs (e.g., A&es k d dendrobatids) should be informative. One would suspect that some behaviors operate only for short periods and in the proper temporal and reproduaive contexts. Do malehsa and R h i m h m not eat their young because they recognize ernbryos for what they are via sight or oifaction, or are the responses to what appears as a potentiai food item inhibited? In other cases, exci-

tation (e.g., brooding by either sex; Townsend and Moger 1987) rnust occur. The behavior of hatchlings rnust be modified to d o w thern to rnount the back of the rnaie at hatching and remain there (e.g., Dendrobates spp., Leiopelma spp., and Sooglosussechellensis) or enter a pouch (e.g., Assa). Maies that carry ernbryos aid adhesion or rnovernents of the young by skin secretions (G. J. Ingram et al. 1975). The gaseous and nutrient interactions of parents and larvae need considerable investigations (e.g., De Prez et al. 1992a, b; Wells 1980), and the potentiai for immunological interactions between adults and ernbryos should be investigated.

Summary
As noted by E. R. D u m (1924), anuran adults and larvae live in different worlds so that the significant changes of rnetamorphosis d o w independent selection on the two life stages. In a functional context, one can consider two partially overlapping genornes w i t h each amphibian egg. Largely different sections of this cornposite genorne rnust be deployed sequentially in order to produce larval features during developrnent and adult features at rnetamorphosis. Within this single genomic program, the different developmental patterns are produced by dterations in gene expression, and aiterations in either section of the genorne rnust not adversely affea the other stage (M. J. Rose 1982). The relative uniformity of rnost endotrophic developmental patterns suggests that the regulatory changes that were ernployed were relatively similar in many cases, but the trajectory of the development of herniphractine hyiids certainly is grossly different. At a time when the phylogenetic relation-

Fig. 7.3. Hypothetical scheme of the sequenti~l expression of various batteries of developmentai and regulatory genes. (A) Typicai exotrophic tadpole. (B) Direct developer. The fact that some characters may be controlled within an inductive cascade is ignored in this demonstration. Abbreviations: BL = blastulation, CL = cleavage, EG =

externa1 gills, GA = gastdation, and HB-F = formation of the head and body morphology of a frog, LI = limbs, NE = neudation, OA = oral apparatus, and TA = tail. Lengths of blocks indicate relative time periods of expression.

188

GISELLE THIBAUDEAU AND RONALD ALTIG

ships of egg-brooding frogs were not known at the famiiial level, Noble (1917) noted that a cornrnon mode of embryology was a strong signal. Endotrophy offers a means of shifting life history dramaticay (e.g., Raff and Kaufman 1983); and depending on the lineage being examined, some of the developmental changes require that such shifts occur relatively early in development. RafTet al. (1987) suggested that such situations are the resuit of heterochronies that occur over relatively short evolutionary intervals (and see Alberch 1985; Alberch et al. 1979; Fink 1982). This suggestion does not imply that the basis of the changes is not genetic or that they are not geneticdy fixed, nor does it imply monophyly among endotrophs or identical regulatory changes in a11 lineages. A hypothesis involving heterochronic changes gains support from the occurrence of similar energetic-developmental constraints in other groups. Such comparisons are most pertinent when made among groups that include developmentai patterns that produce both free-living larvae and direct developers. The discussion by Raff (1987) on the evolution of direa development in sea urchins is an intriguing comparison. Diagrams similar to those of Raff (1987) present specuiative ideas on operative sequences of major gene groups of a direct developer compared to a typical tadpole (fig. 7.3). We suggest that mutations in higher level regulatory genes (= homeobox, hox), perhaps coupled with gene duplication, ailowed for major steps in developmental patterns to be made quickly. Direa development in anurans resuits from a suppression of probable gene famiiies that govern typical larval traits (e.g., larval chondrocranium and hyobranchium; d oral features of tadpoles) and an advancement in timing of other cascades (e.g., aduit head formation; leg ontogeny) that resuit in early initiation and formation of a typical frog morphotype usudy without a tadpole morphotype. We reiterate that endotrophic development is not a simple shortening or telescoping of typical development, and a thorough understanding of the evolution of endotrophy will come only through an understanding of the developmental regulation that causes the retiming of the various events. In summary, the occurrence of frog eggs that develop without the quintessential tadpole suggested by general dogma is intriguing to a h o s t everyone. While considering the outlandish idea that a male frog would seleaively swallow embryos into his vocal sac and later burp up baby frogs, perhaps Lynn's (1961:157) pun marks our amazement the best: "There can be no doubt that this is a case in which

the sudden arrival of quintuplets left the father speechless." Surely it is oniy the relative rarity of finding endotrophic embryos that is responsible for the large gaps in our knowledge. Because of the divergent developmental pattern compared with typical frogs, direa-developing anurans, with Eleutherodmtylus cogui being the most cornrnonly studied species, are prime subjeas for the study of developmental processes (Elinson 1987a; Elinson et ai. 1990). Comparisons among several nidicolous forms with increasing degrees of divergente from the norm also wouid be a viable option. Because of the muitiple occurrences of endotrophy throughout the Anura, we must remain cautious about making premature generalizations. Additional data are needed on the development, endocrine controls (Hourdry 1993), genetics, and physiology of endotrophic anurans as related to reproductive behavior (e.g., Townsend et ai. 1981). Both classicai and molecular approaches wiii be productive (see Hanken et al. 1997). Additional details of egg composition as a funaion of size among species and developmental modes need to be dusuated in light of the effects on development and the energetics relative to breeding strategies. Transplant experiments between developmental types (Elinson 1990), hybridization of exotrophic and endotrophic species of Gartrotheca, and manipuiation of egg components wouid be informative. Surely one is not going to produce an endotroph from an exotroph, but the possibility of generating a developmental variant similar to Bufopeniglenes discussed above might not be totdy out of the question.

ACKNOWLEDGMENTS
In alphabetical order, D. Auth, A. Channing, R. C. Drewes, W. E. Duellman, M. Cherry, M. Hero, D. L. Jameson, A. J. L. Lambiris, W. Magnusson, H. Marx, R. W. McDiarmid, J. I. Menzies, R. A. Nussbaum, J. D. Roberts, L. Rodriguez, F. Slavens, R. Stocks, P. Tolson, and.J. Vindum contributed or loaned valuable specimens for use in this study or assisted our efforts in other ways. W. J. Diehi, R. P. Elinson, J. Hanken, and D. S. Townsend contributed substant i d y to earlier drafts. K. Adler and J. R. Mendelson I11 provided citations and other pertinent information, R. Rappaport commented on certain of the ideas presented above, and J. Hanken kindly supplied several of the photographs used in this chapter.

PHYSIOLOGY
Coping w i h heEnvironment
Gordon R. Ultsch, David F. Bradford, and Joseph Freda

Introduction
Early scientific interests in the study of tadpoles centered mainiy upon developmentai biology and date back to at Ieast the morphologicai investigations of Jan Swammerdam (e.g., 1737-1738). Physiologicai studes, aiso inspired by an inquisitiveness about developmentai processes, are generally dated more recently, usually from the seminal work of Wilhelm Roux (1888), who tested the preformation theory of development. He destroyed one blastomere of the two-celi stage of a frog with a hot needle and concluded that the remaining celi could not form a complete embryo. This conclusion was later proven to be incorrect (McClendon 1910); frogs have radial and indeterminate cleavage, and the effeas R o a saw were caused by not separating the dead celi from the live one. Nevertheless, Roux is usually considered the "father of experimental embryology." In a sense, Roux can aiso be considered to have been the initiator of studies on the general physiology of tadpoles (see B. I. Balinsky 1981; Browder et al. 1991; Carlson 1988; J. A. Moore 1972). Interests in nondevelopmentai aspeas of the physiology of tadpoles were insignificant untii the last two to three decades. An increase in the number of papers on tadpole physiology during that period parallels a general proliferation of studies of the comparative physiology and physiologicai ecology of lower vertebrates. In this chapter we concentrate on those aspeas of physiology that are especially relevant to the ecology of tadpoles, particularly respiration physiology, thermai relations, and ion and water balance. Discussions of amphibian energetics by Sede (1987) are applicable to severai subjeas presented below. Staging notations are from A. C. Taylor and Koiiros (1946) in Roman numerais and Gosner (1960) parenthetically in Arabic numerais (and see table 2.1).

In the Arnphibia, as wei as in other groups of anirnals, it often seems that the larvae receive less attention than the adults.
R. M. Savage 1962:24

Respiratory Physiology
It only has been within about the last 20 years that an appreciable amount of research has been done on the respiration physiology of tadpoles. As with other vertebrates, tadpoles need to take up oxygen, eliminate carbon &oxide, and maintain an appropriate acid-base status. Beyond this rather generic assertion, further generalizations are risky because of the diversity of rnicrohabitats in which tadpoles are found and the variety of gas exchange organs that are present. Furthermore, most studies of gas exchange and acid-base baiance have been done on North Arnerican species, and the majority of those have been on the bulifrog (&na catesbeiana). Finally, there are apparently physiologicai clines (e.g., hemoglobin funaions) occurring in this wide-ranging frog

190

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

that require caution about drawing general conclusions even about this single species. This review, therefore, deals largely with ranid tadpoles, but the reader must bear in rnind the impressive variety of tadpoles, particularly those in the tropics (e.g., Duellman 1970; Inger 1985), about which we have practicaliy no physiological information.

Hemoglobin Functwn
Hemoglobin (Hb) greatly increases the 0, carrying capacity of blood. Reports of the 0, capacity of bullfrog tadpole blood range from 7.8 (McCutcheon 1936) to only about 3 v01 % (Pinder and Burggren 1983). McCutcheon (1936) first reported that the P,, of tadpole (bullfrog) H b was considerably lower than that of an adult, although there is not enough information to state this as a general case for anurans. A low P,, is presurnably adaptive for an animal that is primarily a water breather, especialiy if it lives in vegetated areas that are prone to seasonal and diurnal hypoxia (Nie et al. 1999; Ultsch 1973, 1976a). McCutcheon (1936) also reported a negative Bohr shift in tadpole H b contrary to the positive Bohr effect found in most animals, including adult bullfrogs. A shdt to the left of the 0, dissociation curve is adaptive if an animal lives in a hypoxic environment that is also hypercarbic. This s h h enhances 0, loading from the hypoxic environment, and hypoxic aquatic environments are often hypercarbic (Ultsch 1973, 1976a). However, the presence or absence of a negative Bohr shift in tadpole H b solutions is dependent upon the buffer solution used (Aggarwal and Riggs 1969; K. W. K. Watt and Riggs 1975), and more important, the Bohr shift of whole blood is positive (Johansen and Lenfant 1972; Pinder and Burggren 1983). Thus, the potential ecological adaptive value of a negative Bohr shift is apparently not realized in whole animal~. McCutcheon's (1936) finding of an ontogenetic increase in P,, has been corroborated for H b solutions (reviews by Broyles 1981; Riggs 1951; SulLivan 1974) and for whole blood (fig. 8.1; Hazard and Hutchison 1978; Johansen and Lenfant 1972; Pinder and Burggren 1983). As is the d e in blood from other species, the P,, of whole tadpole blood is considerably higher than that of stripped H b solutions because of the presence of cofactors such as nucleoside triphosphates (Aggarwal and Riggs 1969; Sullivan 1974). However, the increase in P,, from tadpole to adult is not related to organic phosphates, which decrease with metamorphosis (Hazard and Hutchison 1978; Johansen and Lenfant 1972; Pinder and Burggren 1983) and is caused by larval H b being abruptly replaced with adult H b at about Taylor/Kollros XXII-XXIV (Gosner 4 4 4 5 ; fig. 8.2; Just and Atkinson 1972; Moss and Ingram 1968b; Theil1970). The hemoglobins of both larval and adult buiifiogs are composed of several components (Just and Atkinson 1972; Moss and Ingram 1968a; Wise 1970). Of the four components (I-IV) of bullfrog tadpole blood (K. W. K. Watt and Riggs 1975), components I and I1 predominate in young tadpoles and 1 have higher 0, affinities than components 1 1and IV, which predominate in older tadpoles. Because the lungs of young bullfrog tadpoles are less developed than those of older ones

O
O

20

40 60 80 OXYGEN TENSION (mrn Hg)

1O0

RED CELL 30ORGANIC PHOSPHATES 10 (PMIs Hb)

TAD

INT

ADULT

Fig. 8.1. Oxygen dissociation curves and erythrocyte organic phosphate concentrations of tadpoles (TAD),intermediate stages (INT), and adults of Rana catesbeiana; n = calculated Hill coefficient. After Johansen and Lenfant (1972).

(Atkinson and Just 1975), the higher affinities could be an adaptation to aquatic hypoxia for a stage that is more dependent upon aquatic 0, uptake. There seems to be much less flexibility in H b function in tadpoles than in adult frogs. Pinder and Burggren (1983) subjected bullfrog tadpoles and adults to 28 days of combined aerial and aquatic hypoxia (70-80 rnm Hg; 20"23C) and compared changes in respiratory properties of the blood to normoxic controls. There were no changes in hematocrit, RBC count, H b concentration, mean corpuscular H b concentration, 0, capacity, Bohr effea, Hill's coefficient, or intraerythrocytic concentrations of nucleoside triphosphates (ATP + GTP, and 2,3-DPG); a slight decrease in P,, from 9.2 to 7.0 mm H g was discounted as physiologically unimportant. In contrast, adults showed significant increases in hematocrit, a doubling of 0, capacity, and a decrease of 11mm H g in the P,,. The authors concluded that the adult responded to hypoxia by altering the respiratory properties of the blood, while tadpoles enhanced the efficacy of the gas exchange organs by changes in their morphology (Burggren and Mwalukomo 1983; see below). The fact that the responses appeared in adults, which do not experiente hypoxia because they breathe air, but not in tadpoles, which live in an aquatic environment much more prone to hypoxia, is curious. Perhaps adaptations for adults are related to underwater hibernation as a compensation for their high P,, relative

PHYSIOLOGY

191

to that of tadpoles, but because these experiments were done at 20"-23"C, further experiments at low temperatures are needed to test this hypothesis. A final comment deais with the caveats mentioned above. Essentially ali of the studies of H b function have used bulfiogs, and the sources of the animais have varied. A number of studies (particularly earlier ones) involving ontogenetic comparisons used tadpoles from one area and adults from another, rnixed sources of animais, or used animais with unknown latitudinal origins (e.g., McCutcheon 1936; Moss and Ingram 1968a, b; Riggs 1951). Significant physiologicai clines relating to respiration and acid-base balance have been demonstrated in other ectotherms (e.g., turtles, Ultsch et ai. 1985), and similar clines may exist in anurans. For example, Moss and Ingram (1968a) noted that tadpoles from different suppliers frequently showed differences involving the major H b components, and Agganvai and Riggs (1969) found that both the propomons of the H b components and their arnino acid compositions varied with the origin of tadpoles. Therefore one must largely restrict conclusions from studies to date about H b function to bulfrogs and then be cautious about particulars.

Oxygen Uptake
Gatten et ai. (1992) reviewed the literature on metabolic rates of amphibian larvae (see aiso Lovtrup and Werdinius 1957) and concluded that only at 20C was there enough data to permit generalizations. They found that the routine metabolic rate was essentialiy the same for anuran and saiamander larvae and that the pooled estimate of the metabolic rate of amphibian larvae was about 50% higher than that of

adult saiamanders and about 33% lower than that of adult frogs. The intermediate position of the amphibian larvae rnight be partialiy caused by the relatively low metabolic rates of salamanders, combined with (at least for anuran larvae) the high water content and high gut volumes of tadpoles relative to the adults. A number of studies address the relationship between stage of development in tadpoles and VO,, but it is risky to generalize. The majority of studies were done under conditions that would not result in determinations of routine or standard VO, (e.g., in respirometers where animais were in vessels that were shaken continuously; Fletcher and Myant 1959; Funkhouser and Foster 1970; Funkhouser and MUs 1969; Sivula et ai. 1972; Wds 1936) or in a flow-through respirometer where the excurrent PO, (therefore aiso the PO, about the animal) was hypoxic (Etlun 1934). Most studies did not correct for the effects of the large increase in body size as development proceeds, for changes in the propottion of body mass as water, or for changes in the relative contribution of the gut and its contents. A number of studies did not give the tadpoles access to air. Feder (1982) recognized and avoided most of these objections in his study of the relationships of body size, developmentai stage, and VO, in four species of tadpoles at 25C; he concluded that dry mass accounted for most of the variation (59%-90%) in VO,, at least up to metamorphic stages (Taylor/Kollros XX-XXV, Gosner 4 2 4 6 ) . Feder (1981) aiso found that 91% of the variance in VO, could be attributed to dry mass in Xenopw l& tadpoles at 20C. Among species, there is some indication of adaptive differences in the metabolic rates of larvae, but data are not plentiful. Bufi tadpoles, for ex-

I
I

:t + m/------@
1

d ~ +

Adult type
------+--------Q
I
I I I

B
66
I

Q
I

P
I

'
I I I I I H H +

XV

XX

1 WK

7WK

ADULT

DEVELOPMENTAL STAGE
Fig. 8.2. Relative proportions of larval- and adult-type hemoglobins as a function of developmental stage (Taylor/Koilros) of Rana catcsbeiana. After Just and Atkinson (1972).

192

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

arnple, have relatively h g h metaboiic rates even when con- D, fig. 8.3) without any change in 0, conduaance. If the sidering their small size (Feder 1982; Noland and Ultsch animal can reduce its SMR by halfand double its 0, conduc1981), which could be considered adaptive if a high meta- tance simultaneously, then it can lower its Pcby a factor of 4 bolic rate results in a faster developmental rate that would (25 mm Hg, p o h t C, fig. 8.3). Assurning that these are the mximal possiblc adjustments in SMR and 0, conductance, d o w them to escape from a drying pond. Gattcn et ai. (1992) concluded that the average Q,, for the range of potential adjustments in P, resulting from ail anuraii larvae was 2.14 over the temperature range of 15"- combinations of changes i SMR and 0, conductance is den 25C but did not generalize about Q values over lower fined by area ABCD. , Decreases in SMR (metabolic depression) of ectotherms temperature ranges. It would be advantageous for the Q , to increase corisiderably as temperature falls, particularly at can be induced by a nurnber of factors, including diving (in temperatures associated with overwintering (i.e., a "meta- air-breathing animais), lowering temperature, torpor and hibolic depression" to considcrably below that expeaed from bernation, and hypoxia or anoxia (reviews by Hochachka the usual value for Q,, of 2-3 observed at higher tempera- 1988 and Ultsch 1989). Mechanisms normaiiy involve biotures; see Ultsch 1989). A high Q,, would indicate a meta- chemical regulation of rate-limiting enzymes (review by Hoboiic depression that would be adaptive to both a reduced chachka and Somero 1984). Increases in the whole-body 0, food supply (if ovenvintering tadpoles feed) and the poten- conduaance can be attributed to any mechanisms that tiaily severe hypoxia that can occur in northern ponds during can favorably alter the rate-iimiting step in the transport of the winter. There is some indication that metabolic depres- 0, from the extemai environment to the cell. Potential sion occurs in tadpoles at low temperature (as indicated by O,-exchange limiting mechanisms have been grouped into a h g h Q,,; Bradford 1983 and G. E. Parker 1967), but the those involving the movement of 0, across the surface of case for a metabolic depression associated with low tempera- the external gas exchanger (diffision lirnitations), delivery of ture, hypoxia, or both, is not compeiling. There is a meta- blood to the external gas exchanger (perfusion limitations), bolic depression during starvation in tadpoles (Bradford and or circulatory limitations between the external gas exchanger Seymour 1985), and adult anurans and other arnphibiaris and the ceiis (convective limitations; reviews by Feder and reduce their metabolic rates 70%-82% whiie estivating Burggren 1985; Piiper 1982; Piiper and Scheid 1977). As mentioned above, some evidence indicates that meta(Seymour 1973; Van Beurden 1980; Whitford and Meltzer 1976; review by Pinder et ai. 1992). Therefore it is likely bolic depression may occur in tadpoles in response.to low that there is an increase in Q in tadpoles at low tempera- temperatures during ovenvintering. Most tadpoles that , overwinter are ranids, which during warmer seasons can use tures (also see Thermal Relations). Among vertebrates that are exclusively water breathers their lungs to supplement aquaticO, uptake from hypoxic (e.g., fishes, aithough the following discussion is also rele- water. However, ice covcr during the winter would force the vant for air-breathing vertebrates), it is typical for VO, to be tadpoles to rely exclusively upon dissolved O,, and a metairidependent of water PO, over some ecologicaiiy relevant bolic depression would be highly adaptive. Whether a metarange from the PO, of air-saturated water downward. Such bolic depression in response to diving occurs in tadpoles is animais are caiied metabolic 0, regulators, as contrasted open to question and may deperid on the importante of air with metabolic 0, conformers whose VO, declines steaddy breathing to a given species. Feder (198 1)found that Xemwith declining ambient PO,. The range of naturaiiy oc- pus laevis tadpoles reduced VO, by 20%-40% during forced curring PO, over which metabolic 0, regulation can occur submergence at 25C relative to conuols with access to air, is one measure of the degree of adaptation to hypoxia. The while there was no eEect of forced submererrice on thc VO. V PO, at which regulatiori is no longer possible, and VO, of Rana berlander larvae. l n later studies on tlle samc two starts to decrease with further lowering of ambient PO,, is species, there was no decrease in the VO, during forced subtermed the criticai 0, tension (PJ. The lower the Pc of an mergence of either species relative to those with access to air animal, the better adapted it is to hypoxic water. At ambient (Feder 1983a; Feder and Wassersug 1984). Thus thc more 0, tensions below the P,, the relationship between VO, and usual finding is no change in total VO, with submergence at PO, is usudy dose to linear. The slope of this linear rela- near-normoxic PO,. Considering the relatively smaii contritionshp between vO, and PO, below P, has bcen termed bution of puimonary respiration to total VO, in normoxic the 0, transport capacity of the animal (Beckrnbach 1975) water (Burggren and West 1982; Feder 1983a; Feder and or the whole-body 0, conductance (J. T. Duke aiid Ultsch Wassersug 1984), this is not a surprising result. However, in 1990); the latter was named in anaiogy to the thermai con- hypoxic water (less than about 70 mm Hg), the aquatic VO, duaance of endotherms. of Rana berlandim' and Xenopw uevis tadpoles is much reChanges in metabolic rate and whole-body 0, conduc- duccd in forcibly submerged animais, and access to air does tance can effect changes in P, (fig. 8.3). Consider an animal not raise the total VOZto control (normoxic water with acthat is a metaboiic 0, regulator with a given standard meta- cess to air) levels (Feder 1983a; Feder and Wassersug 1984). bolic rate (SMR) and a P, of 100 mm H g (point A, fig. 8.3). Thesc findings suggest that the set point for aquatic VO, If the animal can double its 0, conductance, P, wiil reduce is lowercd. The possibility that such a controlled response by halfto 50 mm H g (point B, fig. 8.3) without any change occurred is supported by the observation that ifRana berlanin thc SMR. Alternatively, if the animal can reduce its SMR dieri tadpoles were exercised in water with a PO, of 44-70 by one-half, P, wdl stiii reduce by half to 50 mm H g (point rnrn Hg, the total VO, averaged 80% of control values and

PHYSIOLOGY

P 0 2 at AIR-SATURATION
Z

-+

c
C

SMR

(3

SMR

AMBIENT OXYGEN TENSION (mm Hg)

Fig. 8.3. Theoretical effects of changes in the standard metabolic rate (SMR, expressed as 0, consumption) and whole-body 0, conductance (O,C, calculated as the slope M of the left portion of the relationship between 0, consumption and ambient 0, tension) on the criticai 0, tension (P,). A given P, can be affected by changes in SMR and by changes in 0,C. For example, P, at 100 mm H g (A) can be reduced to 50 mm H g by either haiving the SMR (D) or doubling the 0,C (B). The PO, of air-saturated water is shown as an approximate range for 0 - 0 C See text for detds. '4.

reached as much as 200%, indicating that the ability to take up O, at control levels was still present (Feder 1983a). Whether tadpoles can increase their whole-body 0, conductance for auatic resoiration has not been s&did directly, but indirect evidence suggests they can. A number of short- and long-term strategies are presumably avadable to them, including cutaneous capdlary recruitment (review by Feder and Burggren 1985), adaptive changes in the morphology of the skin and gis (see below), and modifications of hemo~lobinfunction. If tadooles can aiter their wholebody 0,~onductance(e.g., as the salamander Siren lacertina; J. T. Duke and Ultsch 1990), those individuais within a species living in routinely hypoxic environments may have lower P, vaiues than those from routinely normoxic environments. The effect of declining water PO, on aquatic VO, is of interest even if tadpoles can make up a decrement in aquatic VO, by pulmonary respiration, because surfacing increases predation risk (see below). Thus, a low P, wdl be ecologicaiiy adaptive if water is the only source of O, (determined experimentally by forced submergence), and one might expect to h d the lowest values for P, among species that frequent hypoxic waters. In North Arnerica these situations generdy confront species that breed in permanent ponds. Conclusions drawn from the relevant data (table 8.1) should be ten-

tative. One qualitative conclusion that seems fairly well supported is that the Pcof tadpoles is higher than that of fishes (Ott et ai. 1980; Ultsch et ai. 1980b, 1981a) and lower than that of most aquatic saiamanders (J. T. Duke and Ultsch 1990; Ultsch 1976b; Ultsch and Duke 1990), indicating that the efficacy of tadpole mechanisms for breathmg water are intermediate between the two. There is a trend toward a lowering of P, with fding temperame, which could be expected because of the decreased SMR (fig. 8.3), but data are sparse. The Pc would be expected to increase with body size as the surface-to-volumeratio f d s in tadpoles ifthe skin is an important gas exchange site (see below). This was found to be the case by Feder (1983b) for resting Bufo woodhousii, but the results are mixed for B. t e w e h and Rana sphenocephala (Noland and Ultsch 1981). Moreover, Feder (1983b) found no relationship between body size and Pc in R. bwlandieri, but Crowder et ai. (1998) found the P, of R. catesbeiana tadpoles to be relatively constant with body size through Taylor/Koilros XVI (table 8.1). More data relevant to this point for species with late lung development would be especiaiiy interesting. Gatten et ai. (1992) cded attention to the paucity of physiologicai data on anuran larvae in their review. They reported that the ability of tadpoles to increase mass-specific aerobic rates of 0, consumption (i.e., metabolic scope) is

194

G O R D O N R . U L T S C H , D A V I D F. B R A D F O R D , A N D J O S E P H F R E D A

Table 8.1 Estimates of criticai oxygen tension (P,. mm Hg) of various water-breathing tadpoles relative to developmental stage (Taylor/KoUros 1946), body mass (g), and temperature ("C) Taxon Stage
II-VI1 IX-XIV II-XII M-xv

Mass

Temperature

pc

Referente
Noland and Ultsch 1981 Noland and Ultsch 1981 Noland and Ultsch 1981 Noland and Ultsch 1981 Feder 1983b

Bufo terrestrtj

Bufo woodhousii Pseuduphyne bibronii

Bradford and Seymour 1988b

Rana berhndien'
I-XV

Feder 1983a Crowder et al. 1998 Crowder et ai. 1998 Crowder et al. 1998 Crowder et ai. 1998 Crowder et ai. 1998 Crowder et ai. 1998 Crowder et al. 1998 Crowder et ai. 1998

Rana mtesbeiana

Rana muscosa
Bradford 1983

Rana pipiens
Helff and Stubblefield 1931

Rana sphenocqhah
I-v M-m I-v M-xrv

Noland and Ultsch 1981 Noland and Ultsch 1981 Noland and Ultsch 19.81 Noland and Ultsch 1981 Feder and Wassersug 1984

x " Ucl& eP .
I-XX a D qmass. bEstimatedfrom increases in lactate concentration and visual inspecon of graphs. <Estimated from increases in iactate concentration after 6 hours submergence.

largely size-independent at 20 and 25"C, compared to the usual dependence of mass-specific routine metabolic rate on body mass. The differing relationships between body mass and metabolic rate for resting and activity VO, means that one cannot make a general staternent about the amount by which a tadpole can increase its metabolic rate with activity from general equations for the two situations. Such a comparison can be made for lizards (activity VO, about 6 times the resting VO,; Bennett and Dawson 1976) and salamanders (average increase about 5 times; Gatten et ai. 1992). O n a species-by-species basis, the average increase of 12 times in aduit anurans indicates an efficient oxygen transport capacity. Apparentiy tadpoles are less capable of increasing their VO, than adults, with factorial increases of about 3 times, but this is a preliminary conclusion based on only two studies (Feder 1983b; Quinn and Burggren 1983). The abihty to increase VO, may be of l i d e ecological necessity because most tadpoles are not continuously active. Even when stirnulated to high rates of activity when escaping predators, the usual strategy is t o dash away and hide. Such burst activity requires only a few seconds at most, and because an instantaneous acceleration is involved, these ac-

tivities are most efficientiy powered by available stores of ATP and other high-energy compounds such as phosphocreatine. The regeneration of these compounds can then be accomplished aerobicaliy without the need for lactate production. Gatten et al. (1984), for example, found that tadpoles of three species had little lactate production after 30 sec of intense swirnrning at 4"-27C; tail muscle phosphocreatine fel1 t o nil in the one species in which they measured it (Rana catesbeiana). While it is likely that tadpoles rarely accumuiate lactate in the field as a resuit of activity, they d o have a capacity for lactate anaerobiosis, which may be used when water becomes hypoxic. Anaerobiosis during hypoxia wouid be espec i d y important among tadpoles that ovenvinter, when ice cover prevents air access (e.g., several species of ranids, Collins and Lewis 1979; review by Pinder et al. 1992), and in species that do not ovenvinter but have poorly developed lungs (e.g., bufonids). The role of anaerobiosis during ovenvintering among amphibians, larva1 or aduit, has not been studied. In laboratory experiments, aduit frogs are intolerant of anoxia relative t o other aquatic ectotherms, even at temperatura as low as

PHYSIOLOGY

195

S0C (review by Pinder et ai. 1992). Yet Friet and Pinder (1990), in a preliminaty study of hibernada of adult Rana catesbeiana in Nova Scotia, found that the PO, of the water kvivithui 50 cm of the bottom was as low as 5 mm Hg, with a sustained PO, of less than 20 mm H g for more than 8 wveeks. Their findings~ suggest that bullfrogs are much more tolerant of severe hypoxia in the field than would have been deduced from experimental studies. Whether this confia is because of different origins of animais, acclimation to progressive hypoxia, or some other effect remains to be seen. Because tadpoles tend to be more tolerant of low PO, at low temperatures than adults (Bradford 1983), it is likely that tadpoles are aiso more capable of anaerobiosis and perhaps of tolerating the resultant lactic acidosis. However, the physiologicai mechanisms that permit overwintering among tadpoles have not been studied. There have been severai studies of activity metabolism in tadpoles, including the role of anaerobiosis. While burst activity seems to be largely aerobic in tadpoles (Bennett and Licht 1974; Gatten et ai. 1984), continuous exercise can lead to appreciable lactate accumulation. For example, Bennett and Licht (1974) found that 30 sec of rapid swimming by buiifrog tadpoles at 20C increased the resting lactate concentration by oniy 33% but that 10 min of slow swimming raised laaate concentrations by 270%; by exercising the tadpoles to exhaustion, Quinn and Burggren (1983) induced bullfrog tadpoles to accumulate lactate to levels twice as high as those in the Bemett and Licht (1974) study. Quinn and Burggren (1983) conduaed a thorough anai-

ysis of the consequences of exhaustive exercise in buiifrog tadpoles and adults. They found that blood and muscle lactate concentrations at rest were about the same in both stages and that muscle laaate concentration increased by 5-6 fold in both after exercise. Adult blood lactate concentration peaked wivithui 1 min &er exercise, but it took 30 min to eak in the larvae. which indicates a slower release to the blood, reconversion in situ, or both, rather than by export to the liver. Adults showed the usual increase in VCO, that results from lactic acidosis when bodv bicarbonate stores are titrated, but this was not the case for larvae (fig. 8.4), or at least it appeared so when the first measurements were taken 45 min after exercise. However. the ~ersistent low vaiue for the respiratory quotient (fig. 8.4) suggests that body CO, stores were being replenished (i.e., repayment of a "CO, debt"). This in turn suggests that there may have been a postexercise peak in VCO, that had dissipated by the time of the first VCO, measurement. Tadpoles may be able to eliminate excess CO, more rapidly than adults because they are smailer and because they c G use lungs, gilis, and skin for CO, elimination. By having multiple sites for CO, exchange, the tadpoles may aiso be less susceptible to respiratorv acidosis during: exercise. " The importance of lungs to tadpoles in hypoxia is obvious, but they are aiso important during activity. Starnina increases with access to air even when swimminp:in normoxic water, and hypoxia decreases stamina more in lungless Bufo americanus tadpoles than in the lunged tadpoles of Rana berlandievi and Xenupus h s and (Wassersug and Feder 1983).
V

TIME ( h )
Fig. 8.4. Respiratory responses of tadpoles of Rana catesbeiana. (Uppw)Respiratory exchange ratio (R = V C O ~ ~ Oat )rest and after exhaustive exercise at 23C; the hatched area indicates a sustained decrease in R. , &mm) Means + S. E. of 0, consumption (sqrutres)and CO, elimination (Crtan~les) under the same conditions; solid symbols indicate valurs that increase significantlyfrom resting levels. After Quinn and Burggren (1983).

196

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA where M is the rate of gas transfer, D the diffusion coefficient, S the effective surface area for gas exchange, dP the partial pressure gradient across the skin, and 1 the distance over which diffusion must occur (water [or air]-to-blood barrier). It has been a p e d that gas exchange across the integument of amphibians is primarily diffusion-limited (Clemens and Feder 1990; review by Feder and Uurggren 1985) and that &f can be regulated only by varying dP (e.g., changing the internal gas pressures or moving to a more favorable respiratory microenvirocment). However, Burggren and his coworkers argued that at a given dl: M can be regulated by changing the effective values of either S or I. They proposed that the effective cutaneous exchange area can be varied by capillary recruitment and derecruitment (Burggren and Moalli 1984; Feder and Burggren 1985; also Feder et al. 1988) and that the apparent thickness of the diffusion path can be decreased by convection that reduces the thickness of the boundary layer (Burggren and Feder 1986; Feder and Pinder 1988); this layer is potentially important in limiting cutaneous gas exchange (Booth 1990; Pinder and Feder 1990). Whether such cutaneous regulatory changes occur, and whether they are effective in altering cutaneous is an open question. For example, the cutaneous VO, of the skin of aduit Rampdpiens in water with access to air fell as water PO, fell (Vitalis 1990), which implies that mechanisms avaiiable for increasing cutaneous ~ 0 , during progressive hypoxia were not effective. Because the frogs could breathe air, it could be argued that the shortfall in cutaneous VO, was sirnply made up by an increase in pulmonary 0, uptake, whlch negated any need for cutaneous regulatory mechanisms to be brought into play. Ln contrast, the much larger and predominantly skin-breathing salamander Cqptobran&us allganknss can maintain its cutaneous VO, over the same range of water PO, (Ultsch and Duke 1990) that the frogs in the study by Vitalis could not (or did not). Therefore, among the amphibians in which the question of the regulation of cutaneous in water has been addressed, there is no definitive answer; data on tadpoles are lacking. Apparently changes in the rate of gas transfer, particularly 0, uptake, can occur among aquatic amphibians over both short (Feder et al. 1988; Malvin and Hlastala 1986a, b, 1988) and long terms (J. T. Duke and Ultsch 1990), but mechanisms and their control remain conjectural (Malvin and Riedel1990; Pinder et al. 1990). The only study involving tadpoles (Burggren and Mwalukoma 1983) showed that chronic exposure of bullfrog tadpoles to hypoxia resulted in a thitlning of the skin, a d o u b h g of capillary mesh density, and a reduction of the blood-water barrier by one-hall. Anurans develop external gills prior to hatching and retain them for several days after hatching. The external gills arc replaced by the internal g a s coincident with the development of the operculum; the internal gills are lost by Taylor/ Kollros XXIII (Gosner 45; Malvin 1988). In bullfrogs between Taylor/Ko1lros V and XX (Gosner 30-42), the gills are highly vascularized and remain a constant proportion of body mass; by Taylor/Kollros XXII (Gosner 44) the importance of the gills is diminished as resorption starts and the

However, the ecological importance of the ability to accumulate lactate is most likely related to hypoxia tolerance. From this viewpoint it is interesting that Feder (1983b) found that Bufii woodhousii tadpoles swimming to exhaustion accunlulated no more lactate than those at rest in water with a PO, of 19 rnm Hg. The effects of prolonged elevations of lactate dur'ig chronic hypoxia have not been studied.

Gas Exchan~e Partitioning Tadpoles, like many aquatic salamanders, may have several sites of gas exchange (e.g., gills, skin, and lungs) and therefore may exchange gses with air, water, or both. The extent to whlch gas exchange is partitioned among sites and between media depends on a host of factors, including the presence or absence of lungs, body size, respiratory gas concentrations in the water (at the moment and over time), temperature, stage of development, level of metabolism, and microhabitat (e.g., Jia and Burggren 1997a, b). To understand the complexities of gas exchange in tadpoles, one must consider the morphological characteristics of the gas exchangers and their vascular supply. A number of species, such as all bufonids studied to date and some torrent-dwelling species (Feder 1981; Nodzenski et al. 1989; Wassersug and Heyer 1983), have rudimentary lungs or do not develop lungs until metamorphosis. These species are likely more dependent on cutaneous gas exchange than those with lungs. Microhyiids (R. M. Savage 1952; Wassersug and Pyburn 19871, some leptodactylids (Wscrsug and Heycr 1988), and some pipids (Malvin 2988) have a reduced development of the respiratory portions of the gills, which Ikely lowers the ability of the gills to exchange gases with water. Ranids have all three sites fu~tctional from an carly age (R. M. Savage 1952). Because bullfrogs and other ranids are most commonly smdied, the discussions herein are heavily weighted toward that group. Probably the only universally important gas exchange organ among tadpoles is the integument. Both terrestrial and aquatic tadpoles possess d ~ typical gas-permeable skin of e amphibians, and all amphibians exchange gases, especially carbon dioxide, through it. The degree of gas exchange cannot be predicted directly from morphology. For example, the importance of the skin of tadpoles as a gas exchanger for oxygen has been questioned by De Saint-Aubain (1982, 1985) and McIndoe and Snlitll (1984b); they noted that cutaneous vascularization is sparse relative to that of aquatic salamanders and adult anurans. However, tlle skin can be an important site of 0, uptake (see below). There also may be no coupling between the degree of skin vascularization and the presence or absence of lungs. De Saint-Aubain (1982) found no morphologtcal Merences in the cutaneous vascular system between the tadpoles of Bufo b u . and Ram tempmaria, although the latter has well-developedlungs and the former has only rudimentary lungs (R. M. Savage 1952) and presumably is dependent upon aquatic respiration. Perhaps the small size of Bufo larvae negates any need for an increased cutaneous 0, conductance. In considering the potential of the skin as a gas exchanger, Fick's law of diffirsion is appropriate: M = D(S)(dP)/I,

m,

m ,

PHYSIOLOGY

197

sill capdiaries move deeper into the epithelium (Atlunson in hypoxic water, and the rnicrohabitat of ranid tadpoles can
& i 1975). Just

It is adaptive for aquatic amphibians to have variable perh i o n patterns that are related to environmental 0, tensbm. Tadpoles in weli-oxygenated water exhibit cardiorespiraton synchrony when at rest (Wassersug et al. 1 9 8 1 ~ )In . merely hypoxic water, when the lungs are the primary organ of 0, uptake, 0, would be lost to the water if gdi perfu&n nTasinevitable. Most, if not a, gilled aquatic saiamanbers have shunts that either direa blood from the ventral to the dorsai portions of the aortic arch before it enters the gili kmeliae or bypass the respiratory portion of the larneliae by h n i n g directly from the afferent to efferent branchiai arteris (C. L. Baker 1949; Darneli 1949; Figge 1936; Malvin e 1988; Malvin and Boutilier 1985). While the physiologic i data are not as extensive as for saiarnanders, some tada poles possess giil shunts (De Saint-Aubain 1981, Bufo bufo md Rana temporaria; McIndoe and Srnith 1984b, Litmia mhdi), and tadpoles likely exert control over the degree of shunting (De Saint-Aubain 1985). As with saiamanders Figge 1936; Malvin 1985a, b), the proportioning of blood h n - between the respiratory portions of the gills and the stiunts is presumed to be controlled by selective vasoconariccion, vasodilation, or both, perhaps under the influence ot'circulating catecholamines. , with external giils among other amphibians (Bond h 1960), the degree of development of anuran internal giils is responsive to the concentrations of dissolved respiratory ( b v t r u p and Pign 1968). Bulifrog tadpoles in Taybr;'Koiiros XV-XX (Gosner 40-42) exposed to a water PO, ot'7C80 mrn H g for 4 weeks increased the size and nurnber of'gili iiaments; chronic hyperoxia (387-430 mm Hg) had w d e a (Burggren and Mwaiukoma 1983). Although tadpoles in hypoxic water in that study had access to a gaseous phase, the gas had the same PO, as the water, so it is uncerQin how great the morphologicai response of the giils would have been had the tadpoles had access to normal air, espea d y because they were at a stage where the lungs are weli k e l o p e d . Nevertheless, the presence of a morphological response capability was evident. The lungs of bulifrog larvae gain weight frorn Taylor/ Koilros V-XV (Gosner 30-40); afier that, lung absolute neight remains relatively unchanged, but their relative a-eight increases as body weight falis (Atkinson and Just 1975). These authors reported the degree of lung septation ro be minor at Taylor/Koliros X (35) and increasing considm b l y by stage XXII (44) as tail resorption progresses. This does not mean that the lungs are not functionai in early stages, because even early-stage bulifrog tadpoles surface occasionally to breathe and routinely have gas-filled lungs (Crowder et al. 1998; Just et ai. 1973; also Orlando and Pinder 1995). The ability to breathe air is not cruciai to developrnent or survival in bulifrog larvae through Taylor/Koliros XXI rGosner 43) when submerged in normoxic water at 20C, but later stages wiil drown (Crowder et al. 1998; Just et al. 1973). It appears that ali stages breathe air (Crowder et al. 1998; Just et al. 1973). The utility of breathuig air is obvious

become quite hypoxic (Nie et ai. 1999; Noland and Ultsch 1981). The reason tad~oles breathe air in normoxic water may be to keep the lunis inflated; ifthey did not breathe air, the lung gases would be lost as 0, and then N, followed the diffusion gradients out of the lung. This situation is similar to that of diving insects that use a gas bubble as a f f i s i o n giil (K. Schrnidt-Nielsen 1983). Lungs disseaed from premetamorphic bulifrog tadpoles that have been living in normoxic water with access to air float (Crowder et al. 1998; also Just et ai. 1973), but if the animals are denied air access for a day, the dissected lungs sink, which indicates that most or ali of the lung gases have been lost. This reasoning is supported by the observation of West and Burggren (1982) that bulifrog tadpoles in Taylor/Kollros XVI-XIX (Gosner 4041) did not ventilate their lungs during 1 hr of hyperoxia (about 600 mm Hg), while ventilation (albeit low-leve1 at l/hr) did occur in normoxia. A possible explanation is that the lungs did not deflate in hyperoxic water; had they done so, the tadpoles probably would have breathed. West and Burggren (1983) also showed that pulmonary stretch receptors must be in the lungs. The loss of lung gases has aiso been suggested by Feder and Wassersug (1984); Xenopus L m i tadpoles in normoxic water became negatively buoyant afier 10-63 rnin of submergence, and a negative hydrostatic pressure of - 1atrn would not cause these tadpoles to release an air bubble. A r would be released from the lungs of tadpoles kept in normoxic water with access to air if they were Similarli tested. Burggren and Mwalukoma (1983), in the studies of rnorphologicai changes in hypoxia-acclimated Rana catesbezana tad~oles discussed above. found an increased lung volume and cava density in response to hypoxia. Hyperoxia had no effect on tadpoles, and neither hypoxia nor hyperoxia affected adults. Because the animals were breathing both air and water of the same PO,, it is not certain what the changes in lung structure in the tadpoles would have been ifthey had been k e ~in hwoxic water with access to normoxic air. The t f a a tha; adultS8howed no changes may be caused by their higher degree of lung development; that is, 70-80 rnm H g may not be a severe enough hypoxia to present a significant respiratory chaenge. Tadpoles of D i s c ~ l o s s u s p exposed i to 426 and 710 rnm H g 0, for 15-30 days in both the air and the water phase had greatly suppressed lung development (De Quiroga et ai. 1989). The degree of suppression was greatest in earlier stages and least in newly metamor~hosed animals. which led to the conclusion that the effea I was prirnarily one of delayed lung development caused by the excess aquatic oxygen supply. These authors suggested that the reason they found a response, whde Burggren and Mwaiukoma (1983) did not, was that the PO, used in the latter study was not high enough to trigger a response; it is aiso possible that the difference in responses was speciesspecific. Finally, it is not known if the lungs are merely accessory respiratory mechanisms that enable a tadpole to live in water with a PO, below the P, determined during forced submergence experiments (i.e., the "ecological Pc" is less than the
L 2

198

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

.>"
40

w.. ..
....,...
(IV-V) (XVI-XIX) -.A~ills

Fig. 8.5. The relative contributions of giiis, skin, and lungs to gas exchange in tadpoles and adults of Rana catesbeiana at 20C. (Ufler) O, uptake (vo,). (Lower) CO, elimination (vco,). Tadpoles and metamorphosing individual staged according to Taylor and Kohos (1946). Dotted lines connect to presumed values. After Burggren and West (1982).

"experimental Pc"; see Nie et ai. 1999), or if the lungs can be sufficient in an environment where the water is chronic d y and severely hypoxic. Saiamanders that live in such environments (e.g., Amphiuma, Siren) have well-developed lungs and can inhabit areas where the water remains virtually anoxic for months (Ultsch 1976a). Some lunged tadpoles can tolerate extended hypoxia (e.g., Lannoo et ai. 1987, Osteopilus bbrnneus), but whether they can tolerate continuous anoxia is unknown (see Costa 1967). Aquatic tadpoles l a c h g or with poorly developed lungs partition gas exchange between the skin and the buccopharyngeai epithelia, with the main exchange surface in the latter being the gills. Bufonid tadpoles, which are essentially lungless, have gills with well-developed respiratory lamellae. It seems to be a general d e that while there are aquatic tadpoles with reduced or absent lungs and others with reduced or absent respiratory lameliae, there are none in which both of these respiratory structures are reduced or absent. Such might be the case if cutaneous respiration is sufficient and the tadpole is unlikely to encounter hypoxia. If such a tadpole exists, one rnight find it in relatively cool torrent habitats and the tadpole wouid be s m d , or both, and therefore have a favorable surface-to-volume ratio for integumentary gas exchange. Arcaphus tadpoles inhabit cool mountain streams, and whde they have reduced lungs, the respiratory portion of their gills is not diminished (Gradwell 1971a, 1973). Bufo tadpoles, which are uniformiy s m d regardless of aduit size (Werner 1986), have rudimentary lungs and fimctionai gills. Tadpoles that obtain oxygen exclusively through the skin are not known, but their nonexistence wouid not seem to be because it is physiologicdy impossible for them to exchange gases solely in this way. Some large aquatic saiamanders are (Ultsch and Duke 1990, Cvyptobranchus) or can be (Shield and Bentley 1973, Siren) predominantlv skin-breathers in normoxic water. N o attemDt has been made to partition gili and cutaneous oxygen uptake in any tadpole that lacks a significant pulmonary capability. Such an experiment on Bufo tadpoles wouid be particuiarly interesting because the sparse cutaneous vascuiature (De Saint-Aubain 1982, Bufo bufo) suggests that the skin is an inefficient gas exchanger. The relative roles of the skin and gilis in gas exchange of lunged tadpoles in water of near-normoxic PO, and with access to air have been studied. Burggren and West (1982) used &na catesbeiana tadpoles and four-year-old aduits, both from Massachusetts; the larvae were of early (Taylor/ Kollros rV-V; Gosner 29-30), intermediate (XVI-XM; 4041), and late (XXIII-XXrV, 45) stages. Late stages had lost their gds, so measurements were oniy of pulmonary and cutaneous exchange; early stage tadpoles were presumed to be mainly aquatic breathers and were not permitted access to air. This constraint made their pulmonary component O%, which is probably an underestimate even in water of high PO,. The authors stated that intermediate-stage larvae are obligate air breathers at 20C. However, bulifrog tadpoles from Aiabama submerged in aerated water at 20C at Taylor/ Kollros V-XII (Gosner 30-37) can develop to at least stage XXI (43) without access t o air (Crowder et ai. 1998). These
i

contrasting resuits suggest an important effect of latitude of origin or developmentai PO, history of the tadpoles. Burggren and West's (1982) study of bulifrog tadpoles showed that the major proportion of the WO,was through the skin in d stages, as is the usuai case in amphibians, and that the cutaneous VO, was dominant over branchial VO, in the premetamorphic stages (fig. 8.5). This effect was not caused by an inefficiency of the gills, which extracted about 60% of the 0, from the water passing over them; this rate compares favot-ably with the giilis of ieleost fishes (but see West and Burggren 1982, where the 0, extraction by bulifrog gilis was oniy 43%). In addition, the high extraction rates were paired with high gill irrigation rates of about 951 min. Experimentaily induced increases in gill irrigation rates which indicates that 0, exdid not raise the branchiai change across the gills was not diffusion-limited. Aiso, the ~redminance c&meous of is unaffected bv increases in totai VO,, at least when the increases are ca&ed by increases in temperature, and remains at about 60%-80% from 15"-33C for both &na berlandieri and R. catesbeiana (Burggren et ai. 1983). One explanation for the high cutaneous VO, may be that a significant proportion of the tadpole tissues (and therefore VO,) are peripherai (e.g., large tail surface) and cutaneous difFusion may suffice for those tissues (Burggren and West 1982; De Saint-Aubain 1985). The most ecologicdy relevant aspect of gas exchange partitioning in tadpoles is between media and not among sites. ~avin~-an ab&ty to use air as an 0, source and c, sink d o w s tadpoles t o inhabit microenvironments from which they might othenvise be excluded by intolerable water hypoxia, hypercarbia, or both. A number of studies have deait with arjal and aquatic gas exchange (especidy 0, uptake) as a fimction of developmentai stage, water PO,, temperature, and degree of activity. Severai studies of &na catesbei-

m,, m,

.>
40 20
...I.____

+--+
o
........
Gills

Lungs

".i_::-----.
Skin

$/ -

Skin

+/+

Lungs ADULT

(XXill-XXIV)

DEVELOPMENTAL STAGE

PHYSIOLOGY

199

ana tadpoles related aeriai and aquatic respiration. West and Burggren (1982) studied Taylor/KoUros XVI-W (Gosner 4 0 4 1 ) tadpoles in water of 140, 82, 43, and 21 rnm H g (see Crowder et al. 1998). There were progressive increases in the frequency of lung ventilation with increasing hypoxia (l/hr at 140 rnm H g to 51/hr, with much variability, at 21 mm He) with an indication that tidai volume aiso increased. Average giil irrigation fi-equency and heart rate were unafFected by hypoxia (aiso Xenupus laevis larvae, Feder and Wassersug 1984; see Tang and Rovainen 1996), but there was a transient decrease d o t h the frequency and amplitude of giil ventiiations in hypoxic water after an air breath. Marian et ai. (1980) found an increase in surfacing frequency with a decrease in water PO, in Euph(yctis cyanophlycts, but in contrast to the studies just mentioned, they aiso found an increase in the frequency of gill ventilation as hypoxia increased. West and Burggren (1983) concluded that both lung idation (via stretch receptors) and the PO, of lung gas a e c t the giil irrigation ninute-volume. They suggested that the decrease in giil irrigation immediately &er an air breath could be adaptive in preventing diffusive 0, loss to the hypoxic water. Feder and Wassersug (1984) examined aerial and aquatic VO, in Xenopus b i s larvae, which lack giil fiiaments and presumably use the eniarged giil filters, buccopharyngeai epithelia, and skm for aquatic gas exchange. In water with more than 100 mm H g of O,, the aerial component of VO, was a relatively minor 17%, as was the case in buiifrog tadpoles (Burggren and West 1982). At a water PO, of 75-85 mm Hg, the frequency of lung-breathng started to increase fi-om control vaiues of about 2/hr to a high of 45/hr at 25-50 mm Hg. When the water PO, became very low, there was an indication of a reduction in overa VO,, demonstrating that the lungs were not able to make up the shortfail in thiaciuatic VO,: * laaate did not increase. There was much scatter in the data, so a suggestion that these results indicate a depression of metabolic rate in hypoxic water must remain tentative. In addition. Wakeman and Ultsch (1976). in similar studies with three species of salamanders ranging from completely to partially aquatic, found that total VO, could be down to a water PO, of 20 maintained in all three s~ecies mrn Hg. Whether a metabolic suppression (or a controiied reduction) occurs among tadpoles in severely hypoxic water probably wiii aiso be dependent upon such factors as degree of lung development and temperature. It is diflicult to draw general conclusions about the extent other of partitioning witho~tconsiderin~ variables. For exarnple, the developmentalstage has significant effects on partitioning related to differences in the relative development of lungi and giils and the large increases in tadpoli body size between hatching and the late premetamorphic stages (Burggren and Doyle 1986). The giii-lung-skin partitioning of VO, in buiifrog tadpoles was relatively independent of temperature from 15"-33C in spite of increases in metabolic rate (Burggren and Pinder 1983). The relative importance of air breathing wili depend upon species-specific differences in the efficacy of exchange mechanisms, and these vary greatly. For example, Osteopilus bmnneus, a tropical
"I
L 1 ,

liyiid, is an obiigate air breather in its natural environment and does not irrigate its buccai cavity (Lannoo et ai. 1987). Although Burggren and West (1982), Burggren and Pinder (1983), and Feder and Wassersug (1984) have shown that air breathing is infrequent and that pulmonary respiration is a minor component of total VO, for tadpoles in normoxic water. it is clear that an increased deoendence on air breathing wdl occur in hypoxic water. In order to minirnize predation risk, one would expect tadpoles to use one or more of a variety of adaptations, includmg selection of highPO, habitats, unpalatability, minimizing exposure time during breathing, acclimatization to hypoxic habitats (see below), and adaptive behavior patterns associated with air breathing. The selection of habitats with high PO, wouid reduce the dependence upon air breathing and wouid be adaptive if all other factors were equai. In habitats where the distribution of dissolved 0, is heterogeneous, PO, is normay highest in open (i.e., lacking dense vegetation) water (Ultsch 1973, 1976a; Wakeman and Ultsch 1976). Open water is not a good place for tadpoles to hide or find food, and when they do surface, they are much more likely to be detected by predators (Feder 1 9 8 3 ~ ) . Noland and Ultsch (1981) studied the relationships among temperature tolerance, hypoxia tolerance, and habitat seleaion in lungless tadpoles of Bufo tewestris and lunged tadpoles of Rana sphenocephaiu. Water temperature and dissolved 0, (DO) were measured from March to August at nine sites where one or both species occurred, and an attempt was made to samole all-microhabitats at each site. After these measurements, each microhabitat at each site was sampled to determine if either of the species was present. ~ h e ;found a trend toward a lowering of the criticalthermal maximum with increasing stage of development (aiso Bufo woodhousii, Shermaii 1980). Bufo tadpoles had a C , about 2C above that of Rana at all stages; higher acT, clunation temperatures resuited in higher values of C ,, T. Critical 0, tensions (Pc)of submerged larvae were relatively low at 20152 rnrn H " i d tended to increase with tempere ture but showed no clear relationship with stage or between species. Several conclusions can be drawn from the field data (fig. 8.6). (1) The range of temperature and D O was similar at sites containing Bufo and those with Rana (some sites had both species). (2) Bufo tended to live at higher temperatures (median of 30C) than Rana (median of 24"C), largely because of their occurrence in shallow water and ternporary ponds. (3) Eighty-two percent of Bufo larvae were found at a D O of more than 50% air saturation, and 92% were found above their Pc. (4) Oniy 37% of the Rana larvae were found in water with a DO more than 50% of air saturation, whde 38% were found below their P,. From these data, Noland and Ultsch concluded that Bufo breed in temporary ponds partly because such ponds tend to have both a high D O and hghtemperature reiative to permanent ponds. ~ h high D O e is obviously adaptive to a tadpole that lacks lungs, while both high D O and high temperature wiil increase the already relativeh hieh metabGlic rate of Bufo larvae. speed their de, " velopment, and increase the probability of metamorphosis before pond drying (see also Dupr and Petranka 1985; Pat-

( * ) a ,

GORDON R. U L T S C H , D A V I D F. B R A D F O R D , A N D J O S E P H F R E D A

50% SATURATION

C*

. ..
Om.

RANA SPHENOCEPHALA

DISSOLVED OXYGEN (ppm)

Fig. 8.6. Dissolved oxygen-temperature polygons for some aquatic habitats in Alabama. These areas represent the sum of possible combinations of dissolved oxygen and temperature within the habitats. The presence of tadpoles (Ramsphaocephalu and Bu@ tmertpis) at a location is indicated by dots and their absence by circles. The critical 0, tension (P,) for each species and 50% saturation level of dissolved oxygen are both indicated as a function of temperature. After Noland and Ultsch (1981).

terson and McIncktan 1989). Kana, on the other hand, breed in more permanent ponds that will tend to have lower DO levels, and the tadpoles compensate by breathing air during the lengthy develop~nental period. Nie et al. (1999) also studied the effects of ambient PO, and temperature on the distribution of tadpoles (Rana catesbeiana) in Alabama. They found that during the summer, the tadpoles in one of the study ponds were frequently found in water with a PO, below their P, in spite of the lack of predatory fishes and the fact that 56% of the pond bottom had water with a PO, greater than the PCof the tadpoles. Tadpoles caught in submerged traps in the hypoxic areas drowned, which indcates that air breathing was obligate in those microenvironments. If a tadpole lives in shallow water to ensure being in a hgh DO environment or breathes air in hypoxic water, it is exposing itself to predation. Predation is probably the major cause of mortahty of tadpoles, especially in permanent ponds, and may be a major selective force that has resulted in many anurans breeding in temporary (fish are absent) ponds (see chaps. 9 and 10). Moreover, predation may determine anuran community structure (Britson and Kissell 1996; Cortwright and Nelson 1990; Heyer et al. 1975; Woodward 1983; see Axeisson et a[. 1997) and can elirninate entire populations of tadpoles (Cortwright and Nelson

1990; Sexton and Phillips 1986; D.C. Smith 1983; see Barreto and Moriera 1996). Some anurans that breed in temporary ponds have relatively unpalatable larvae (e.g., R. A. Griffiths 1986; Reading 1990; Voris and Bacon 1966; Wassersug 1971; Wilbur 1987; some Bafa), which would allow them to be exposed in shallow, well-oxygenated waters, but in general temporary-pond breeders have tadpoles that are more palatable than those of permanent-pond breeders (see chap. 9). Among anurans that breed in permanent ponds, tadpoles of those with extended developmental periods may also be less palatable than those with shorter developmental periods. Bullfrog tadpoles, wMe apparently not absolutely unpalatable, will be avoided by some predators if there are other anuran larvae available ( h s e and Francis 1977; Woodward 1983). The relationshius between behavior and air breathing have not been exteisively studied. The decreasing frequenG of air breathing with increasing water PO, (Crowder et al. 1998, Wassersug and Seibert 1975; reviews by Feder 1984, Burggren 1984) intjlcates that pulmonary respiration is an accessory means of gas exchange in tadpoles. Using lungs only when necessary shodd lower the predation risk associIn ated with breathing at the surface (Feder 1983~). addtion, when a tadp& does surface t* breathe air, ;he breath is taken quickly and followed by a rapid descent to minimize

PHYSIOLOGY

201

exposure time. A few reports of synchronous air breathing among aggregating tadpoles are known (e.g., R W. McDiarmid, personal commuriication, Leptodaczy~us; Altig and Christensen 1981, Rana heckscheri; M. S. Foster and McDiarmid 1983, Stuart 1961, Rhinophvynus domali; N. D. Richrnond 1947, Scaphwpus holbrookii);this aaivity has been interpreted as an antipredator behavior among fishes (Kramer and Graham 1976). Tadpoles lacking functionai lungs are not necessarily unable to benefit from atmospheric oxygen. "Air gulping" is a cornrnon behavior in fishes in hypoxic water (Ultsch 1989) that can increase blood 0, transport (Burggren 1982). Even i severely hypoxic water the upper few millimcters will be n oxygenated, and if an aquatic animal can irrigate its gills with this water, it can use atmospheric 0, indirectly. S m d fishes with upturned mouths are weii adapted for this behavior (Kramer 1987; W. M. Lewis 1970).Bufi woodhousiz tadpoles appear to use the latter strategy, as they niove to the surface in hypoxic water but apparently do not gulp air (Wassersug and Seibert 1975).

Carbon Dwxide Eliminatwn andAcid-Base Balance The majority of CO, elirnination of d ampliibiails is cutancous, and tadpoles in normocarbic water would not be expected to be an exception (reviews by Burggren 1984, Burggren and Just 1992). This supposition was true for Taylorl Koiiros XVI-XIX (Gosner 4 0 4 1 ) bullfrog tadpoles in a flow-through respirometer, where 98% of the VCO, was aquatic (57% cutaneous, 41% branchiai) and only 2% was pulmonary even though 17% of the VO, I a s pulmonary (Burggrcn and West 1982). As with aquatic VO,, aquatic VCO, partitioning in resting tadpoles is not affeaed by developmental stage (and therefore body mass) throughout the premetamorphic stages. The dominant role of the skin in CO, elirnination that is characteristic of the adult is assumed when a larva cntcrs metamorphic clirnax at Taylor/Koiiros XX-XXV (Gosner 42-46; Burggren and West 1982). There is no reason to beiieve that the partitioning of VCO, between air and normoxic water would change sigd c a n t l y with an increase in pulmonary respiration. If the water becornes hypcrcarbic, a tadpole wouid have to increase its plasma PCO, in order to maintain an outward diffusion gradient for aquatic CO, elimination. Even a relatively siight environmental hypercapnia would be a potentidy severe acid-base chaenge. The plasma PCO, of tadpoles is normay very low aiid closer to that of fishes than to that of other amphibians (reviews by Heisler 1986; Toews and Boutilier 1986; Ultsch 1996; see Busk et ai. 1997). According to the Henderson-Hasselbalch relation, an increase in blood PCO, of 3 mm H g in a tadpole such as Rana catesbeiana represents about a doubling from normal values. This would cause much more of an effect on plasma pH than wodd be the case for aquatic salamanders with a normal plasma PCO, 2-6 times that of a tadpole (Ultsch 1987, 1996). Normal acid-base variablcs among tadpoles resemble those of fishes that iive i normoxic water (Heisler 1986; n Ultsch 1996; see Ultsch and Jackson 1996), aithough the

data are limited (Erasmus et ai. 1970-1971; HelE 1932; Tust et ai. 1973). Perturbations of acid-base status can be expected from the same sources as with fishes: exogenously from water aciddication and environmentai hypercarbia and endogenously froni accumulation of acidic metaboiites, especiay lactate. The reactions of tadpoles to acidic precipitation and acid mine drainaee are disc;ssed below. ~ h effects of water hve percarbia oGadpoles have received almost no attention. H;percarbia eiicits an increased dcpendence on pulmonary resiiration in aquatic salamandrs ( ~ a k e m & and ltsch 1976), as was found for the late stages of bullfrog tadpoles (Infantino 1989); increasing water PCO, up to 12 rnm H g had iittle effect on rates of giil or lung ventilation in bullfrog tadpoles in Taylor/Koiiros IV-XV (Gosner 2 9 4 0 ) . A ventilatory response had developed by Taylor/Kollros XVI-XIX (Gosner 4 0 4 1 ) , when the frequency of breathing doubled and the frequency of giii irrigation reduced by half; these changes would be consistent with a pulmonary hyperventilation that would partidy conip~nsate a respiratory acidofor sis and a reduction in aquatic VOZthat might slow the rate of CO, entrance. For tadooles of earlier stages with sma " lungs and for lungless species, this strategy is unavailable. Extraceiiular pH compensation would have to involve increases in pl&ma con~entratiorisof HCO;, a mechanism cornmon in fishes (reviews by Heisler 1986, Ultsch 1996) but lacking in the aquatic saiamandersAmphiuma and Siren (Heisler et al. 1982), which often inhabit hypercarbic urater. These salamanders d o w the extraceilular p H to f d during hypercapnia but defend intraceiiular p H by clevating intraceiiular concentrations of HCO;, partly by exchanging Clk for HCO; from the environment. With no field data on the upper iimits of PCO, in tadpole microenvironments, the leve1 of respiratory acidosis that might exist in natural populations is uiiknown. Tadpoles may lack effective mechanisms for coping with high CO, tensions and therefore avoid hypercarbic water. Thc accurnulation of lactate shouid present an acid-base chdenge.' and it is clear that laaate des accumulate both during exercise and severe hypoxia (see above). Howcver, as with most other sources of acidosis among tadpoles, there are no data on mechanisms of acid-base comoensation to the acidosis that accompanies lactate formation. In aquatic nutles, lactate accumulation results initiay in both metabolic and respiratory acidosis, as HCO; will be titrated at a greater rate than the animai can eliminate the resdtant CO, (review by Ultsch 1989). In tadpoles, because of their multiple and efficient modes of CO, elimination and theit relatively s m d body size (and therefore favorable surfaceto-volume ratio for cutaneous CO, elimination), a significant contribution of respiratory acidosis is not to be expeaed when laaate accumulates; t h s wodd ameliorate the p H disturbancc. In bullfrog tadpoles, lactate can be eliminated by the gds, s h , and kidney, aithough elirnination is not important relative to gluconeogenesis in processing lactate that results from exhaustive exercise (Quinn and Burggren 1983). These paths of elimination might be irnportant during chroiiic laaic acidosis induced by long periods of anV

202

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

aerobiosis such as might occur during ovemintering in hypoxic water.

Thermal Physiology
Of all the physicai parameters in the aquatic environment, temperature is perhaps the most dramatic in its effea on the physiology, ecology, and behavior of anuran larvae. For example, environmental temperature dramatically affeas the time taken to reach metamorphosis, which can be criticai t o the survivai of an individual faced with a drying habitat or the onset of winter. Temperature aiso affeas differentiation and growth rates, body size at metamorphosis, mechanisms of gas exchange, rates of energy metabolism, and undoubtedly many other physiologicai parameters documented in other eaothermic vertebrates. Moreover, the limits of temperature tolerance and temperature-dependent life history traits of anuran l m a e are generally related to the geographic distribution of the species. The thermai environment of all anuran larvae is variable in space and time, often extremely so in shallow water exposed t o intense insolation. For example, tadpoles of Rana catesbeiana may encounter seasonai temperatures ranging from near 0" t o over 3SC, and near-shore temperatures may increase by more ti~an20C in less than 6 hr (Ldlywhite 1970; Menke and Claussen 1982). Water temperature adjacent t o tadpoles of some species i temporary ponds may n regularly rise during the day to 42C (H. A. Brown 1969; Dunson 1977). Anuran larvae have generally accornmodated to variation in the thermai environment by (1) possessing the physiologicai ability to function over a wide range in body temperature, (2) behaviorally regulating body temperature, and (3) physiologically adjusting to changes in environmentai temperature (i.e., acclimatization, or in the laboratory, acclimation). O n an evolutionary scale, variation in the thermai environment has resulted in natural selection for differences among anuran larvae in a number of temperature-dependent physiological or life history traits. Temperature Tolerante Lzmits The lirnits of temperature tolerance of an animal are commonly described by a criticai thermai maximum (R,) and criticai thermal rninimum (CT-). These usually are defined as the temperatures at which "locomotor activity becomes disorganized and the animal loses its ability to escape from conditions that wili promptly lead to its death" (Cowles and Bogert 1944). The C , for most anuran larvae, when comT, pared at the same stage of development and at intermediate acclimation temperatures, is between about 38" and 42C (table 8.2). Extreme vaiues are severai degrees higher (44.9"C, Bufo marinus; fig. 8.7; table 8.2), and C , is apT, parently often approached in nature. For example, Cyclorana cultrpes, C. piutyuphaiu, and Litoria mbella occur in Australian pools at 39.2"C (Main 1968), Lzmnonectes canwivovus at 42C (Dunson 1977), Scaphwpw couchii at 39"40C (Mayhew 1965), Pseudaois tmimata at 36.SC (Hoppe 1978), Osteopzlus septentrionalis suspeaed at 42C (H. A.

Acclimation temperature O 35C A 30C O 25C

DEVELOPMENTAL STAGE
Fig. 8.7. Critical thermal m b a () a , and critical thermal minima (arn) of larvae of Bufi marinus at three acclimation temperatures and variou stages of development. Dotted liries connect mean responses of larvae averaged over aii acciimation temperatures. Staging foiiows Gosner (1960). After Floyd (1983).

Brown 1969), and a number of Australian species at 35"45C (Tyler 1989). CT,, of tadpoles appears to be related to the natural environmentai temDeratures to which the animais are ex~osed. Such a pattern is weli established for anuran embryos (Bachmann 1969; J. A. Moore 1949b). The highest reported C , vaiues for tadpoles occur in species tht typicay inT, habit shailow ponds exposed to intense insolation (e.g., Bufo marinus and other species, Gastrophryne caroliansis, and Limnonectes cancrivoms): Among tw species of hylids and two species of pelobatids (table 8.2), lethal temperature is directly correlated with the maximum temperatures likely t o be experienced in the field (H. A. Brown 1969). In Ascaphus &i,-a species inhabiting coo1 streams, C , may be excepT, tionally low, as tadpoles aiways avoid temperatures above 22C in a thermai gradient (De Vlaming and Bury 1970). Pseudmi tmseriatain Colorado has a stgnificant& higher C , at low elevation (e.g., 38.6"C at 1530 m, Taylor/ T, Koiiros XX [Gosner 421) than at high elevation (3725C at 2710 m; Hoppe 1978). Likewise, in R. sylvatica heat tolerante decreases with both increasing elevation and latitude

PHYSIOLOGY

203

Table 8.2 Preferred body temperatures (PBT, as mean selected temperature) and C , and CT,,for anuran l m a e at various stages and accliT, mation temperatures. In studies including more than one acclimation temperature or larval stage, an intermediate acclunation temperature and I T, ,o b at stage closest to Gosner (1960) stage 39 were chosen. Values U parentheses are the extreme means of C , or a , s e ~ e d any larval stage or acclimation temperature. FTP = final thermai preferendum, (i.e., acclimation temperature equals PBT; Reynolds and Casterlin 1979) and RT = room temperature. Acclimation Temperature ("c) 10 FTP 22.5 20 22.5
-

Species ASCAPHIDAE Ascaphus tnrei BUFONIDAE Bufi amwicanus

Stage

PBT

a, ,

%
De Vlaming and Bury 1970 Dupr and Petranka 1985 Beiswenger 1978 Cupp 1980 Beiswenger 1978 Karlstrom 1962 Karlstrom 1962 Straw 1958 Beiswenger 1978 Floyd 1984 Floyd 1983 Heanvole et ai. 1968 Davenport and Castle 1895 Noland and Ultsch 1981 Sherman 1980 Cupp 1980 H . A. Brown 1969 H. A. Brown 1969 Dupr and Petranka 1985 Hoppe 1978 Cupp 1980 Heanvole et al. 1968 Cupp 1980 H. A. Brown 1969 Gosner and Black 1955 H. A. Brown 1969 Dunson 1977 W o h u t h et al. 1987 W o h u t h and Crawshaw 1988 Dupr et ai. 1986 Hutchison and Hi 1978 Lucas and Reynolds 1967 Menke and Claussen 1982 Workman and Fisher 1941 Wiilhite and Cupp 1982 D. F. Bradford, unpublished data W o r h a n and Fisher 1941 Casterlin and Reynolds 1979 Lucas and Reynolds 1967 Noland and Ultsch 1981 Dupr and Petranka 1985 W o r h a n and Fisher 1941 Cupp 1980 Manis and Claussen 1986

B. boreas

22.5 27 30 21 25 27
-

20 HYLIDAE Osteupilus septentrionalis Pseudaclzr regilia f?hiseriata LEPTODACTYLIDAE Lepto*lus albilabvk MICROHYLIDAE Gartruphvyne carolinensis PELOBATIDAE Scaphiopus couchii S. holbmokii Spea hammondii RANIDAE Limnanectes ca~tcrivwus R. cmcadae R. cutesbeiana 25 20 ETP
-

20 21 20 25 18-20 26.5 22-25 21 21 24 26.7 RT 15 20 20 4 20 FTP RT 27 FTP 20 20 22 temperature that 50% of the tadpoles can tolerate indelinitely).

.'Incipient lethal temperature (i.e., m&um bLethaltemperature, dead in 1-2 hr.

204

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

(Manis and Claussen 1986). In contrast, no sigiiificant difference in C , was found benveen populations ofleptodmT = tylus albilabris at 15 and 570 m elevation in Puerto Rico (Heanvole et ai. 1968). The C , also appears to correlate with environmentai T, temperatures, although it has been determined for tadpoles of only a few species (table 8.2). In Rana muscosa, a species in which tadpoles ovenvinter in frozen lakes, C , is below T, -2C. The C, T , is near or below 0C in R. catesbeiana, a temperate species that ovenvinters in some populations (Coiiins 1979), a n d h a p h w truei, a species inhabiting coo1 temperate streams. In contrast, the tropical Bufo marinus has a much higher Ct,, (i.e., 6.0C or greater and dependent on acclimation temperature and larvai stage; fig. 8.7). Similarly, in the temperate species Scaphiopus holbrookii, which undergoes larvai development entirely during the surnmer, C , is 8.5"C in larvae acciimated at 18"-20C. T The temperature tolerance range of anuran larvae is large in comparison to most eaothermic vertebrates. In species T, in which both CT,, and C , have been determined, the temperature tolerance range for a single stage of development and acclimation regime is 34C in Bufo marinus, 37C in R. catesbeiana, and 28C in Scaphwpus holbrookii (table 8.2). This range is even greater if acchation effects are included. For example, B. marinus tadpoles at Taylor/Kollros VI (Gosner 31) can tolerate a temperature range of 39C if acchated prior to exposures to the low and high temperatures (fig. 8.7; Floyd 1983; ais0 Abe and Godinho 1991). Data are not available for the temperature tolerance limits of tropical forest species that experiente relatively s m d variations in temperature. It might be expected that tadpoles of these species would have narrower temperature tolerance ranges than species exposed to wider temperature variation. However, among adult anurans, tropical species do not gene r d y have narrower temperature tolerance ranges in comparison to temperate forms, aithough their temperature tolerance iimits are higher than those for temperate species (Brattstrom 1968). The typicaiiy higher CT,, of tadpoles relative to adults is a striking difference in the thermai relations of larvai and adult anurans. This is the case for Bufo americanus, B. exsul, B. marinus, B. woodhousii, Gastrophlyne carolinensis, Rana catesbeiana, R. pipiens, R. sylvatica, and Scaphiopus holbrookii (Cupp 1980; Floyd 1983; Hathaway 1927; Sherman 1980; Straw 1958; compare data in table 8.2 with Brattstrom 1968). The only apparent exceptions to this pattern are Pseuh r i s tnjeriata, in which larvai m appears to be similar , to adult CT,, (Hoppe 1978), and B. canums, in whch larvai C T ,, is lower than adult C, T , (Karlstrom 1962). The finding that R. pipiens tadpoles die more quickly at high temperature than adults (Orr 1955) may have been caused by differences in body size. The generdy higher a , , tad, ,of poles than adults may reflea the need for tadpoles to minimize the time to complete metamorphosis and escape inimicai environmentai conditions and aquatic predators (see below). Alternatively, this difference may reflect the greater iikelihood for tadpoles than adults to become trapped in warm surroundings. Adults on land continudy lose heat

by evaporation (Tracy 1976) and often have the option to move to aiternative thermai microenvironments (Lillywhite 1970). Both larvai and adult anurans regulate body temperature behaviordy by moving or changing posture among available thermal microenvironments (Brattstrom 1962, 1970). When placed in a thermal gradient in the laboratory, anuran larvae invariably selea a certain temperature or temperature range more frequently than others. The arithmetic mean of these selected temperatures is often defined as the preferred body temperature (PBT; Reynolds and Casterh 1979). Many biochemicai and physiologicai processes, such as active metabolic rate, maximurn sustained speed, growth rate, food conversion efficiency, and learning and memory capabilities, are optimai at or near the PBT (Huey and Stevenson 1979; Hutchison and Maness 1979). Among anuran larvae, important processes known to change as a function of temperature include dzerentiation rate, growth rate, body size at metamorphosis, mechanisms of gas exchange, and metabolic rate (see below). Clearly, temperature regulation aiiows maximization or control of diierentiation and growth rates (Smith-Giii and Berven 1979). For example, in Pseudacris ornata, Rana clamitans, and R. sylvatica, differentiation and growth rates increase with increasing temperature until an inhibitory temperature weli below CT,, is reached (fig. 8.8;

i1

Growih

TEMPERATURE ('C)
Fig. 8.8. Temperature dependence of differentiation rate and growth rate for individual Ranapipiens. Growth rate is cornputed as the regression coefficient of body v o l ~ r n eon .days up to Taylor/Koiiros XVII ~ ~~ (Gosner 40). Differentiation rate is computed as the regrcssion coefficient of stage on days through Taylor/Kollros XX (Gosner 42). Lines connect geometric rneans at each temperature. After Smith-Giii and Berven (1979).

PHYSIOLOGY

205

uble 8.2; Berven 1982; Berven et ai. 1979; Harkey and Scmlitsch 1988; Smith-Gili and Berven 1979). Temperature ulection rnay be critical to a tadpole faced with a deteriorati aquatic environment, the end of the growth season, or q ni increased risk of predation. Behaviorai temperature reguhxion also rnay serve to rninimize exposure to deleterious mn-tures abd conditions. For ex&vle. PseudanZs ornata lan-;e have fewer deformities and greater average survivai a h e n raised at intermediate temperatures of 20" and 25C tban when raised at 10" and 30C (Harkev and Semlitsch 1988). In overwintering species, avoidance of the lowest mperatures rnay minimize the probability of entrapment in ice that rnay lead to death by freezing or anoxia (Bradford 1984b). A temporary elevation in PBT, termed behaviorai fever, occurs in d major groups of ectothermic vertebrates iNuger 1978). Behavioral fever can be induced in tadpoles by injection of kied Aeromonm hydrophih, a common pathopin fishes, ampiubians, reptiles, and mamrnals (Casterlin and Reynolds 1977). This response has an adaptive vaiue in some ectotherms by increasing survivai rate after infection a i & iive A. hydrophila (Kluger 1978). However, behaviorai kver in tadpoles rnay aiter their behavior and increase their nisceptibility to predators (Lefcort and Eiger 1993). The mechanisms of heat exchange involved in behaviorai thermoregulation M e r between l&al and adult anurans because only adults can thermoregulate on land where radiation and evaporation are important mechanisms of heat exchange (Liilywhite 1970; Tracy 1976). In water, conduction and convection predominate as heat exchange mechanisms. Consequently, the body temperature of a tadpole is determined alrnost entirely by the temperature of the surrounding nater. Neither tadpoles nor frogs are known to possess any physiologicai means of controhng heat flux when in water (Brattstrom 1979). Nevertheless, solar radiation rnay contribute direaly to the heat input of tadpoles in shallow water (see below). The PBT of most anuran larvae studied is between 28" and 32C (table 8.2). These values are roughly 10C below the CT,,for each species, aithough PBT does not appear to be well correlated with CT,, There is some evidence for a weak correlation between natural environmental temperature and PBT. For example, in Rana clamitans, R.pipiens, and R. sylvatiea in the northeastern United States, tadpole PBT is duealy correlated with the beginning date of the breeding season and the northernmost distribution of the species (J. A. Moore 1949a; Workman and Fisher 1941). Moreover, the two species with the lowest known PBT (table 8.2; Rrcaphus mez and R . sylvatua) inhabit the coldest environments (also see Miranda et ai. 1991). The precision of temperature selection varies considerably among species and larvai stages. For example, Pseudacris tris&ata and Rana sphevweqphala behave as weak thermai selectors, especidy early in larvai development, whereas Bu@ am'canus behaves as a strong thermai selector throughout development (Dupr and Petranka 1985). For the developmental stages showing the most precise temperature selection in these species, the standard deviation for seleaed tem1

perature in groups of tadpoles is 2.2"C in E triseriata, 2.1C in R. sphevwcephala, and 1.6"C in B. amevicanus. The corresponding value in R. eatesbeiana, in which selected temperatures were determined for individuais rather than groups, is 0.7"C ( W o h u t h and Crawshaw 1988). In contrast to CT-, the PBT of tadpoles shows no general pattern in comparison to PBT for adults. Rana cmcaue tadpoles have higher PBTs than adults ( W o h u t h et ai. 1987), R . pipiens tadpoles appear to have a similar PBT to adults (Casterlin and Reynolds 1979), R . catesbeiana have similar or lower PBTs than adults (Hutchison and Hdl 1978; Lillywhite 1970; Lucas and Reynolds 1967; Wollmuth and ~rawshaw 1988), andRreaphhs m e i tadpoles have lower PBTs than adults (Claussen 1973; De Vlaming and Bury 1970). Data are not avadable for tropical forest anurans tha; experi'ence a narrow temperature range or iive in therm d y homogeneous habitats. It is possible that some species in these conditions do not behaviordy regulate body temperature like some tropical forest-dwelkg fizards ( ~ u and ~ e Webster 1976).

Ontogeny of Temperature Tobrance and Reguhtion The temperature tolerance limits of anuran larvae typically change considerably during ontogeny. The most striking and consistent change is a decrease in CT,, near the completion of metamorphosis (about Taylor/Koliros XXI-XXV [Gosner 43-46]; fig. 8.7), as in Bu@ amerieanus, B. marinus, B. terrestris, B. woodhousii, Gashopblynecarolinensis, Pseudacris triseriata, Rana sphevwctphah, and R . sylvatua (Cupp 1980; Floyd 1983; Hoppe 1978; Noland and Ultsch 1981; Sherman 1980). The only exception is R. catesbeiana, in which this pattern was evident at only one of two acclimation temperatures (Menke and Claussen 1982). Prior to metamorphosis, CT,, remains fairly stable over a broad range of stages in most species, although significant changes are sometimes apparent (above referentes; Gosner and Black 1955). Ontogenetic changes in the CT,has received iittie attention. In B u . marinus the change in CTm, with stage is pronounced and is largely the opposite of the change in CT(fig. 8.7). In determinations of CTmin three groups of for stages of R. catesbeiana, CTmi,was near 0C and varied l i d e among the stages (Menke and Claussen 1982; see Dainton 1988, 1991). Based on these limited data, the temperature tolerance range of anurans (i.e., C , to CT-,) appears to be reduced T, during the final stages of metamorphosis, rather than shified t o a lower level, as postulated by Cupp (1980) and Floyd (1983). The characteristic reduction in CT,,,= the apparand ent reduction in temperature tolerance range at the end of metamorphosis rnay be consequences of biochemical and morphologicai rearrangements occurring at this time (Sherman 1980). Floyd (1983) noted that the times of least tolerante to temperature extremes in B. marinus coincide with the times of greatest morphological and biochemicai rearrangement early and late in larval development. The thermai preference of tadpoles also changes during larval development. Throughout most of development, PBT generdy increases with larvai stage (e.g., De Vlaming and

206

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

Bury 1970; Dupr et al. 1986; Dupr and Petranka 1985; Floyd 1984; Hutchison and Hdi 1978; Lucas and Reynolds 1967; Woiimuth and Crawshaw 1988; Wollmuth et al. 1987; Ascaphus truei, Bufo americanus, B. marinus, Pseudacris triseriata, Rana cmcaae, R. catesbeiana, R. pipiens, and R. sphenocephala). However, in R. cascaae and R . catesbeiana (Oregon), PBT declines substantiaily during final metamorphic stages (Woilmuth and Crawshaw 1988; Wollmuth et al. 1987), whereas it appears to remain high at this time in B. am.canus, l?tkeriata, R. catesbeiana (Oklahoma), and R. sphenocephala (Dupr et ai. 1986; Dupr and Petranka 1985; Hutchison and Hill 1978). Precision of temperature selection generaily increases throughout larval development, along with the increase in PBT (e.g., Bufo americanus, Pseudacrii triseriata, Rana catesbeiana [Kentucky], and R . sphenocephala; Dupr et ai. 1986, Dupr and Petranka 1985). Standard deviations of selected temperature in groups of tadpoles of these species range from about 4"-8C in early developmental stages to about 2"-3C in later stages. In R . catesbeiana, in which individual tadpoles were monitored rather than groups, PBT and the precision of temperature regulation also generaily increased during development (Wollmuth and Crawshaw 1988). However, both precision and PBT decreased at the end of development (i.e., Taylor/Koilros XXII-XXV [Gosner 44461). The standard deviation for selected temperature in this study varied among stages fiom 0.7O4.6"C. The adaptive significance of ontogenetic changes in PBT is not clear. Floyd (1984) argued that ontogenetic changes in PBT of B. marinus are correlated with changing thermal characteristics of the natural microhabitats, and PBT is not coupled to ontogenetic change in temperature tolerance. Dupr and Petranka (1985) argued that the general increase in PRT during development rnay reflea greater selection pressure for rapid development at the end of the developmental period; this is a time when larvae become increasingly susceptible to habitat deterioration (Wilbur 1980) and predation (S. J. Arnold and Wassersug 1978; Wassersug and Sperry 1977), and the growth season may be ending. Finay, ontogenetic changes in PBT may spatiaiiy segregate age or size classes and thereby potentiaily reduce competition or cannibahsm, or encourage the formation of size-specific, cooperative aggregations (Alford and Crump 1982; Dupr and Petranka 1985). Although age segregation ofR. mcadae occurred in the field, groups were in such close proximity that neither cannibahsm nor competition were likely to have been affected ( W o h u t h et ai. 1987).

son and Maness 1979; and references therein). In ali amphibian larvae examined (table 8.2), temperature tolerance limits are significantly affected by the temperature of acciimation. For example, an increase in acciimation temperature of Bufo marinus from 25" to 35C resiilted in an increase in CT,= of about 2.5"C and an increase in C , of about 1C T (fig. 8.7). Among five other species (Bufo,Gastrophryne,Pseuda&, Rana), an increase in acclimation temperature from 10" to 30C resulted in an increase in CTm, of about 1"-3C (Cupp 1980), and an increase in acclinlation temperature from 15" to 35C in three species (Osteopilus,Scaphiopus, and Spea) resiilted in an increase in lethal temperature by about 4C (H. A. Rrown 1969). Such effects are probably essential for surviving temperature extremes in natural conditions, as a number of species have been observed in the field thriving T , (see above). The in water at temperatures near their C, effect of acclimation temperature on PBT is much less distinctive than its effea on temperature tolerance iirnits. InAscaphus truei and B u . marinus, there is no effect of temperature acclimation (De Vlaming and Bury 1970; Floyd 1984), whereas a complex interaction was found among acclimation temperature, stage of development, and PBT in Rana cascadae,R. catesbeiana, and R.pipiens (Hutchison and Hdi 1978; Lucas and Reynolds 1967; Wollmuth et al. 1987). The time required to acclinlate completely to a new temperature has not been documented for anuran larvae. In adult anurans about two to four days are required for C T ,, and CT,, to s t a b h e after a large change in acclimation temperature (Brattstrom 1968; D u e h a n and Tmeb 1986). Possibly as little as 24-36 hr are required in larvae, and t h s time has been used as the minimum required for measuring the "final thermal preferendum" (= the temperature uitimately selected regardless of previous thermal experiente; Casterlin and Reynolds 1979; Dupr and Petranka 1985; Reynolds and Casterlin 1979). In addition to acclinlation, which is a gradual process, tadpoles of Rana catesbeiana undergo heat and cold "hardening" (Menke and Claussen 1982). In this process tadpoles exposed briefly to near-lethal temperatures become more tolerant to subsequent exposure to these temperatures within 90 rnin.

Other Infuential Factors A number of faaors (e.g., time of day, season, iight intensity, photoperiod, and nutritional state) other than stage of development and thermal history (i.e., acclinlation) influente theri%oregulatory behavior or-temperature tolerance in anuran larvae ( D u e h a n and Tmeb 1986; Hutchison and MaTemperatureAcclimatwn ness 1979). The magnitude of the effeas of these parameters Over a period of hours to days in a laboratory setting, tad- on PBT and tempeiature tolerance lirnits gener&y is smail poles accommodate to changes in the thermal environment in comparison to the effects of thermal history. by physiologicaiiy adjusting to new temperatures (= temDiel and seasonal cycles i temperature tolerance and n perature acclirnation; Hutchison and Maness 1979). When PBT have been found in manv eaothermic vertebrates. alsuch adjusunents occur in the natural environment, the pro- though it is not established whether ail cycles are t d y encess is cailed acclirnatization (K. Schmidt-Nielsen 1983). dogenous (Hutchison and Maness 1979). Among anuran' Temperature acclimation has been discussed extensively for larvae a diel Dattern in PBT was observed in Rana clamitans adult amphibians and for ectotherms in general (e.g., Dueli- and R. pipiek but not in Ascaphus truei (Casterlin and Reynman and Tmeb 1986; Hutchison and Dupr 1992; Hutchi- olds 1979; De Vlaming and Bury 1970; Wiiihite and Cupp

PHYSIOLOGY

207

1982). Seasonal differences in PBT have been found in R. during the daytime as water temperature increases. Tadpoles catesbeiana and R. pipiens (Lucas and Reynolds 1967; Woil- commonly select the warmest temperatures available (e.g., muth and Crawshaw 1988). In tadpoles of R. catesbeiana at Beiswenger 1977,1978; Bradford 1984b; Brattstrom 1962; Taylor/Kollros XI-XVI (Gosner 36-40), PBT was above Muilally 1953; Noland and Ultsch 1981; Tevis 1966; Woll30C during surnmer and in the mid-20s during winter. In muth et al. 1987; Bu) boreas, B. canorus, B. hemwphlys, B. contrast, earlier stage tadpoles selected the same temperature punctatus, B. terresCm,,Rana c a s h , and R. muscosa). Overthroughout the year. Aithough the cues underlying this sea- wintering R. muscosa select the warmest temperature below sonal diierence in later stage tadpoles of R. catesbeiana are the thermocline in frozen lakes (i.e., 3"4"C; Bradford unknown, a seasonal dzerence in PBT would unction to 1984b). For some anurans (e.g., Beiswenger 1977, 1978; minirnize rate of diierentiation at a time inappropriate for Brattstrom 1962; Noland and Ultsch 1981,Bufo a m a n u s , rapid development and metamorphosis (Woilmuth and PseudamS crucfer, Rana boylii, and R. sphevwcephala) surface waters become sufficiently warm that they are avoided or no Crawshaw 1988). Although tadpoles of many species are positively or neg- longer sought. A comparison of the thermal relations of Bufo terremi atively phototaxic (Dueliman and Tmeb 1986; Wolimuth et ai. 1987), daytime PBT does not diier significantly from and Rana sphenocephala in the field and laboratory demonnighttime PBT of Rana pipiens (Casterh and Reynolds strates relationships among tadpole temperature tolerance 1979). Many positive phototaxic responses may occur in an- limits, tadpole thermoregulatory behavior, and the therticipation of increasing water temperatures that foilow expo- mal characteristics of the environment (Noland and Ultsch sure to sunlight (Beiswenger 1977; Dueliman and Trueb 1981). Tadpoles of B. tememi typicdy select the warmest 1986). Photoperiod d e a s the PBT of second-year but not temperatures available in shaiiow water, which often results PBT increases when in body temperatures above 30C. In contrast, R. pipiens first-year tadpoles of Ascaphus -i; photoperiod increases (De Vlaming and Bury 1970). Photo- avoids shaiiow water when temperatures become warm, and period affects C , and CT,,, in B. marinus tadpoles in a body temperatures are typicaiiy weil below 30C. These T, way that appears to increase survival under potentially dan- dzerences correlate with a higher CT,, in B. temescris than gerous thermal conditions (Floyd 1985). At high acclima- in R. pipiens (table 8.2). Aiso, the body temperature of B. tion temperatures, a long photoperiod increases R,,and temesh-zj in the field more frequently is near the CT,, than at low acclimation temperature, a short photoperiod reduces R. pzpiens The conspicuous behavior that maximizes body CTmm. magnitude of these effects in B. marinus is a s m d temperature in B. temesh-zj, a species that frequently breeds The fraction of the changes in C , and CTm,,caused by ther- in shdow, temporary waters, is viewed as a strategy for T, mal history. shortening developmental time and thus increasing the Nutritional state affects PBT in many ectothermic verte- probabiiity of completing metamorphosis before the habitat brates, includiig aduit anurans. The increase in PBT that oc- dries. In contrast, R.pz;piensusudy inhabits more permanent curs when the animal is digesting food (Ldiywhite et al. habitats than Bufo and does not develop as rapidly. More1973) presumably maximizes rate of digestion and con- over, Rana habitats typically have lower ambient temperaserves energy when digestion is not occurring. This pattern tures than Bu$ habitats. may be irrelevant for many anuran larvae, which are virtudy Dense aggregations of a few to thousands of anuran larnever without food in the digestive tract. No data are avail- vae frequently occur (see chap. 9). Aithough a number of able to assess the effect of nuuitional state on PBT in tad- factors are involved in the formation and maintenance of poles. Nutritional state affects thermal tolerance limits in these aggregations, water temperature is often the dominant Bufo americanus and B. marinus by decreasing the range of factor (Beiswenger 1975, 1977,1978). For example, in Bufo tolerance during food deprivation (Cupp 1980; Floyd ametiuznus, dispersed tadpoles are activated in the morning 1985). However, the magnitude of the reduction in both by increasing light intensity and move to shdow water on species is s m d in comparison to the effects of thermal his- sunny days where they accumuiate in warm microhabitats tory on temperature tolerance h i t s . (Beiswenger 1975). Once tadpoles reach sufficient density in the lighter and warmer microhabitats, feeding and social Themzoreg~tlatwn n d h r e g a t w n Behavwr a activities draw them into dense aggregations. At night as Aithough selection by tadpoles of a preferred body tempera- water temperatures decrease in the shdows, tadpoles return ture is evident in the laboratory, a host of factors (e.g., light to warmer, offshore water. Beiswenger (1975) argued that intensity, food availabiiity and quality, dissolved oxygen, aggregating is a way for tadpoles-to exploit food more popuiation density, presence of predators, and tadpole social efficiently, engage in certain social activities, and seek suitinteractions) other than temperature may d e c t microhabi- able temperatures more effectively. In contrast, Brattstrom tat selection in the field (Beiswenger 1975, 1977; Lemckert (1962,1970) and Woilmuth et al. (1987) argued that aggre1996; Noland and Ultsch 1981; Woilmuth et al. 1987). gations in B. boreas, Pseudacris crttcfer,l? regila, &na boylzi, Nevertheless, temperature is often the dominant factor de- and R. c a s h occurred because of, and to augment, the termining microhabitat seleaion, and thermoregulatory be- thermal characteristics of the local microhabitat. Clearlv. havior has been documented in a number of studies. dense aggregations in many species are often observed in the Tadpoles in many temperate ponds typicatiy lie offshore warmest niicrohabitats in a pond. at night in warmer, deeper water and migrate to shdows Brattstrom (1962) provided experimental evidence that

i'

208

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

tadpoles aggregate, in part, because they absorb more solar crease from 13" to 23C resuits in about a 10-fold increase radiation than the surroundmg water, which resuits in more in differentiation rate and a 6-fold increase in growth rate in heat being absorbed in water containing aggregations. Beis- Ranapipiens (fig. 8.8; Smith-Giii and Berven 1979). Similar wenger (1977) described observations both supporting and patterns are also found in R. chmitans and R. sylvatica (Berrefuting this argument. The use of white porcelain pans in ven 1982; Berven et ai. 1979). which much of the radiation rnay have been refleaed by the Environmentai temperature has a ~ronounced effect on white substrate is a limitation in evaluating the data in Bratt- length of the larvai ieriod by virtu; of its iduence on strom's (1962) paper. Further experiments are needed in differentiation and growth rates (Smith-Gill and Berven which the radiative properties of the substrate in the test sys- 1979). In some species, reducing the time to metamorphosis to a rninimum is critical because tadpoles must complete tem better represent those in the field. metamorphosis prior to drying of the habitat or onset of Temperatwe, Dzflerentiatwn, and Crowth Rate other inimical conditions. Behavioral temDerature selection The influence of temperature on differentiation and growth resuiting in maximal differentiation rate wouid be expected rates probably is the most dramatic effect of environmentai (e.g., Brattstrom 1962; Sherman 1980; Wolimuth et al. temperature on the ecological physiology of anuran larvae. 1987). The differential effea of temperature on differentiaThis influence determines larval body size and length of the tion and growth rates promotes emergence from ponds durlarval period (Reques and Tejedo 1995; Smith-Giii and Ber- ing the warm part of the year, particularly if ponds become ven 1979), which in turn are directly related to survivai and overheated during pond drying (Etkin 1964). The same reproductive success (Berven 1982; Berven and Gdl 1983; mechanism mhibits emergence during the inappropriate Berven et al. 1979; C o h s 1979; Crump 1989a; Harkey cold season. In some species and environments, temperature may be and Semiitsch 1988; D.C. Smith 1987). Typicaiiy, an increase in temperature resuits in faster rates of development sufficiently low to make ovenvinteriig of larvae obiigatory and growth, shorter time to metamorphosis, and smaiier (Berven et al. 1979; C o h s 1979; Coiiins and Lewis 1979; Smith-Giii and Berven 1979). Overwintering occurs in a body size at metamorphosis. A model that predicts the timing of metamorphosis based number of anurans at high elevation and latitude where averon differentiation rate has been proposed and tested in the age temperatures are lower and the growing season is shorter laboratory and field by Smith-Gdl and Berven (1979). In than at lower elevation and latitude. For examde. Rana Ranapipiens, differentiation rate (defined as the slope of the catesbeiana larvae in Louisiana rnay complete metamorphosis stage-age relationship) is strongly correlated (r2 = 0.95) in a few months, whereas individuals in Michigan may not with the length of the larval period, whereas growth rate is metamorphose untii their third summer (Coiiins 1979).Asconsiderably less correlated (rZ = 0.51). Growth rates of caphus trei tadpoles in subalpine sites in Washington require R. pipiens and R. sylvatica larvae in some circumstances show five summers to complete larval development (H. A. Brown no significant relationship to length of the larval period. 1990b). In R. ylvatica in montane environments, a oneGrowth pararneters are often poor predictors of timing of month difference in ovi~ositionresuits in nearlv a vear , , metamorphosis largely because growth and differentiation difference in completion of metamorphosis because larvae are affected differentiaily by environmental factors such as from eggs oviposited later must overwinter (Berven et ai. temperature and population density. Also, growth rate is 1979). At extreme elevations and latitudes, low temperamav be the factor h:tine functionaiiy dependent upon differentiation rate, whereas tures that Drevent metamor~hosis the converse is not true. This is because growth and develop- the geographic range of a species (Smith-Giil and Berven mental rates are both functions of circulating hormone lev- 1979; Uhlenhuth 1921). els, which are in turn dependent upon thestage of differAmphibian larvae are typicaiiy larger when raised at low entiation reached. Nevertheless, growth rate frequently is temperature, a phenomenon that appears to be general for related to the duration of larval development (e.g., Wiibur larval ectotherms (Berven 1982; Berven and GU 1983; Berand Coilins 1973) because conditions favorable for differ- ven et al. 1979; C o h s 1979; Etkin 1964; Harkey and Sementiation often are also favorable for growth (Smith-Giil and iitsch 1988; Koiiros 1961; Smith-Gdi and Berven 1979; Uhlenhuth 1919). For example, R. pipiens raised at 13" and Berven 1979). Temperature is a major factor determining larvai differ- 23C in the laboratory differ in body size by about 3-fold at entiation and growth patterns, although other factors such aii stages, including metamorphic climax (fig. 8.9; Smithas density, food abundance and quality, and water quality Gdi and Berven 1979). In R. sylvatica in the laboratory and are clearly important (Berven 1982; Berven and Giil 1983; in R. clamitans in both the field and laboratory, a 10C deHarkey and Semiitsch 1988; Pandian and Marian 1985b; crease in temperature results in about a 2-fold increase in Smith-Giii and Berven 1979; Viparina and Just 1975; Wil- body size at ietamorphic clima (Berven 1982; Berven and bur 1980; Wdbur and Coilins 1973). As temperature in- Gil1 1983; Berven et al. 1979). Larger size at lower temperacreases, rates of differentiation and growth both increase ture can be attributed to the greater effect of temperature on until an mhibiting temperature is reached (Berven 1982; Be- differentiation rate than on growth rate (fig. 8.8; Berven rven et al. 1979; Harkey and Semhtsch 1988; Smith-Giii et ai. 1979; Smith-Giii and Berven 1979). Thus. iarvae at and Berven 1979). Such an increase is predictable based on low temperatures gain more mass and are larger at any given Arrhenius enzyrne lunetics. For example, a temperature in- stage, including metamorphic clima, than at higher temI
L ,
V

PHYSIOLOGY

209

fects of temperature, rather than genetic or maternal effeas. Nevertheless, significant genetic differences were found. For example, in R. clamitans genetic differences were found in the tmperature at whch-dfferentiation and growth rates were maximal, in temperature sensitivity of differentiation and growth rates, and in stage-specific growth rates. These genetic dfferences were counter to the observed clinai Gariation. For example, in cold-temperature experiments designed to simulate mountain-top conditions, montane tadpoles developed faster and completed metamorphosis earlier and at a smder size than did lowland tad~oles. Thus. naturai seleaion in R. chmitans has favored the shortest possible larvai periods given the ambient temperature constraints. This pattern is viewed as a case of counter-gradient selection, in Ghich naturai seleaion favors genot$es that minimize the range of phenotypes potentidy induced by the environments aiong a gradient. A similar counter-gradient pattern exists between environmentai temperature and developmental rate in anuran embryos (H. A. Brown 1967; J. A. Moore 1949b).
DEVELOPMENTAL STAGE
Fig. 8.9. Body size of larvae of Ranapipiens reared at 13"and 23C through metamorphic dimax (Taylor/Koiiros XX,Gosner 42). Vertical bars were not defined in original ieference. After Smith-Giii and Berven (1979).

peratures. A secondary consequence of larger body size in ~oplobatrachus ti>erins appeai-s to be increased nergetic efficiency in converting tadpole tissue into frog tissue (Pandian and Marian 1985b). The differentiai effeas of temDerature on rates of differentiation and " growth are evident in the field as weii as the laboratory. In Rana clamitans larvae that had not completed metamorphosis by September, differentiation ceased whereas mowth continued (Smith-Gill and Berven 1979). ~ v e n t u dthe larvae ceased both differentiation and gowtk i as temperatures dropped to below 10C. In the foiiowing spring, growth commenced earlier, at a lower temperature, than did dfferentiation. Subseciuentlv. these larvae metamorphosed at a larger size than'those'that both grew and differentiated at faster rates during the previous surnrner. Larvae with the slowest growth and differentiation were largest at metamorphic climax (Berven et ai. 1979; SmithGill and Berven 1979). Clinai variation in larvai development and morphology associated with temDerature differences has both environmental and genetic components. In Rana chmitans (lowland and montane populations) and R. ylvatica (lowland, montane, and tundra populations), the relatively cold environments at high elevation and h g h latitudes shorten the available breeding seasons, slows tadpole developmentai rates, prolongs the larvai periods, and increases the sizes of larvae at d stages in comparison to the lowland sites (Berven 1982; Berven and Gill 1983; Berven et ai. 1979). Complementary field and laboratory studies (including reciprocal transplant and breeding experiments) demonstrated that this clinai variation is primariiy induced by environmentaiefI

TemperatureE$ects on Other Promsses Energy metabolism measured by rate of oxygen consumption of anuran larvae increases approximately exponentidy as a funaion of temperature, a pattern typicai of most ectothermic animais. Qlovalues are 1.6-2.2 in Rana berlandieri (Burggren et ai. 1983; Feder 1985), 1.9-2.4 in R. catesbeiana (Burggren et ai. 1983; Feder 1985), 1.8-1.9 (Noland and Ultsch 1981) and 2.6 (G. E. Parker 1967) in R. sphenocephafa, 1.2-1.8 in Bufo tmestns (Noland and Ultsch 1981), 1.3-2.0 in Xenopw h s (Feder 1985; Hastings and Burggren 1995), and 3.2 in Limnodynastesperonii (E. Marshd and Grigg 1980b). An overd Qloof 2.14 between 15" and 25C has been derived from data for both anuran and caudate larvae (Gatten et ai. 1992). Thermai acclimation of 0, consumption occurs in larvae of three of the four species examined: Rana berhndieri, R pi@ns, and X. 1(Feder 1985; G. E. Parker 1967). The findings are s i d a r to those for many adult anurans and caudates (Feder 1985; Feder et ai. 1984a) and many fishes and invertebrates (Bdock 1955) in which thermai acclimation f i e a s rate of oxygen consumption. Comrnonly, eaothermic animais reared at coo1 temperatures have hgher metabolic rates at a given temperature than animais reared at warmer temperatures (Prosser and Brown 1961; K. Schrnidt-Nielsen 1983). Interestingly, tadpoles of the Australian leptodactylid Limnodynastesperoni show no thermai acclimation of metabolism at any temperature (E. Marshd and Grigg 1980b), and one has to ask if this trait may be a common feature of the larvae and adults of Australian anurans. Temperature f i e a s many other physiologicai variables that are ecologicdy relevant in a wide variety of eaothermic vertebrates. Examples are digestive efficiency and rates (Warkentin 1992a, b), sumivai of disease, maximum locomotor acceleration and velocity (e.g., Lefcort and Blaustein 1995), endurance, active metabolic rate (see Toloza and Diamond 1990a, b), and learning and memory capabilities (Huey and Stevenson 1979; Hutchison and Maness 1979; Kluger

210

GORDON R. ULTSCH, DAV I D F. BRADFORD, AND JOSEPH FREDA

1978). Presumably, many of these variables are iikewise affected by temperature in anuran lamae.

Osmoregulatory Physiology
Our current knowledge of ion and water balance in tadpoles has come from three principie areas: comparative physiology, properties of adult and lamal frog skin, and acidiication of freshwater habitats by acidic deposition. The latter causes mortality of amphbian lamae (e.g., Ling et ai. 1986) primarily through the disruption of Na+ and C 1 regulation. Studies exploring the mechanisms of acid water toxicity have revealed much basic information about ion regulation in tadpoles (e.g., Pierce 1985; Pierce and Harvey 1987; Pierce et ai. 1984). Although osmoregulation in tadpoles has been studied from these different perspectives, our overaii knowledge stili remains incomplete in comparison with our knowledge of gas exchange in tadpoles or osmoregulation in fishes. Thus far, ion regulation in tadpoles appears to be similar to ion regulation in teleost fishes. Significant differences between tadpoles and teleosts include an active uptake of ions across the skin, giils that are less developed morphologicdy, and metamorphosis providing a gradual transition to air breathing. Many interesting detais regarding ion and water balance in tadpoles await discovery. In this section, our current state of knowledge is presented and promising areas for research are suggested. Body Compositwn The ionic composition of the plasma of tadpoles, like most freshwater vertebrates, is markedly hyperosmotic and hyperionic relative to the surrounding environment. Ions of Na+, CI- and HCO; (table 8.3) are the major constituents responsible for 50%, 30%, and 15% of the osmotic pressure of plasma respectively (Just et al. 1977). The concentration of Na+ in bullfrog tadpoles is much more dilute than in typical teleost fishes (90 and 150 mEq/liter, respectively). In the crab-eating frog (Lzmnonectes ~ a n ~ v m sNa+ and C 1 are ), also the most important ionic constituents, but their concentrations are about 50% higher than in bullfrogs acciimated to similar conditions (table 8.3; M. S. Gordon and Tucker 1965). Limnonectescancrivorus inhabits braclush water ponds and both tadpoles and adults can survive prolonged exposure to fil-strength seawater. The plasma osmolarity of

spadefoot toad (Spea hammondii) tadpoles is also higher than that of bullfrog tadpoles, and Funkhouser (1977) hypothesized that this is an adaptation for sumivai in ephemerai ponds that exhibit high salinities caused by evaporation (also see Ferrari 1998). Little is known about the plasma ion concentrations of other species of tadpoles; as this information becomes available, it wiil be interesting to see if the concentrations of plasma electrolytes are correlated with the concentrations f saits in theirhabitats. The osmotic pressure of tadpole plasma increases steadiiy throughout development and is highest during the later stages (Degani and Nevo 1986; Funkhouser 1977; J. E. Richmond 1968). Osmolarities of 160 mOsmol/liter in premetamorphc stages of Rana catesbeiana tadpoles increased to 259 mOsmol/liter at the end of metamorphic climax (J. E. Richmond 1968). Just et al. (1977) reported a much smder increase (181 to 198 mOsmol/liter) during development of R. catesbeiana. Spea hammondii tadpoles aiso experienced a similar increase in plasma osmolarity at metamorphosis (Funkhouser 1977). This increase is presumably because of the increase in concentrations of plasma electrolytes resulting from a characteristic dehydration that occurs at metamorphosis (Just et al. 1977). This phenomenon is in keeping with the general concept that aquatic animals benefit from lower plasma ion concentrations because of a reduction of osniotic and ionic nradients between the animal and the water. Hematocrit remains relatively stable during development and averages about 30% in R. catesbeiana tadpoles (Just et al. 1977). Few measurements of osmotic concentration and comDosition of tadpole blood have been reported for species other than large ranids. Because of this iimitation, many investigators have re~orted and water concentrations as wholeion body concentrations, and large interspecific and intraspecilic variations exist (table 8.4). Caution must be taken when interpreting experiments that measure whole-body concentrations of ions and water because several factors influence their levels. The recent feedmg history must be considered because a large percentage of the mass of a tadpole is ingested material in the gut, and the ionic content of the food can directly influence the mass-specific ion content of tadpoles. The mass of the gut contents also can have an indirect effect on mass-specific ion content by infiuencing body mass. In addition, body ion and water content changes dramaticdy
U

Table 8.3 Ionic cornposition, osmotic pressure and hernatocrit of the pla srna of tadpoles Ionic cornposition (mEq/liter) Species
Limnonectes ~anrrivorus~ Rana catesbeiana

Na' 153 90 96-109

K+
5 4 3 4

CI99 68 57-60

HCO,

Osmotic Pressure (rnOsm/liter) 272 70-195 1-200

Hernatocrit
(%)

Referente
M. S. Gordon and Tucker 1965 Alvarado and Moody 1970 Just et ai. 1977 S. C. Brown et ai. 1986 Funkhouser 1977 Funkhouser 1977

25 8-24

aLimnonectes canwivorus was acclimated to 100% freshwater.

PHYSIOLOGY
Table 8.4 Ion content and percent body water of five species of Rana tidpoles lon Content (p,Mol/g dry body mass) Species of Rana Na' 573-1080 573 369 60-70'
-

Body water
C1(%)
-

K+
339-355
-

Cat+

Referente
Freda and Dunson 1984 Freda and Dunson 1986 Freda and Dunson 1984 McDonald et al. 1984 Bradford 1984b Freda and Dunson 1984 Freda and Dunson 1984

catesbeiana clamitans muscosa pip'ens sylvatica

'units
=

1365-1815 400-800 p,Eq/g wet mass.

30-35" 507 300-600

45-5 5'
-

3040' 300-825

88-91
-

87-96

as a tadpole develops. Freda and Dunson (1984) reported the body ion and water concentrations of wood frog (Rana ~lvatiuz) tadpoles every week from one week posthatching until metamorphosis. From days 7-14, body sodium concentrations increased from 400 to 800 pEq/g dry mass. This presumably occurred from a combination of active uptake across the skin and gds and the initiation of feeding. Body water also increased from 80% to 96%. The net result was that tadpoles gained weight very rapidly early in development because of the uptake of water. For example, dry mass increased by only 20%, whiie wet mass increased by 300%. This rapid increase in body mass early in development has ecological and physiological significance. Many predators (e.g., newts and salamander larvae; Brodie and Formanowicz 1983; C m p 1984) of tadpoles are gape-limited, that is, they feed on what they can fit in their mouths and swallow. Tadpoles should be most vulnerable to predation during early stages of development, and a rapid volume increase after hatching would tend to reduce predation. Throughout the remainder of development of R. sylvatica, body Na+concentration and body water gradually declined. Body Na+ concentration has been shown to be linked with tadpole size for other species. For R. catesbeiana, body Na+ concentration is negatively correlated with dry mass (body Na+ concentration = 255 g - 0.485, r2 = 0.94; Freda and Dunson 1986). Sidarly, Bradford (1984a) reported that the concentrations of body water and solute are negatively correlated with body mass, and Zamachowski (1985) found that percent body water deciined as R. temporaria tadpoles grew. The mechanisms causing this relationship are not known, but ifthe extracellular fluid space shrinks during growth and metamorphosis, body water and solute concentrations would be expeaed to decline. A second possibility is that as tadpoles grow, they contain a higher percentage of tissues (e.g., bone) that might reduce mass-specificconcentrations of water. McDonald et al. (1984) showed that body Ca" concentrations nearly triples in R. clamitans during premetamorphic and metamorphic stages. T h s presumably occurs as a result of limb development and increased cakification of the skeleton. Species comparisons are also confounded by environmental conditions (e.g., temperature, sahnity, and pH), which can have large effects on body ion and water composition (see below), and intraspeciic variation in body ion

and water concentrations. For example, Freda and Dunson (1984; see Pehek 1995) reported that R. chmitans tadpoles coilected from an acidic bog and acchated in the laboratory to water with a pH of 5.8 had a body Na+ concentration of 369 pEq/g dry mass. McDonald et al. (1984) reported a body Na+ concentration of 60 pEq/g wet mass (= 600 pEq/ g dry mass, assurning 90% water) for R. chmitans tadpoles coilected from an undescribed pond (see M. Uchiyama and Yoshizawa 1992). Conclusions about any general effects of D Hreauire further studies because the different s~ecies that I have been tested have not been of the same age or size or maintained under comparable conditions (see Bradford et ai. 1994; Schmuck et al. 1994; Verma and Pierce 1994). Water Balance As discussed in the previous section, the body fluids of most tadpoles are hypersmotic and hyperioiic .relative to the freshwater habitats in which they iive. These osmotic and ionic gradients are the driving force for the continual osmotic uptake of water and d i i v e leak of plasma ions. As with freshwater fishes, tadpoles remain in ion balance by excreting copious quantities of dilute urine and by actively transporting ions across external epitheiia. Few direa measurements of unidirectional exchanges of water in tadpoles have been reported. This is surprising because the body water concentration of tadpoles changes during growth and development and is strongly influenced by temperature (Bradford 1984b). Water may be taken up across the gds, skin, or gut. Tadpoles drink the external medum, probably during the course of feeding. Alvarado and Moody (1970) reported d r i n h g rates (Rana catesbeiana) of 0.014 mi/g.h (see Territo and Smits 1998 for a discussion of physiological calculations based on various measures of body mass), ofwhich 0.006 ml1g.h is absorbed. S. C. Brown et ai. (1986) reported preliminary measurements of urine production to be 0.006 ml/g.h, and Mackay and SchmdtNielsen (1969) and B. Schrnidt-Nielsenand Mackay (1970) reported urine flow i R. clamitans at about 0.02 mi/g.h. n Because the mass of these animais did not change during the measurements, urine flow rate approximated the uptake of water across the giiis, skin, and gut. Detailed studies ofwater uptake and loss across the gds, skin, gut, and kidney remain to be done.

212

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

Ion Rt~uiution Tadpoles continuously lose Na+, CI- and other ions because of the large concentration gradient between their body fluids and the water. In addition, ions are lost in the urine. We are not aware of studies that quanti@ the relative importance of these different loss routes. Several studies reported loss rates of Na+ and CI- for Rana catesbeiana and R. clamitans (Aivarado and Moody 1970; Dietz and Aivarado 1974; Freda and Dunson 1984; McDonald et ai. 1984). Uptake of Na+ and CI- generaiiy exceeds loss so the animais remain in positive ion baiance. Possible routes of active uptake of ions include the gilis, skin, gut, and buccopharyngeal epithelium. Initially uptake was thought to occur only at the gilis because Na+ and CI- was lost from solutions used to perfuse the gilis, but uptake was not increased if the persate was aiiowed to contact the skin (Aivarado and Moody 1970). In addition, studies of isolated larval skin did not detect active uptake or a transepithelial potential (TEP; Aivarado and Moody 1970). In support of the notion of the lack of dermai transport, active transport and TEP across the skin abruptiy appear late in metamorphosis and are correlated with an increase in skm Na+-K+-ATPase(Boonkoom and Aivarado 1971; Kawada et ai. 1969; R. E. Taylor and Barker 1965). This conclusion has been challenged because stripping the skins from the underlying musculature caused considerable stress to the preparation. In addition, no specific precautions were taken to prevent edge damage. Cox and Aivarado (1979; see Aivarado and Cox 1985) developed techniques for the study of isolated larval skin that rninimized dissection and edge damage. They found that larval skin actively transports Na+ inward at a rate of 5%10% of that of adult skm. Their work suggests that the basal membrane in the skm of larval and adult anurans is similar. The difference in the rate of Na+ transport may be caused by the absence of amilorideinhibitable Na+ channels in the apical membrane of larvai skin. The addition of nystatin (forms cation-selective channels in lipid bilayers) to the outer bathing solution stimulated Na+ influx to rates comparable to untreated adult skms (Cox and Aivarado 1983; also Zamorano and Salibin 1994). The many factors that influente the active uptake of Na' include temperature, external ion concentration, water hardness, and pH. Interspecific differences may occur, but conclusive data are lacking. The uptake mechanisms for Na+ and CI- display typical saturation kinetics. The Km and V for , Na+ uptake in Rana catesbeiana tadpoles were 0.2 mEq/liter and 0.25 ~ E q / g - h (Dietz and Aivarado 1974). Freda and Dunson (1986) reported a Km and V for Na+ transport , in R. catesbeiana to be 0.28 mEq/liter and 13.4 ~Eq1g.h on a dry mass basis (= 1.61 p,Eq/g.h; assuming 90% body water). The chloride transporter was saturated over the range of 0.2-3.0 mEq/liter and the V for C1- was 0.10-0.15 , ~Eq1g.h(Dietz and Aivarado 1974). The location and type of ceils responsible for Na+ and C1- transport have not been identiied. Na+-K+-ATPaseactivity in microsomal fractions of giil homogenate correlates weil with the intensity of Na+ transport (Boonkoom and Aivarado 1971). Na+-K+-ATPase

has also been measured in the skin of Rana catesbeiana tadpoles (Kawada et al. 1969). The transport of Na+ and CI- are independent. For example, acetazolamide inhbits the branchial influx of CIwithout affecting the influx of Na+ or the efflux of Na+ or CI- (Dietz and Aivarado 1974). The counter ions for Na+ transport are thought to be H + and NH,', while CI- is exchanged for HC03. These findings bring up another important ~ o i n t In addition to their function in sait and water . baiance, the gilis of tadpoles are also important sites of acidbase regulation and NH,' excretion. Dietz and Aivarado (1974) reported net NH,' excretion in Rana catesbeiana tadpoles of 260 nEq1g.h. McDonald et ai. (1984) reported net losses of NH; for Rana clamitans of 634 nEq1g.h. In addition, NH,' excretion in postmetamorphic animais was 12% of that in premetamorphic animais, indicating that the weilcharacterized switch from ammoniotely to ureotely had been completed. In the terrestrial BufO bufO, this switch starts at the time when forelimbs are developed under the operculum and accelerates with f o r e h b emergence but does not occur in the aquatic Xenupus luevis (Munro 1953).

Effects of Temperature When tadpoles are exposed to cold temperatures (e.g., 4"C), they rapidly increase in mass because of the uptake of water. The water content of Rana muscosa tad~oles increased bv 6% I after 1month of exposure to 4"C, after which water content remained stable for 7 months (Bradford 1984a). Between 7 and 12 months of exposure, water content again increased slightiy. During the first month of exposure, the osmotic pressure of the peritoneal fluid dropped by 14%, approached control values after 7 months, and then again declined at 12 months. It appears that water uptake is accommodated by the extraceilular fluid volume (ECF; Bradford 1984a). In a related study, S. C. Brown et al. (1986) studied the weight loss in cold-acchated (5C) tadpoles after return to warmer temperatures. Rana catesbeiana tadpoles transferred to 11" and 18C lost 7% and 10% of body weight after 5 days, whde plasma Na+ concentration increased by 28% and 21%, respectively. Daily treatment with ovine prolactin or growth hormone prevented weight loss or the concornitant increase in plasma Na+ concentration. Neither propylthiouracil nor arginine vasotocin had an effect. These authors concluded that the changes in plasma solutes reflected changes in ECF volume. The adaptive significance, ifany, of cold-induced edema in larval or adult anurans is not known. It would be interesting to know ifwater gain was simply because of an inhibition of osmotic regulatory mechanisms at low temperature, or if it is a specific physiological adaptation to winter conditions. Effects of Salinity Frogs occur in aquatic habitats that range widely in salinity. Many habitats are extremely dilute, such as soft-water bogs, rainwater fiiled temporary ponds, and water in granitic basins. Rana catesbeiana tadpoles can be reared in distiiled water (Aivarado and Moody 1970). Similar to frogs and fishes, exposure to distilled water or very dilute solutions stimulates

PHYSIOLOGY

213

the sodium transport system. Elevated V for Na+ trans, port and an unchanged Km (Dietz and Aivarado 1974; Freda and Dunson 1986) may reflect an increase in the total number of transport sites. Freda and Dunson (1986) also found that exposure to high Na+ concentrations increased Km and depr&sed V,,,,. ~ h e s changes in the characteristics e of the transport mechanism would help maintain Na+ balance in dilute water and reduce energetic costs of ion regulation in water with high concentrations of Na+. It is not known whether branc6al or s h permeability is adjusted in response to water saiinity. Some amphibians occur in highly saline habitats such as ponds and streams d u e n c e d by saltwater spray or tidai movements and desert ponds as they evaporate. The majority of field studies have focused on the distribution of aduits with referente to salinity (Christrnan 1974; Ruibai 1959, 1962). Manv s~ecieshave been found in braciush salt marshes, and it appears that species vary in their tolerance to salinity. A few specialized species (Bufo viridis and Limnonectes cancrivoms) can survive in 40% and fuli-strength seawater, respectively (M. S. Gordon 1962; M. S. Gordon and Tucker 1965; M. Uchiyama et al. 1990). Very few studies of the distribution of tadpoles with regard to habitat salinity have been re~orted. The best:known euryhahe amphibian is the crab-eating frog (Limmnectescancrivoms) of southeastern Asia. The aduits and tadpoles may be found in braciush water ponds and mangrove swamps. The tadpoles can survive exposure to fdi-strength seawater (M. S. Gordon and Tucker 1965; M. Uchiyama et al. 1990; M. Uchiyama and Yoshizawa 1992) and have been found in ponds ranging from 16% to 75% seawater (M. S. Gordon and Tucker 1965; Dunson 1977). However, M. S. Gordon and Tucker (1965) couid not induce tadpoles to metamorphose in the lab if water salinity was griater than 20% Seawater, and the largest tadpoles were obsemed in the most saline ponds. These obsemations suggest that L. cancrivoms tadpoles cannot metamorphose at high salinities, but this suggestion awaits more rigorous verification. The adults are osmoconformers and are able to survive in highly saline waters by elevating plasma urea, which raises plasma osmotic pressure and thereby elirninates the driving force for the dehydration that would occur in saltwater. Larvae are osmoregulaters; ifwater salinity is increased from 100% freshwater to 100% saltwater (a more than 10-fold increase in osmotic pressure), plasma osmotic pressure only doubles (M. S. Gordon and Tucker 1965). Over d salinities (0%-100% seawater), 90%-100% of the osmotic pressure of plasma is caused by Na+. K+. and C1-. Similar to marine teleost fishes. these tadpeles osmoregulate in seawater by d r i n h g th hyperosmotic medium and by excreting a s m d amount of isosmotic urine. Aithough it hai never bein verified, the tadpoles probably excrete salts by an extrarenal pathway such as Na+ transporters in the skin and gis. M. S. Gordon and Tucker (1965) also found that tadpoles become osmoconformers as froglets (Taylor/Kollros XXV, Gosner 46). In 80% seawater, starting at Taylor/Kollros XX (Gosner 42), plasma osmotic pressure increased and plasma Na+, K+, and CI- concentra, L

tions did not change. This pattern was presumably the resuit of the initiation of urea production. More detailed experimental data are clearly needed on the physiology of this interesting species.

Effects o LowpH f Concern over acidic precipitation has stimuiated a great deal of recent research on the effects of acidic water on amphibians. Lowered rain pH over the northeastern United States, southeastern Canada and Europe has been implicated in the depression of alkalinity and pH of many freshwater habitats. Acidic mine drainage associated with coal and metal strip mining can also seriously impact freshwater habitats (Kinney 1964; Porges et ai. 1986). Aithough anthropogenic acidification has received much attention, it is generdy not realized that many freshwater habitats are acidic because of naturdy occurring humic acids or the biological activity of Sphapum (reviews by Gorham et al. 1985; Kilham 1982; see Rosenberg and Pierce 1995). Naturdy acidic waters have been described throughout the world: major branches of the Amazon and Congo rivers (Marlier 1973); boreal peat bogs of North America, Europe, and Asia (Gorham et al. 1985); many waters of the eastern and Gulf Coastal plain of the United States (Gosner and Black 1957%) and the southwest cape of South Africa (Picker 1985; Picker et al. 1993); heathlands of Great Britain (Beebee and Grifin 1977); swamps of Maiaysia (D. S. Johnson 1967); and ponds of northern Austraiia (T. E. Brown et al. 1983). Detailed discussions of the distribution of acidic waters in the United States, the relative tolerance of different species, and the ecological effects of acidic waters are reviewed by Freda (1986) and Pierce (1985). Tadpoles are more tolerant of low environmental pH than embryos but less so than aduits (Freda 1986). Amphibians as a group are much more tolerant than fishes. The primary toxic effect of low pH is dismption of Na+ and CI- balance (Freda and Dunson 1984; McDonaid et al. 1984). Acute exposure to lethal p H mhibits the active uptake of Na+ and C1- while causing a massive stimuiation in passive Na+ and C1- loss. InAux of Na+ is linearly related to pH and 100% inhibition occurs near pH 4.0. The loss of Na+ increases exponentidy just above the lethal pH. In d species tested, death occurs when body Na+concentration is reduced by about one-half. The increased efflux of Na+ and C1- at low pH is quantitatively responsible for the reductions in body ions, and variation in the rate of loss defines intra- and interspecific variation in acid tolerance (Freda and Dunson 1984). The inhibition of Na+ influx may be caused by direct competition of H + and Na+ for the carrier sites (which may not be the sarne for the two ions), as increasing the concentration of Na+ in acidic water elevates Na+ uptake (Freda and Dunson 1986). Ion loss presumably occurs across the s h and giils. In fishes it is thought that Na+ loss at low pH results from the leaching of calcium from the branchial epithelium, which opens up tight junctions and compromises the structurai integrity of the gdi epithelium (Freda and McDonald 1988; also see Suffler 1996). Chronic exposure to sublethal p H generdy causes a 20% decline in body Na+concentration and stimulates compensa-

214

GORDON R. ULTSCH, DAVID F. BRADFORD, AND JOSEPH FREDA

tory mechanisms to help maintain ion balance (Freda and Dunson 1984, 1985, 1986). The influx of Na+ in Rana pipiens tadpoles was first depressed but then exceeded conGol values during seven days of exposure to pH 4. 5. Efflux of Na+ was initidy stimulated but later declined to levels just above controls (Freda and Dunson 1984). Similarly, McDonald et al. (1984) reported that Na+ influx and efflux in R. ciumitans in closed chambers where the water was initiaiiy of DH4 . 0 4 . 2 returned to control values after 7 hr of emosde. Aithough the inhibition of influx is not significant during lethal exposures, restoration of influx during sublethal exposures is important in maintaining ion balance. Freda and Dunson (1986) showed that exposure to low p H exposure acts simiiarly to other salt-depleting conditions (e.g., distiiied and soft water) by increasing the V for Na+ trans, ~ o r tThe reduction in sodium efflux that occurs after mo. longed exposure h a s not been studied but presumably involves a reduction of Na+ permeability of the giiis, skin, or both.

Summary
To understand the ecological physiology of tadpoles as a group, it is essential to examine phylogeneticdy diverse tua and diverse environmental conditions. At present, most of the knowledge about tadpole ecological physiology is based on very iunited subsets of taxa and environmental settings. For example, most ecophysiological studies have been done with temperate zone species, rather than tropical species, yet the greatest diversity of taxa are tropical. Even within the temperate zone, the preponderance ofwork has been iunited to members of the Bufonidae and Ranidae. In part, this is because relatively few investigations have been conducted on amphibian larvae in comparison to aduit amphibians and other vertebrate groups. Also, these studies have often sought to understand how amphibian larvae d s e r from aduits and other vertebrates rather than to understand how tadpole biology M e r s among species and environments. Investigators have tended to work with local species or species easiiy maintained in the laboratory. Future ecophysiological studies should ask questions concerning the interaction of environmental parameters and the physiology of tadpoles in environmental settings that are

poorly known. Studies of ontogenetic differences wouid be helpful (e.g., Burggren 1984; Burggren and Fritsche 1997). Examples may be temperature relations of tadpoles in tropical forests with little fluctuation in temperature and gas exchange of tadpoles mhabiting bromeliads. To the extent possible, such studies should be designed to compare phylogeneticdy disparate taxa and different settings in order to distinguish between phylogenetic and environmentaiiy induced patterns. We shouid also attempt to understand the physiology of tadpoles in the context of the many considerations faced by the individual (e.g., obtaining food, avoiding abiotic and biotic hazards, and maintaining an internal environment within tolerable lunits; Feder 1992). For example, behavioral temperature regulation can expedite development, and air gulping can compensate for low oxygen levels in the water. In contrast, the former may lead to stranding, and both activities may render an individual more susceptible to predation (e.g., Lefcort and Eiger 1993). Our challenge is to understand the trade-offs between the physiology-driven processes and other considerations. This task should be interesting, although perhaps difficuit, because many amphibians can tolerate wide variations in some physiological parameters (e.g., body water, temperature, and osmolyte concentration) in comparison to other vertebrates, especiaiiy birds and marnmals (Feder 1992).This is exemplified for tadpoles of many species by their relatively wide temperature tolerance ranges (table 8.2). Aiso, many aspects of the physiology of tadpoles change during larval development. This apparently wide tolerance of some physiological conditions combined with ontogenetic changes in physiology may aiiow tadpoles considerable flexibility in accornmodating to the spatial and temporal variations in their abiotic and biotic environments.

ACKNOWLEDGMENTS
We thank T. Dietz, A. Riggs, and R. J. Wassersug for editing parts of this chapter. Additional cornnients were provided by W. Burggren and M. Feder. Bradford acknowledges the environmental Science and Engineering Program at the University of California, Los Angeles, for support during the early stage of preparation of t h s chapter.

BEHAVIOR
Interactions and Their Consequences
Karin vS. Hoff, Andrew R. Blaustein, Roy W. McDiarmid, and Ronald Altig

Introduction
A tadpole is an energy-gathering,growing, nonreproductive larvai stage in the biphasic anuran life cycle. Various authors (e.g., Wdbur and Cohns 1973) have modeled the impacts that different aquatic environments have on tadpole gr&th and feeding, and growing to the largest size in the shortest period remains a driving force for selection. Optirnizing growth enhances the probability of completing metamorphosis and reaching the reproductive stage. Forms of paedomorphosis that result in a larval individual capable of reproduction occur in saiamanders but are unknown in anurans. As a resuit, behaviors usuaiiy associated with sexual activity and species recognition (e.g., courtship displays) among reproductive adults are absent in larvae. Instead, the behaviorai repertoire of tadpoles is restricted to activities that augment development, growth, and survival to metamorphosis. Aggregations of tadpoles in nature are relatively common (see Duellman and Tmeb 1986; Lescure 1968; Lyapkov 1996; P. J. Watt et al. 1997) and may be interpreted broadly as ecophysiological, metamorphc, feeding, and social responses. Relative to aggregative behavior, we focus primarily on feeding and social interactions. Three basic subject areas provide the organization for this chapter: (1) aspects of general maintenance (feedmg, respiration, thermoregulation, and locomotion); (2) dispersion and social behaviors (habitat selection. social interactions. and kin recognition); and (3) sensory abilities (vision, cutaneous senses, olfaction, and chemical communication) and learning. Tadpole behavior is not weii studied, and we point out specific areas that need attention. i notations of developmental stages foilow the system of Gosner (1960; see table 2.1 and fig. 2.1).

I find studying the behaviour of animais in their natural surroundings a fascinating hobby.
Tinbergen 1968:iii

General Maintenance Feedin~

Tadpoles usuaiiy are considered highly speciahzed, fiiterfeeding herbivores (e.g., Duellman and Tmeb 1986). Most ingest planktonic material from the water column, obtain organic materiais from pond sediments, or scrape material (i.e., aufwuchs, periph~on)from submerged substrates. Given the diverse and son~etimesbizarre morphology of their feeding structures, the variety of sizes and kmds of food items recorded in their guts, the absence of ceiiuiase for digesting plant materials, the diversity of microhabitats occupied, and our lack of basic knowledge of their diets determined from conventional gut content analyses convinces us that tadpoles are better thought of as opportunistic omnivores or detritivores.

216

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG sites (benthic, midwater, surface) throughout the water column, and have characteristic morphologies and behaviors. To demonstrate t h s ecological and morphological diversity and to provide some background for reading this chapter, we illustrate two somewhat hypothetical larval comm~mities, one typical of a forest pond 9.1) and the other of a forest stream (fig. 9.2). V T selected our examples from the NeoCe tropical region because that is a fauna with whch we are familiar. It also is broadly representative of faunas in other tropical regions, which are ecologicaily arid taxonomicaily diverse compared to tliose in comparable habitats in temperate zones. More detailed illustrations of these and other tadpole morphotypes are shown in figure 12.1, and their associated mouthparts are shown in figure 12.2. Heyer (1973b) described the microhabitat use, feeding and swimming behavior, and diet of pond-dwehlg tadpolcs of 1 7 species in a dry forest site in Thailand, and Inger (1986) a l d Ingcr et ai. (1986a) reported a comparable study for 12 stream and 4 nonriparian tadpoles of 16 frog species from a wet forest site in Borneo. Pond tadpoles of the African frog Xenupus laevis are typicaiiy positioned head downward at about 4S0, while those of&alychnis p w e l l i (Duellman and Trueb 1986) and otlier phyllomedusine hylids usuaiiy position thcmselves upward a t .about 45'. suspension-feediig microhylid tadpoles usuaiiy float horizontdy near the surface or in midwater, and tadpoles of Rbinopbvynus dorsalir usuaiiy swim horizontaiiy. The buccal purnping mechanism is similar in at least ail rasping tadpoles, but species d 8 e r in the efficicncy of entrapment based on the sizes (i.c., 0.126 in Wassersug 1972 to greatcr than 200 p.m in Seale 1980) of particles that are capGred. These differences in efficiency have been correlated with variations in skeletal elements of the purnp (Wassersug 1972; Wassersug and Hoff 1979) and density of the giil aters (Wassersug 1972). Tadpoles without keratiriized mouthandXenopus) cannot graze on plant parts (e.g., G ~ o p h l y n e material that is too laree to fit into their mouths. and tad" poles lacking ultraplanktonic entrapment surfaces (e.g., Hymenocbirur) feed on items large enough to be retained without entrapment surfaces. Wassersug and Hoff (1979) noted specific behavioral correlates with buccal pump design. Deflcaion of the floor of the cavity by a levcr aaion of the ceratohyal causes the buccal cavity to act as a hydraulic pump. The area of the bucal cavity floor, the length of the lever, and the amount of deflection determine maximum buccal volume and maximum suction force (negative pressure) generated by the pump. Microphagous tadpoles tend to have large buccal volumcs and short lever arms and regulate fecding rate by frequency moduiation (FM) of pumping frequency (e.g., Xenopus laevzs). Tadpoles with relatively longer lever arms vary buccal volumes and pressures by amplitude modulation (AM) the ceraof tohyal defleaion (e.g., Rana yl17atica). Suctorial forms that rely on buccal suction to adhere to surfaces (e.g., hcaphus) have large lever arms and smail buccal volumes to generate high pressurcs. Macrophagous tadpoles have large volumes and long lever arms. The validity of morphological modeling to predict feed-

Stimuli emanating from food sources that initiate locomotor or feeding responses (e.g., P. H . Harrison 1990) also have not been studied well. Svstematic and com~arative evaluations of the food habits of tadpoles are uncommon (but see Diaz-Paniagua 1985; Heyer 1973b; Inger 1986 and referentes therein) and generaily not very informative because many ingested items pass undanlaged through the gut, whiie other soft-bodied oreanisms and bacteria are not dea tected. Such observations have led some authors to suggest that the real sources of food for tadpoles may be dissolved inorganic and orgaxiic nutrients, bacteria, and viruses (Heyyr 1973b; Inger 1986). The contribution of these s m d itcms to tadpole diets remains unstudied in nature, and that part of thematerial that a tadpole ingests that is actually digeSted is not known. Items re~orted from the V t s of tad~oles m include sand. detritus, viruses, bacteria, smaii unicellular organisms (protists), algae (diatoms, filarnentous green, euglenophytes), plant fragments of assorted sizes, pollen grains, hngi, various kinds of s m d animals (axmelids, cladocerans, copepods, gastrotrichs, insccts, nematodes, rotifers, tardigrades, water mites), anuran eggs, and heterospechc and conspecific tadpoles (Costa and Balasubramaniam 1965; Diaz-Paniagua 1985; J. D. Harrison 1987; Heyer 1973b; Inger 1986; Lajmanovich 1994; Sabnis and Kuthe 1980; Sekar 1992). Tadpoles also are known to feed on fecal material (Steinwascher 1978a) and swldry typcs of carrion (= necrophagous group of Beiswenger 1975). In al cases, caution must be exercised l in interpreting feeding habits and their possible correlations with morphology. A. Haas (personal communication) recently found intact ephemeropteran larvae in an unidentified ~oophis tadpole from ~ a d a ~ & c aThis situation was unexr. pected in a suctorial tadpolc that lives in very fast water and maintains position with an exceptionaily large oral disc. Research has focused on evidence of food selection from several approaches: stomach content analysis (e.g., Jenssen 1967); comparative aspects of buccal anatomy related to feeding (e.g., Viertel 1985; Wassersug 1980); comparative mechanical restrictions on feeding behavior based on anatomy a i d limitcd experimentation(e.g., Wassersug and Hoff 1979); feeding dynamics of a smaii number of species in the laboratory (e.g., Sede 1980); and relationshps between feeduig modes and microhabitat use, oral morphology, diet, and phylogenetic groups (e.g., Altig and Johnston 1989; Diaz-Paniagua 1985; Heyer 1973b; Ingcr 1986). Warkentin (1992a, b) examincd the effects of temperatiire and Uumination on microhabitat use and feeding rates of Rana clamitans tadpoles. Most tadpoles feed by producing currents that carry particles into the bucal cavitv and across food entraDment surfaces. A tadpole opens its mouth (ser fig. 4.16), depresses to the floor of the buccal ca~ity draw in water, closes the mouth, and then elevates th floor of the buccal cavitv to pump water across food entrapment surfaces and out through the spiracle(s) (e.g., De Jongh 1968; Gradwell 1972b, c; Kenny 1969b, c; Severtsov 1969a; Wassersug 1972). Tadpoles occur in countless aquatic habitats, feed at many

(fig.

BEHAVIOR

Fig. 9.1. An ecological diorama of a hypotheticai composite tadpole community in and near a neotropical pond. The mdpoles of six species with distina morphotypes that might occur there are iustrated exhibiting typical behaviors (e.g., feeding, swimming, hiding, etc.). Top center: Hyla b m m e l e (Hylidae), an arboreal (Tjp 1) tadpole that occurs in the central axil of a bromeliad. Top right: Hyla miroqhau (Hyiidae), a midwater macrophagous form with reduced mouthparts searching for s r n d crustaceans among stems of aquatic plants. Lower

right: Scinuxstuu@n (Hylidae), a nektonic midwater form with high fins that scrapes food from leaves and stems and other undenvater surfaces. Right center: Rhinophynushalk (Rhinophrynidae), a filter feeding tadpole that swims in large, midwater schools. Left center: Lcp t o ~ 1 w p e n t ~ L (Leptodactylidae), an omnivorous benthic tadus pole that often is carnivorous on hatchlings and small tadpoles. Lower left: B u . murinus (Bufonidae), a black, benthic, rasping tadpole that sometimes aggregates.

ing behaviors is supported by studies of feeding dynamics in 1997; aiso see Caidweii and De Arajo 1998 and Nguenga Ranu ylv& and Xempus laePrj (Sede et ai. 1982 and Sede et ai. 1997). In their review of arboreai tadpoles, Lannoo and Wassersug 1979; see feeding dynamics below) and aiso et ai. (1987) tabulated species known to consume frog eggs; is consistent with the suggestion (Wassersug and Seibert other hylids that breed in phytotelmata (i.e., bromeliads and 1975) that the buccal pump design of tadpoles refleas feed- tree holes), including Phrynohym resinzjhm (R. W. McDiaring more than respiratory needs (see Respiration below). mid, personal observation), Osteocephalus d iqmeurii (R W . . Polis and Myers (1985) tabulated 14 species fiogs McDiarmid, personal observation), 0.oophqgus (Jungfer and whose tadpoles ate conspecific tadpoles or eggs, and other Schiesari 1995); Anothecu spinosa (Jungfer 1996) and 0. species have been added to the list (e.g., Osteopilus ieptencvio- phnieps (L. Haugen, personai cornrnunication) can be added d j Crump 1986; aiso see table 10.2). Crurnp (1986) dis- to that list. ChirUcdw etfingm' (Rhacophoridae)nests in tree I, tinguished between opportunistic cannibalism (such as oo- holes and broken or cut bamboo stems and has oophagous phagy and necrophagy, in which the prey does not attempt tadpoles (Ueda 1986; Karn et al. 1996). Tadpoles ofPhilautus to evade the predator, and cannibahsm, in which conspecifics sp. aiso found in a tree hole in Thailand apparently feed on are attacked, killed, and eaten (e.g., Summers and Amos frog eggs, likely their own (Wassersug et ai. 1981a).

218

K A R I N vS. H O F F , A N D R E W B L A U S T E I N , R O Y M c D I A R M I D , R O N A L D A L T I G

Fig. 9.2. An ecological diorama of a hypothetical composite tadpole community in and near a neotropical stream. Likely occurring representative morphotypes are shown in different rnicrohabitats and exhibi ~ typical behaviors (e.g., feeding, swimming, hiding, etc.). Top g right: Thmopa miliari (Leptodactylidae), a semiterrcstrial tadpole that lives on the wet surface where rivulets of water flow over a vertical rock faces. Lower right: Cocbra~~ugvanulosa (Centrolenidae), a fossoria1 tadpole that buries within leaf packs in stiU.water. Right center: O a > p ~ p y b u r n(Microhylidae), a psamrnonic tadpole buried in sand i

in shailow, slow-moving water. Left center: P h ~ h ~ u t t a(Hyt a lidae), a neustonic tadpole feeding from the surface film in shallow water with an umbeiiiform oral apparatus. Lower left: Hyh annata (Hylidae), a suctorial tadpole that attaches to rocks i fast water. Top n left: Elcuthemda&u~ ridenr (Leptodactylidae), embryos of a &rem developing species that placa its eggs among fallen leaves that often accumulate in the axils of understory plants. Top center: PhynuLyll~ resintjitm (Hylidae), an arboreal (Type 2), omnivorous, often times oophagous, tadpole that occurs in tree holes.

Cannibalism may be more common than reported, espen ciaiiy among tadpoles of frogs that breed i smaii, ephemerai ponds and phytotelmata (e.g., bromeliads and tree holes) that are subject to unprediaable drying or where resource shortages and overcrowding may be common and survivorship low (Bragg 1957, 1965; Crump 1986, 1990, 1992; Downie 1990b; see chap. 10). Cannibaiistic tadpoles most commonly eat conspecifics of different sizes or develop-

mental stages. Large tadpoles may eat eggs (Crossland 1997; Crossland and Aiford 1998; Heusser 1970a) and hatchlings (Crump 1983), and premetamorphic tadpoles may prey on metamorphosing individuais (Bragg 1957; C m p 1986). Heusser (1970a) speculated that conspecific oophagy may be yet another reason for synchronous oviposition, and this argument can be extended to synchronous metamorphosis. n Reports of tadpoles eating conspecific tadpoles i the field

BEHAVIOR

219

have not aiwavs made it clear whether the orev were aireadv dead before they were eaten. However, conspecific tadpole cannibaiism has been observed in the field (Cmmp 1990, Hyhpseudupuma); Petranka and Thomas (1995) argued that explosive breeding in Rana sylvatica reduced egg and tadpole cannibalism. In laboratory studies, Cmmp (1986) dowed tadpoles of Osteopilus septencrionalis to me&norphose in aquaha with emergent surfaces to reduce the likehhood of drowning. The presence of these surfaces did not eliminate cannibalism by younger tadpoles in the sarne aquaria, and rnetarnorphosing tadpoles (Gosner 42-43) in both experimental treatments (i.e., those that were not eaten by conspecifics) climbed up the verticai glass waiis. Aii metarnorphosing froglets survived in the control treatments (i.e., those without younger conspecifics). It seems k e l y that cannibaiism, at least in this species, is active predation by younger (Gosner 36-39) tadpoles, perhaps siblings. Cmmp (1986) speculated that metamorphosing tadpoles were more vuinerable to predation because they had diminished locomotor capacity compared to either tadpoles without forelegs or froglets without taiis. Wassersug and Sperry (1977) aiso demonstrated that anurans in the middle of metarnorphosis (four legs and a tail) do not swim as fast as tadpoles younger than Gosner 42 or hop as far as froglets older than stage 45. It may be that tadpoles prey preferentiaiiy on conspecific tadpoles when they are most vuinerable; t h s is the same stage at whch they are aiso preferentiaiiy preyed upon by other species (e.g., Tbamnuphis, S. J. Arnold and Wassersug 1978). Some tadpoles (e.g., Hymenocbims and Lepidobatracbus) are specialized for consuming large food items (Wassersug and Hoff 1979). and some are activelv oredaceous on heter, ospecifics. The hypertrophied jaw serrations of the tadpoles of Otopbryne pyburni (Wassersug and Pyburn 1987) and Mantdu@ylus lu.ubrtr (R. Altig, personai observation) superficidy appeai to be specializa~ons carnivory but are for probably used to sort particles during filtration processes. Wassersug and Pyburn (1987) suggested that the tadpole of Otopbrynepyburni feeds by passive aspiration caused by water flowing past the aperture of the elongate spiracle tube. An arnazing example of feeding behavior and parentoffspring com&unica<on has been dlocumented in &e Centrai Arnerican frog, Dendrobatespumilw. After 10-12 days of development, tadpoles hatch from eggs (clutchesrange from 5 to 20 eggs; mean of 10 clutches = 10.9) laid and fertiiized on horizontal dry leaves (Weygoldt 1980). The femaie parent transports a single tadpole on her back to a suitable bromeliad and releases the tadpole in a leaf ai with water. Perixi odicdy (1-9 day intervals), she returns, inspects the leaf axil holding her tadpole, and deposits an unfertilized egg which the tadpole eats. When a tadpole of D. pumilw detects (see Sensory Abilities, Vision) a frog visiting the bromeliad axii in which it lives, it begins a speciai tail-vibrating "dance" that can be seen by the visiting frog. Instead of swimming with the usual taii undulations.' the \ I creates noticeable cirtadoole cular movements beneath the water surface by stiffening and rapidly vibrating its taii. If the visitor is the femaie parent coming to feed her tadpole, she backs into the axd and the
L ,
, A

tadpole swims against her hindquarters to stimulate her to oviposit nutritive (unfertilized) eggs. The tadpoles of D. pumilw appear to be obligate egg feeders, and attempts to raise them on other foods have failed (Wev~oldt 1980). This iu remarkable behavior characterized by posthatching parenta1 care, communication between a tadpole and its femaie parent, and the use of nutritive eggs to feed ones offspring was unknown among arnphbians Ghen Weygoldt published his extraordinary observations with captive frogs. This behavior was later observed in D. pumilw in the field in Panarna (Brust 1993) and k e l y is characteristics of aii eight species in the group (e.g., ~ G g f e r 1985, D. qecwsus; ~i&ermann and Zimmerman 1981,D. bistriunicus). Another remarkable case of tadpole-mother communication that aiso is associated with debosition of nutritive eggs was described by Jungfer (1996) for Anotbeca spima. Tadpoles of this species normaiiy are found in tree holes, and in the laboratorv thev show a oeculiar behavior when the female parent enters the basin containing her larvae. The norm d y inactive tadpoles swim around the mother, touching her with their snouts and "sucking" slightiy on her skin; just seconds before nutritive eggs are extmded, the tadpole movements become faster and more hectic, sometimes including bites on the extnided cloaca of the mother (Jungfer 1996). Usuaiiy, the eggs are eaten imrnediately by tadpoles (fig. 9.3A). Other species of bromeliad and tree-hole breeding frogs aiso may rely on deposition of eggs to provision their tadpoles. Are the eggs laid in a bromeliad that already contains tadpoles provided by the femaie parent of the tadpoles? Are the eggs nutritive and f e d e (i.e., capable of developing into tadpoles) or nonfertile and provided only as a food source for the tadpoles? What is the role of the maie frog in these situations, and is there any communication between the tadpoles and the parent(s)? The answers to these and other intriguing questions are beginning to appear. In their description of Osteocepbalus oopbap, Jungfer and Schesari (1995) noted that femaies, clasped by a maie, returned at intervais of about five days to brmeliads that contain their tadpoles and deposit eggs that are eaten by the tadpoles (fig. 9.3B). If no tadpoles are present the eggs develop. The tadpoles apparentiy are oophagous and starve if eggs are not provided. The tadpoles of CbirZ?calusezfin~eti(Kam et ai. 1996) and Mantelh 1aeYigata may be other candidates for this type of nutrition. Different types of parent-offspring communication, other than those described for some oophagous tadpoles, may be more common than we think. Tadpoles ofLeptodactylus bolinianus move toward the source of smaii waves made bv rhvthmic movements of their mother, and by her movements, she leads the tadpoles to other sites (Weiis and Bard 1988; see Dowriie 1996). When the frequency of rain slows and puddles begin to dry, male Pyxicepbalus adspersus constmct channels between puddles and lead tadpoles through the interconnecting channel to other parts of the system that contain more water (Kok et ai. 1989). These reports of parenttadpole communication are exciting observations and provide a fertile arena for future investigation.
\

220

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

Fig. 9.3. Examples of parent-offspring cornrnunication and behavior include the provisioning of numtive eggs to tadpoles and transport of larvae from egg deposition sites to aquatic habitats. (A) Three tadpoles ofAmthccu pimsa (Hylidae) eating nutritive eggs oviposited by their mother (see Jungfer 1996). (B) Tadpoles of Osteoqhdus o p w s oh u (Hylidae) feeding on eggs laid by a amplexing pair, and the fernale is n

the parent. (C) Tadpoles of H m k marmmatur (Hernisotidae)on and beneath the attending fernale parent in a nest cavity in Camo National Park, Ivory Coast (see Rdel 1996 and text for discussion). (D) Four tadpoles atop a nurse frog of Colostethu n c p u (Dendrobatiwiz dae) from the Rio Cenepa drainage in Amazonas, Pem.

Two other interesting examples apparently involve communication between the tadpole and parent. Wager (1929) reported maternal care in Hmisus mammatzw. Femaies remain in a hoilowed-out cavity atop the developing eggs until they hatch. Newly hatched tadpoles are lively and "indefatigable wrigglers." Based on excavations of burrows and other circurnstantial evidence, Wager (1929: 129) commented that ". . . it seems certain therefore that the mother burrows towards the water making a tunnel down which the tadpoles wriggle in a squirming mass until they reach the water." These observations were repeated without qualification by Wager (1965, 1986) and have been repeated in the general literature (e.g., Dueiiman and Tmeb 1986). On reviewing d the published reports, Van Dijk (1985) rejeaed Wager's scenario and noted that others had reported hatched larvae swarrning over the back of an attending female and on the back of another frog, presumably a female. Van Dijk suggested that female Hemhus carry their tadpoles to the water. M.-O. Rodel (personal communication) reported tadpoles foiiowing a female as she constmcted an open channel from the nest to water; during this period some of the tadpoles

were attached to the female. Other observations indicate that ms as ponds fill female H i w open flooded nest charnbers to allow tadpoles to escape. Circurnstantial evidence (e.g., nests more than 100 m from water without any indication of channels, tadpoles in rock pools above the leve1 of the surrounding savannah) suggested to M.-O. Rodel (personal communication) that maternal behavior (digging channels or transporting larvae) of H i u marmmatus may vary demss pending on nest site topography and the surrounding habitat. Although larvae have been observed on female H m s eiw mumuwatus in the nest cavity (fig. 9.3C) and in open channels, no one has actudy observed a female Hemtszw carrying a tadpole across land. Larval transport is weil known in dendrobatid frogs (personal observations; fig. 9.3D) and a few other species ( D u e h a n and Tmeb 1986). Larval transport in dendrobatids usudy involves the male, except in those species where femaies provision the larvae with nutritive eggs (see above). Wells (1978) reported that maie Dendrobates auratzw may use visual or tactile cues to determine when tadpoles can be carried from the nest site (usudy in leaf litter) to a suitable

BEHAVIOR

221

aquatic site (bromeliad or tree hole). When tadpoles become mire active just prior to hatching, attending mdes assume a distinctive posture on the clutch with back arched and ankles pressed together to form a trough. Tadpoles attempted to wriggle onto the maie's back but often feil off. In D. a u r a only one or a few tadpoles are carried at a time; in other species most of a clutch may be transported. Stebbins and Hendrickson (1959) described larval transport in Colostethw subpuncta noting that attachment is by Sticky mucus produced by the nurse fiog and not by the larval mouth. They described the release of larvae but had no information on how the tadpoles got onto the male's back. Additionai data on parent-l&va communication prior to transport couid be obtained fiom cuitured frogs. Active wriggling by the hatchling larvae seems to be an obvious parallel in ail of these cases and ~robablv serves as a visual or baile cue to initiate the appropriate p&enml behavior. This seems to be a fertile area for future investigation. Under certain unknown and seemindv uncornmon circumstances, tadpoles that occur in temporary sites with bare soii substrates form large expanses of tightly spaced feeding depressions. In the sedimentary literature (see Dionne 1969; T. D. Ford and Breed 1970; Maher 1962; Opatrny 1973), these depressions have been called "tadpole holes" or "tadpole nests" for weil over 100 years. J. H. Black (1974, Bufo americanus and Pseuumb s~reckeri;see 1. H. Black 1971) noted these holes once in s k years of observing about 30 pools, and one of us (R. Altig, personal observation) has seen these holes oniy at two sites during the same summer (Bufo t d and Pseuahmh heriata) over 26 years of field
" 2

observations. It is as if the tadpoles are deiiberately seeking s~ecific material beneath the soii surface. The tail of a feedina &dpole vigorously beating against the bottom produces i circular, hexagonal, or pentagonai b e m about the feeding ~ o i n that delimits a deoression about twice the lenmh of t &e tadpole (fig. 9.4). ~ h e s holes are not ripple interferente e marks made simply by tadpoles swimrning (Dionne 1969). The iim that develops on the surface of the water apparently is a bountifui place to feed (e.g., Danos et ai. 1983; Goldacre 1949), but the circumstances that promote this feeding mode is not known. Tadpoles with umbelliform oral discs (see chaps. 3 and 12) specialize i feeding fiom this n iim (fig. 9.5). Because of the orientation of the disc, they feed in a siightly tail-down but upright attitude relative to the surface. Typical benthic and nektonic tadpoles commoniy feed fiom the surface filrn in ponds and skw-moving streams. These tadpoles either turn vertically or roli over on their backs (e.g., species ofBuj5, Hyla, Ranu, and Scaphiopus) when feeding fiom the surface fiim. Hatchlings commoniy hang from the surface filrn (see Wangersky 1976), and one could speculate that they gain some of their first nutrition fiom the surface just as they begin feeding. The details of feeding behavior and optimal foraging are known for only a few species. Feeding dynamics have been investigated mostly in tadpoles of species that are obligate suspension feeders (e.g., Schoonbee et al. 1992, X e q m hmi). Suspension clearance rates and buccal pumping rates are relatively easy to monitor i the laboratory. Sede et ai. n (1982) exarnined the relationships among food concentration, buccal pumping rate, and-ingestion rate in Xempu

Fig. 9.4. Tadpole holes (depressionsi the muddy botmm) formed by feeding tadpoles of Buf anwhXr n (Bufonidae), photographed in June in Jackson County, Mississippi.

BEHAVIOR

223

lua& and fitted the data to either a rnodified Monod spatial distribution of potentiai tadpole foods in ponds must (Michaelis-Menten) or an energy optimization model. Tad- accornpany future laboratory seiection experiments. Ponds pole feeding dynamics resernble those of other suspension- that experiente intense aigai bloorns (e.g., Sede 1980) rnay feeding organisms and provide for both high energy return not present opportunities for selection among food types. to the tadpole and stabhty of populations of food organMany authors have suggested that microhabitat use and isms. Xenopus b s tadpoles, presented with concentrations temporal partitioning (e.g., Degani 1982; Heyer 1976a; Inof food (yeast) ranging from 5 X 10-3 to 5 X 102mglliter, ger 1986; Loschenkohl1986; chap. 10) reduce interspecific increased feeding rates as food concentrations increased until cornpetition for food among tadpoles. Laboratory rneasurea maxirnum ingestion rate was attained. Sede et ai. (1982) ments of growth rates in Hyhgratiosa versus Hyiufemoralis speculated that the rnaximum ingestion rate was determined indicate that one species rnay decrease the growth rate in bv gut packmg or by saturation of the gili iiters. The low another and that this effect is variable among sibships (Trafod concentrations (below about 0.1 mg/liter) at which vis 1980b). Although Steinwascher (1978b) detected both tadpoles cease to feed is presumed to reflect an autoregula- chemical interference and behavioral exvloitation comvetitory rnechanisrn that avoids energy loss during atternpts to tion in laboratory experiients with Rana sphenocephaiu, beharvest scarce resources. Feeding cessation aiso provides a haviors associated with exploitation competition have not concentration refinium for food iterns so that s&ce items been rigorously examined- Larger tadpoies probably push are not elirninated. Work with a population ofRana catesbei- smaiier tadpoles away frorn food patches, aithough this a m in a natural pond showed that phytoplankton popula- would most likely occur only with particulate food sources aons responded reasonably weii and their density stabilized and at high tadpole densities relative to food availability at the concentrations that were predicted by the laboratory (Travis 1980b; Werner 1994). Also, a tadpole of one feeding eqxriments (i.e., d e r aigai bloorns, the concentration grad- rnode (e.g., rasps and swims actively) rnay physicaiiy interually decreased and stabilized at about the refigium concen- fere with a tadpole of another feedmg type (e.g., suspension uation predicted in the laboratory; Sede 1980). Viertel feeder rernains quiescent). Chemical interference rnay have ( 1990, 1991) showed that tadpoles at early stages of devel- no obvious behavioral correlates beyond preferential assoopment rnay have lower threshold concentrations (the low- ciations or avoidance (see Social Behavior below). R. A. est concentration at which feedmg starts) than older tad- Griffiths (1991) noted that cornpetition rnediated by nutripoles. ent depression rnay occur between species in the absence of Farlowe (1928) assumed that tadpoles were indiscrimi- physical interactions. nate suspension feeders and used intestinal contents as an index of the aigai constituents of ponds. Jenssen (1967, Respiratwn Rana ciumitans) and Hendricks (1973, Ranapipiens) found Tadpoles respire through the skin, gills, and lungs (Respirathat foregut contents of tadpoles and water column constit- tory Physiology, chap. 8). Most work on tadpole respiratory uents were indistingushable. Tadpoles also graze on peri- behavior has focused on the rates of ventilation with differphyton and epiphytic aigae (Dickman 1968; see L. M. John- ent oxygen availabilities, development of lung use, partison 1991; Kamat 1962; and Sahu and Khare 1988 for other tioning air and water breathing, trade-offs between respirastudies of tadpole feeding habits), epibenthic algae (Caief tion and feeding, and the relationshp between respiration 1973), and their own fecai peiiets (Steinwascher 1978a). and locomotor performance (e.g., Watkins 1997). Tadpoles The issue of selection among these patchy food sources has that use both giils and lungs respond to changes in dissolved not been addressed, but C. L. Taylor et ai. (1995) showed oxygen (DO) concentration by varying the rate of air breaththat tadpoles can choose among foods of differint nutri- inn. To breath air. tadvoles characteristicaiivswim to the surtionai quaiities. face, quickly ili the buccai cavity with air, and rapidly swim During an examination of ingestion rates of five larvae down again ( = "bobbing" of Wassersug and Seibert 1975; (Bufo amhcanus, B. woodhousii fmieri, Pseudacris crucqer, S. Wong and Booth 1994). Bobbing rates are consistent Rana catesbeiana, and R. sylvatica), Seale and Beckvar (1980) with the state of lung development in tadpoles of Bufo woodassumed that ingestion rates were indicative of food prefer- housii, Pseudamis trt>meiLata, Rana pipiens; these forrns and ence. They demonstrated some differences among species showed a significant negative correlation between D O conthat were fed on single aigal cultures but found no differ- centration &d bobbinc rate once D O was below a criticai ences with R. catesbeiana when six species of green and blue- vaiue. The bobbing r z e of Spea bomb@ons was negatively green aigae were offered. In aii cases, tadpoles appeared to correlated with D O throughout the range of DO concentraadjust feeding rate to food volume or biornass (see Sede and tions used. Air bubbles t&n into the biccal cavity rnay pro\Vassersug 1979). Tadpoles of Pseudacris cmcifer had a much vide oxygen by absorption through the operculum, filter aphigher threshold concentration (= food concentration be- paratus, and waiis of the buccai cavity. The tadpoles of Bufo low whch feeding ceases) than did those of E. w. fowleri or woodhousii, essentiaiiy lungless untii rnetamorphosis, rnay R. catesbeiana when these species were fed a iiamentous not take atmosvheric air into their buccai cavities but mav blue-green alga (Anabaena spherica). These findings suggest come to the oxygen-rich surface water for longer periods of that the concentration range over which food is available time to take in oxygen through the skin (chap. 8). rnay W e r among tadpoles -and rnay afFect seleaion of food Feder (1983a) noted increases in lung ventiiation as a req e s (e.g., Anholt and Werner 1995). Studies of the micro- sponse to aquatic hypoxia in Rana beriundieri, and West and
Y Y
2

224

KARIN vS. HOFF, ANDREW. BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

Burggren (1983) found the converse; forced lung ventila- ature selection has not been identified (also see Behavioral tion per se (as weli as the resulting increase in PO,) contrib- Thermoregulation in chap. 8). There have been numerous uted to the suppression of gill ventilation in Rana catesbei- speculations on its adaptiveness, but there i5 no consensus ana, although the lungs contributed relatively little to gas that any factor(s) (e.g., developmental changes, energy exchange at early developmental stages. As tadpoles ap- metabolism, gas exchange, osmoregulation, predator avoidproach metamorphosis, the lungs increase in importance be- ance) adequately explains the observed behavioral temperacause of increased oxygen demands, and aquatic respiratory ture selection in any species. Most of the hypotheses rebehavior (gili ventilation rate) in Ranapipiens increased with viewed by Brattstrom (1979) remain untested. Lefcort and D O concentration in tadpoles at Gosner 4 1 (Wassersug and Eiger (1993) discussed how a fever caused by a pathogen Seibert 1975). Differential abilitv to use aerial and aauatic influenced tadpole behavior and may influente pathogen oxygen resulted in interspecific differences in locomotor transmission. Cupp (1980) found interspecific, geographical, and ontostamina (time to fatigue) under conditions of hypoxia (Wassersug and Feder 1983). Comparisons of lungless Bufo amer- genetic differences in the critical thermal maxima ( a m ; carolinensis, Rana catesbeiana, icanus versus R. berlandieri and Xenopus laevts (which use Bufo woohousii, Gasr~ophryne both lungs and gdls; see Pronych and Wassersug 1994) andR. sylvatica, but not in Pseudacris trrkeriata).Of those that T showed that air-breathing tadpoles could moderate the re- showed changes in the C , the highest temperature was duction in stamina se& in d species in hypoxic water. tolerated by tadpoles in Gosner 3 0 4 0 . Dupr and Petranka l Breathing air also reduces stamina because of increased (1985) noted similar ontogenetic trends in final thermal premem temperature selected by tadpoles in buoyancy (X. l m k , Wassersug and Feder 1983; Campeny ferendum (ETP; and Casinos 1989), increases exposure and risk of predation thermal gradients) of several tadpoles including PseudanZc (Feder 1984), and reduces growth rate because of the meta- triseriata; the ETP peaked sharply at premetamorphic stages bolic cost of surfacing to breathe (Pandian and Marian and did not track precisely the development of CT, noted by Cupp (1980) and Sherman (1980). Dupr and Petranka 1985~). Ontogenetic changes in buoyancy were described for five (1985) speculated that behavioral temperature selection respecies (Bgfo americanus, Hyh chrysosceli/versicolor;Pseuda- fleas changes in thermal sensitivity of cutaneous coldA s trise&zta macuhta, Rana pipiens, and R. sylaatica) at sensitive neurons (also see Crawshaw et al. 1992; Hamhatchling, larval, and early metamorphic stages by Gee and merschlag et al. 1989; Wolimuth and Crawshaw 1988; WolWaldick (1995). Hatchlings were negatively buoyant and lmuth et al. 1987). sessile; their lGgs were no; idated. L1 b u t ~pipiens were . Indeed, the many ontogenetic changes in the amphibian neutraily buoyant through most larval stages and active in nervous system (Kollros 1981) include changes in the temmidwater; R. piiens larvae were negatively buoyant and ac- perature at which static irnpulse frequency occurred in cutative or sedentarv near the bottom. Ail svecies were nega- neous cold-sensitive neurons (Ranacatesbeiana; Dupr et ai. " tively buoyant and benthic in metamorphic stages. Gee and 1986). Tadpoles at younger stages were not as selective of Waldick (1995) showed that the tadpole lung has a hydro- temperature as older tadpoles, even though developmenstatic role and some species were able to regulate precisely tal trends in static irnpulse frequency of individual neurons their buoyancy to compensate for environmental changes. and behavioral temperature selection were similar. Static irnCampeny and Casinos (1989) examined buoyancy in tad- pulse frequency usuaily occurred at somewhat lower temperapoles of the midwife toad, Alytes obstetricans (see Feder and tures than those that were behavioraily selected (Dupr et Wassersug 1984, Pronych and Wassersug 1994, and Snet- al. 1986), and the development of additional warm-sensitive kova et al. 1995). neurons may also mfluence temperature selection. Such neuSuspension-feeding tadpoles use the buccal pumping rons have not yet been identified. Other aquatic ectotherms mechanism for both food and oxygen uptake. Feder et al. (including sal&ander larvae) modulate temperame selec(1984b) found that tadpoles of Xenopus h i s increased tion in response to hypoxia (Dupr and Wood 1988), but buccopharyngeal respiration in hypoxic water at the expense this has not been examined in anuran larvae. of feeding when prevented from breathing air; food ingestion decreased even though buccal pumping rate increased, and tadpoles likely actively suppressed mucus secretion to Anuran larvae are not shaped like any other aquatic vertereduce food entrapment rate. Wassersug and Murphy brate. Tadpoles have globose bodies, no lateral h s , and a (1987) noted that aerial respiration promoted growth in ob- relatively long, muscular, flexible t d that comprises most of ligate suspension feeding Xenopus. Feder et al. (1984b) spec- the body length (Hoff and Wassersug 1985; Wassersug ulated that the evolution of air breathine in aauatic verte- 1989a; chap. 3). This shape suggests that tadpoles swim " brates occurred as the result of selection for the complete differently from more typical aquatic vertebrates, such as dedication of the buccopharyngeal surfaces to feeding. fishes, and often has led to the assumption that tadpole swimrning is relatively ineffective and uncontrolied. Recent studies h& shown ihat many aspects of the swimrning Although tadpoles ofien aggregate in warmer areas of ponds movements of tadpoles do not differ greatiy from those of and sometimes follow changing patterns of suniight (Bratt- typical subcarangiiform fishes, such as trout (Wassersug and strom 1962; Schley et al. 1997), the mechanism for temper- Hoff 1985).

BEHAVIOR

225

h the sense of large fishes that swim continuously (e.g.,

uma),steady swimming in tadpoles does not exist. Although

d p l e s have impressive endurance while swimming against a slon- current in a circulating flow tank (i.e., more than 90 min for a Bufo tadpole; Wassersug and Feder 1983), rep m e d observations of tadpole swimming generaliy faii into ra-o categories: rapid bursts to evade predators, or very slow svimn&g betwein patches of foodr i midwater ggren g t e s (see Social Interactions below). Occasionaiiy tadpoles mim for more than a few seconds without being actively pursued (e.g., aggregations of tadpoles following patches of sun across a pond; Brattstrom 1962, R. J. Wassersug, personai communication). Even in these cases, steady swimrning of undisturbed tadpoles is best characterized as a series of short swimming bouts. Dudley et ai. (1991) provided information on the effects of shape and limb formation on drag forces during swimming in Rana catesbeiana tadpoles, and Boothby and Roberts (1995) exarnined the effects of tactlle stimdation on escape swimming responses inXenopus laes hatchlings. Parichy and Kaplan (1995) examined the functionai consequences of developmentai plasticity on locomotor performance on hatchlings. Denver (1997) addressed the concept of hormonai changes caused by environmentai suess and their effects on metamorphosis. This form of plasticitv has seldom been considered when discussinp:variations in metamorphic timing and its effects on behavior such as locomotion. For this presentation, we have combined steady and periodic swi&ing and consider only swimming boits tha; last at least three locomotor cycles (tail beats) and show little acceleration or deceleration. Assisted by high-speed cinematography and computer-assisted frame-by-frame analyses, laboratory studies of swimrning have focused on quantitative descriptions of swimrning kinematics. These descriptions were compared to the swimming of other aquatic vertebrates (fishes: reviews by E W. Webb 1975, 1984a, b). Oxner et ai. (1993) presented a mathematicai model of body kinematics in swimming tadpoles. Tadpoles use lateral-undhation of the body to propel themselves through the water (Hoff and Wassersug 1985; Wassersug and Hoff 1985). Waves of bending initiated anteriorly are propagated to the end of the tail (Hoff 1986a) exactly as in other undulating aquatic animais (see E W. Webb 1975). For lentic bufonid and ranid tadpoles, a linear relationship exists between the frequency of the tail beat and the swi&ing velocity. When velcity measured i body n lengths per second, swimming of ali sizes of free-swimming tadpoles up to just before metamorphosis can be described by the same regression equation. Chovanec (1992) compared swimming velocity in Bombina bombina, Bufo bufo, Hyla arborea, and Rana dalmatina at early (26-28), middle (30-33) and late (39-40) Gosner stages. Species differed in velocity and activity; at intermediate &d late stages of development, Bufo and Hylu had high activity with low cruising speeds, especialiy Hylu, which showed the lowest cruising velocities. Rana usualiv swam in short bursts with high ve" locities, and Bombina showed/intermediate patterns. Chovanec (1992) evaiuated thesepatterns with regard to predation
U

by dragonfly naiads (see Sensory Abilities, chemical communication, later in this chapter). In the Wassersug and Hoff studies, the highest steady velocity was about 12 body lengths/sec. The maximuni amplitude of a tadpole tail beat (measured at the tip) increases graduaiiy as swimming speed increases to an asymptotic vaiue of about 0.25 body lengths. Tadpoles with relatively longer taiis (e.g., Rana clamitans or R. septentriomalir compared to B. americanus or Rama catesbeiana) had a relatively lower maximum amplitude, but ali amplitudes were well within the range of steady swimming recorded for many species of fishes. Because of the odd shape of pondtype tadpoles (i.e., heavdy weighted toward the front end as compared to many streamined fishes), a head wobble is apparent during s&ing. ~ u b s e ~ u e study (Hoff and nt Wassersug 1985) has shown that these standard kinematic measures that relate to efficiency and control (e.g., maximum amplitude of the traveling wave at ali points aiong the body) are similar among tadpoles of severai bufonid and ranid species and subcarangiiform (i.e., troutlike) fishes. Indeed, the apparent head wobble is in large part an iilusion created bv &e snout movinp: from side to Side. The vivot point for ihis motion is on $e sagittai line between thi otic capsules (Wassersug and Hoff 1985) and not more posteriorly at the center of the body mass. This suggests that tadpoles have some mechanism for adjusting that point of minimum amplitude (i.e., minimurn laterai displacement) and may invest considerable energy in maintaining stabiiity during swimming. It is reasonabk to expect that &e po&t of minimum laterai displacement would be in the vicinity of the semicircular canais so that equiiibriurn is not effected by the side-to-side movements of the head. Wassersup: (1992) showed that a tadpole can bend its body verticdy & &at upward and downward movements. The mechanicai efficiency of tadpole swimming, measured as the relationshiv btween tGl beat frequenq and swimming speed, f d s Aear the middle of the range fr ali fish species described (Hoff and Wassersug 1985). In a more t e l h e comvarison of the distance traveled ver tail beat among the "average" tadpole, the "average" fish, and ambystomatid saiamander larvae, Hoff et ai. (1989) showed that traveled farther ver tail beat than did saiaanuran tad~oles L mander larvae but not quite as far as the "average" fish. The great similarity between tadpoles and many fish species suggests that mechanicai design constraints on moving through water by laterai undulations are very strong. Although tadpoles do not cruise often or quite as well in ali regards as the best cruising fishes, the mechanicai similarities are more remarkable th& the differences. Tadpoles that maintain position in midwater (e.g., Xerwpus l&; Hoff and Wassersug 1986) present a different mechanicai issue. Some aspects of their swimming mechanics are consistent with the argument that maintenance of stability (lack of side-to-side movements) of the head is conducive to increased feeding efficiency in obligate suspension feeders. Up to about 6 body lengths/sec, Xenopus tadpoles do not use their entire tail for propulsion; they use only the flagellum at the end of the t while holding position (swimming d
V

226

KARIN vS. HOFF, ANDREW BI,AUSTEIN, ROY McDIARMID, RONALD ALTIG

down against their own buoyancy) and gradually recruit surprisingly high maximum speeds, it is clear that most tadgreater lengths of the tail to increase swimming speed. The poles and small fishes would stiii be consurned. Their speeds bending wave on the tail is not initiated at the base of the are inadequate when compared to the recorded velocities of tail until the tadpoles are swimming moderately rapidly (6 feeding events by relatively large, potential predators of tadbody lengths/sec); at these speeds the tadpoles probably poles (i.e., lunging and gape-and-suck feeding in several fish cease feedmg. Restricting lateral movement to the end of the species). The ability t o evade a predator Uely depends on tail. combined with a midventral keel-like extension of the the sensory capabilities of the predator and prey and the tail fin onto the body, probably contribute greatly to main- complexity of the habitat and avadability of cover, as w e l as taining anterior stability (Hoff and Wassersug 1986). At the locomotor abiiities of a tadpole. The neuromuscular control of iocomotion in tad~oies is higher swimming speeds, Bnopus tadpoles increase the tail beat frequency of the entire tail t o increase swimming speed not completely understood. Variations in the distribution of much like more benthic species (Wassersug and Hoff 1985); peripheral nerves in the tails ofRana andxenopw (K. Nishiat these speeds their heads have the characteristic tadpole kawa and Wassersug 1988, 1989) may relate to patterns of muscle recruitmentthat vary with swi&ing speed. Xenopw wobble. Burst swimrning used for predator evasion is another has a long spinal cord and many nerves branching (presummode of swimming. Feder ( 1 9 8 3 ~ ) reported that if Rana ably from different motor neurons) along its entire length berlandieri tad~ole; ~ursuedbv turtl& could extend the that apparently innervate individual (at ieast few) muscle chase (in som: cases Ifor 11 secJor more), the Uelihood of segments (see chap. 6). Rana tadpoles have a cauda equina their evading predation would increase. Those experiments branching (at the point where the spinal cord ends in the were conduaed in a large open pool with no cover for the adult) into large nerves traversing (and possibly innervating) tadpoles and may iustrate a threshold for the attention or many muscle segments. The patterns of innervation are corthe sensory abiiity of the turtie rather than any measure of related with the observed ability ofxnopus to initiate waves of to Increas- of bending anywhere along the tad as they increase the porthe ca~acitv anuran tad~oles evade ~redators. 1 , ing plant density and cover (= habitat heterogeneity) has tion of the tail used with increases in swimming speed, significant positive effeas on the survival of Bu@ tewestrtk in rather than increasing the frequency of the tail beat, as was the presence of visual-hunting odonate naiads and giant wa- shown for tadpoles of Buf and Rana (Hoff and Wassersug ter bugs (Babbitt and Jordan 1996; also see Bronmark and 1986). Rana tadpoles seem to use their tails in larger blocks Edenharnm 1994 and Stauffer and Semlitsch 1993 for be- or functional units (haifthe tail or the whole tail) and initiate havioral responses of tadpoles to fish). Tadpoles observed waves of bendmg only at the base or about at the middle of under more natural conditions (i.e., with cover sites or bot- the tail. tom substrate) make short dashes followed by periods of Waves of bending pass posteriorly as the animal moves immobility (Caldwell et al. 1980). During these predator- fonvard. It has been presumed that these traveling waves of avoiding dashes, tadpoles move with speeds comparable t o bending are accomp&ied by waves of muscle contraction fishes of the same size range (2-10 cm total length; W. (e.g., J. Gray 1936), but Blight (1976) suggested that at least Webb 1978). In studies of electrically induced startle re- three patterns of muscle contraction can produce travelmg sDonses. Hoff (1986b. 1988b) found that maximum veloc- waves of bending. Electromyographic studies of slow swimi b in a short dsh w i correlaied with the proportional and ming in several amphibians (Blight 1976, larval newts; A. absolute amounts of axial muscle arnong tadpoles (i.e., Bufo Roberts et ai. 1984, anuran hatchhngs; and Stehouwer and awricanus, Rana catesbeiana, R. clumita>s, R septentrionaks, Farel 1980, 1981, tadpoles) and other aquatic vertebrates and %nopus laePZr) that varied in siix and shape. Although (Grillner 1974, dogfish; Grillner and Kashin 1976, carp) the performances of size-matched tadpoles were broadly showed that ipsilateral axial muscles can contraa serially t o overlapping, the performance ranks correlated with the mus- ~ r o d u c e wave of muscle activitv that travels with the benda cle mass; Rana septentriunalis was the fastest tadpole with ing wave. Examination of other swimming activities (e.g., 26% of its mass as axial muscle andxnopw laevis ( 3 . 5 4 . 0 starting, stopping, very slow and very fast swirnming) in cm TL) was the slowest with 16% of its mass as axial muscle. Rana catesbeiana showed that traveling waves of bending can Most fishes have greater amounts of axial muscle compared be produced when contralateral muscles are acting simultato any tadpole exarnined (Hoff 1988b). This disparity sug- neously and also can be transmitted passively (i.e., without gests that muscle mass alone is not what permits compara- muscle aaivity in the posterior tail) if the wave is initiated tively good burst swimming performance; overall body anteriorlv. It is not clear whether this rance of locomotor " shape may be an important factor. The maximum absolute mechanisms is specific to Rana catesbeiana or to anurans, or velocity recorded (Hoff 1986a, 1988b) was 104 cm/sec for if it is a general feature of vertebrates that use undulatory a largetadpole, and the hghest relative velocity was 29 body swimming. lengths/sec for a small tadpole (about 2 cm long). The range L. Muntz (1975) described five stages in the behavior of in locomotor performance was very large for a11 species, and B n q w larvae relative to trunk myogenesis: nonmotile many individuais did not respond by swimming rapidly. (NieuwkoopFaber 20-22; myotomic muscles start to Maximum electrically inducible speeds may not be indicative differentiate), premotile (stage 22-24; first striated fibrils of the utility of speed alone to evade predators. Even at these visible, contractions possible), early flexure (stage 24-27; re-

BEHAVIOR

227

flexes and peripherai nerves present), early swimming (stage 28-33), and free swimming (stage 3 2 4 6 ; coordinated swimming possible, muscle c e ~ ut&ucleate). Muscle cells s did not become multinucleate until metamorphosis (Nieuwkooppaber 48-50). The behaviorai development of leg muscles was similar but, as expected, occurred at later stages: nonrnotile (Nieuwkoop/Faber 48-52; diierentiation minimal, nerves present), premotile (stage 53-54; limb trembles and cell striations forming), motile (stage 55-58; lirnb can make stepping movements, celis striated and multinucleate), and fully funaionai (stage 60-63). Hughes and Prestige (1967) studied the behavior of the hind limb during development. Some tadpoles move in ways other than by swirnming. Suctorial tadpoles occur in a number of families and have large oral discs (e.g., Aitig and Brodie 1972; Ascaphus tmei). The l a r ~ e disc aiiows the tad~oie move about and feed to on rock substrates in hgh-energy streams without loosing contact (C. L. Taylor and Aitig 1995; C. L. Taylor et ai. 1996). Having the belly modified into a sucker (e.g., Ranidae: Amoiups, HuUt, and Meristoflenys; Bufonidae: Atelophryniscus, Ateiupus, some Bufo) apparentiy aiiows mhabitation of faster currents (Altig and Johnston 1989), but it is the extension-retraaion c d e s of the oral disc that ~rovide the locomotor force. ~ i k suctoriai and other rheiphilous e tadpoles, those with abdominal suckers can move across the substrate by extension-retraction cycles of the oral disc (= oral h i t c h g of Altig and Brodie 1972). Aithough some stream tadpoles can swim powerfdly (e.g., Formas 1972; Telmatobufi australis),there are no tadpoles that continuaiiy occupy &e water column in fast floGing water like som fishes do. Some nonsuctoriai tadpoles that live in flowing water escape the current by hdmg within crevices, rocky substrtes, or leaf packs. These tadpoles often do not have the typicai morphlogy associated Gith stream inhabitants; they are often fusiform (= vermiform) with less robust tad muscles and low fins. It is &ely that tadpoles of this morphotype swim against currents ( ~ a s s e r s uand Heyer 1983 g and citations therein; R. Aitig, personai observation). Tadpoles of the family Centrolenidae are stream dweliers with fusiform bodes, long tads and low fins, poorly developed eyes, and highly vascularized and lightiy pigmented skl. Vdla and Vaierio (1982) reported coliecting specimens in soft, spongy earth of a stream bank, in moist earth beneath a plant 3 m from the stream, and in leaf packs in the s u e m . They caiied them "fossorial" and speculated that the long tails were usem in pushing the fusiform body (= burrowing) into mud and leaf packs. We have coliected centroleGd tad~oies severai S~ecies of and most often found them in dense leafpacks and decaying organic material aiong streams, sometimes above the waterline. The tadpole of Staurois nutator (Inger and Tan 1990; Inger and Wassersug 1990) is a morphologicaiiy similar burrowing tadpole from streams in the lowlands of Borneo. In addition to being strongly convergent on the centrolenid morphology, this tadpole has a large, uninflated lung (does not breathe air) and, like centrolenid tadpoles, probably relies heavily on cuV

taneous respiration; in this respect it differs from other elongate tadpoles. Other elongate tadpoles live in bromeliads and do very little swirnming (table 3 in Lannoo et ai. 1987). The tadpole of Osteopilusbrunneus has a tail to body ratio of 3.9 and probably is the most elongate tadpole known. Lannoo et ai. (1987) showed that 0. brunneus tadpoles swim very little, ineffectively (i.e., they have to beat their t& twice as fast as a Rana or Xenopus larva to achieve the same velocity), and with excessive lateral motion indicative of low kinematic efficiency. Three explanations had previously been offered to account for the very long tail of 0. brunneus tadpoles (and certain other arboreai, bromeliad dweliing forms as well): cutaneous res~iration. locomotion in a viscous medium. and an agitator to aerate the water. Lannoo et ai. (1987) reviewed each of these postulates and found them somewhat unsatisfactorv. As a result. thev offered a fourth: the tail of , 0 . brunneus serves as a static posturai organ to keep the head directed upward and nearer the water surface. They suggested that the lungs serve for buoyancy and assist in this . . behavior. ~ u l m o n G respiration is important to the tadpole of 0.bmnneus, and this posture saves the tadpoles energy by them not having to swim to the surface. Also a heads-up position could provide the longest tadpoles with an advantage in getting to eggs laid by femaies at the water surface (see previous oophagy discussion). Oldham (1977) described terrestrial locomotion in the tadpoles of two West African frogs, Chiromantis rufescens and Leptopelis hyloides (Hyperoliidae). Leptopelis deposits eggs in a "nest" dug up to 6 cm beneath the surface in moist, soil close to water; &ek the eggs hatch and the tadpoles wait for rain. When the nest area floods, frantic wiggling by the group of tadpoles creates a water-ied cavity. The roof of the cavity eventuaiiy coliapses and the tadpoles ccburrow"to the surface. The tadpoles have unusuaiiy long tads with many muscle bundies per linear unit and are able to reach water by wiggling like eels in very shaiiow water across the substrate (Oldham 1977). In contrast, Chiromantis ru@scens builds foam nests on vegetation, usuaiiy above water; when the nest disintegrates, tadpoles drop into the water and carry on as typicai pond larvae. Sometimes, the nest is not above water and the tadpoles have to make their way across wet soil to water. The tail of C. mfescens tadpoles is relatively shorter and less muscular. In a different scenario, CaidweU and Lopez (1989) reported that tadpoles of Leptodactylus mystaceus aiso make foam in their terrestrial nests to avoid dehydration. Under experimental conditions designed to duplicate hatching conditions, Leptopelis hyloides larvae traveled further and faster and aiways moved on their bellies while propeliing themselves with lateral undulations. Tadpoles of Chiromantis rufescens used less flexure in wiggling G d moved more rapidiy by landing on their sides and "fhpping" with tail movements much like a fish out of water, and some larvae of&ntiahtylus spp. also move effectively in this manner (R. Aitig, personai communication). Leptopelis larvae were able to survive and wiggle toward water longer than Chiromantisin the
i

228

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

same trial conditions. Both species showed some ability to mo&@ the rate of movement with changes in surface moisture and relative humidity, and .ptope1isshowed some ability to move directionally, perhaps through geotaxis. The sinuous wiggling locomotion of lcptopelis larvae was correlated with the longer, more muscular tail and seemed to work better under the conditions tested. Oldham (1977) suggested that the probability of reducing aquatic predation on the defenseless egg stage and having a relatively large and weliadvanced tadpole arrive in the pond constituted the survival value of terrestrial eggs. In these experiments, the tadpoles of Bufo muculatm used for comparison had no ability to move on land. Wager (1986) reported that tadpoles of a related species from South Africa, Leptopelis natalensis, afier hatch& and emerging, moved towaid water as a group. Another pattern of locomotion characteristic of larvae with long, low-finned tails, weii-developed eyes and hind limbs, cryptic coloration, and unusual mouthparts (see chap. 12) we cail "tail flipping." This type of movement is associated with tadpoles d e d "subaerial" by Wassersug and Heyer (1983) and "semiterrestrial" by Altig and Johnston (1989):In mbst other situations, we wouldgive priority to the earlier name, but for obvious reasons we use "semiterrestrial" here. These tadpoles use their tails to flip around on wet, exposed rocks in streams, in films and trickies of water flowing over rocks or from seepages, and on stems and leaves and between axils of plants. Wassersug and Heyer (1983) examined the tadpoles of Cychamphw and Thmopa (Leptodactylidae) and noted their similar morphologies (e.g., elongate and relatively finless tail, depressed body and branchial baskets, compressed and deep jaw sheaths). These tadpoles are obligate air breathers found on vertical rock surfaces where water flow is negligible. Some of these have a flattened abdomen (a possible precursor to a beliy sucker) presumably to increase surface area for cohesion and as an adjunct to lung ventilation. Drewes et ai. (1989) descnbed the tadpole ofArthroleptides martiensseni (Ranidae) and commented on the striking morphological and ecological convergence between their Arthrolepties tadpoles and those of Cychamphus (Heyer 1983a, b) and Thmopa (Cocroft and Heyer 1988).Arthlepties martiemeni tadpoles were coliected from vertical or steev faces of smooth rocks in the stream but outside the 1 main current; many tadpoles were in fissures. These authors noted the large eyes of these tadpoles and found that the hind legs of this species developed much more precociously of than those of tad~oles Bufi or Rana. When disturbed. tadpoles moved hkphazardly by tail flipping, sometimes foi a distance of 1-2 cm per flip. The obvious convereences on the semiterrestrial morphology in tadpoles from several families are notable. Tadpoles of Nannophys q k i s (Ranidae; Kirtisinghe 1958) also are morphologidy similar to those of Arthrohpdes, Cyclmamph, and Thmopa and occur in shallow sheets of water in seeps. The strikingly similar tadpoles of Indirana bedahwi were described by Annandale (1918) as skipping rapidly over damp rocks.

Fig. 9.6. Semiterresmal mdpole of PetmpcdttEsparkcri (Ranidae).Tadpoles were found climbing on slime-covered rock faces along a road cut through a forest in Cameroon, West Africa.

Blommers-Schlosser and Blornmers (1984) noted that tadpoles of the Malagasy frogs in the M.antzahq1wpulcher group can move from one axil of Pandanw to another by violent wiggling; they also have elongate, depressed bodies and low fins on a long tail. Lawson (1993) reviewed the natural history of the tadpoles of the Cameroon species ofPetropedetes (Ranidae). In general, they are elongate tadpoles with long tails, weli-developed hind legs that apparently develop early in the larval stages and small heads. Lawson (1993) noted that eggs are laid on the undersides of leaves on low shnibs and trees along rock and pebble banks of seasonal streams. On hatching, the tadpoles squirm along the branches for some time before dropping to the ground below. In his figure 19, Lawson (1993) showed tadpoles of i? cameronensis on stems and a leaf and a tadpole of i? parkeri with well-developed hind limbs and a long tail sitting on a leaf; Pparkevi tadpoles were found crawling on slimecovered, vertical rock faces covered by trickling water (fig. 9.6), and Petropedetes tadpoles were found at several localities in the Korup National Park actively moving on the forest floor hundreds of meters from the nearest water. Kunte (1998) observed ecologicaily similar tadpoles in India that were probably Indivana kithii. These tadpoles tried to escape the predatory strikes of the snake Ahaetulla nasuta by hopping into cracks and depressions in the surrounding rocks. Tadpoles of H o p l o p w r0pn-iand H. ulugu~nsis (Noble 1929) live in the damp crannies between leaves of wild bananas or within split bamboo stems and have a bilateral pair of fingerlike, fleshy projeaions on the ventrolateral margin of the body at the posterior end of the buccopharynx. These projections move upon contraction of the subhyoideus muscle (chap. 3), and Noble suggested they give an added "kick" to assist the tadpoles as they wiggle from one moist axil to another. It is interesting that the tail of a weliformed embryo is longer than the head and body and wraps across the ventral surface of the developing embryo; hind limb buds are clearly visible in these tadpoles before they hatch but external gills are not (Noble 1929). Recently hatched larvae of H. u l t g u e have weli-developed, inflated lungs that extend two-thirds the length of the body

BEHAVIOR

229

cavity; their size implies that pulmonary respiration is important for these tadpoles (Noble 1929). It is interesting to speculate about the origin and adaptive nature of patterns of taii-flippingJsemiterrestnal locomotion. Undoubtedly, the precursor of this behavior can be found in more typical tadpoles but little information is avaiiable. M. A. Smith (1916) noted that the carnivorous tadpoles of Occidozyda and Phlyno~iussus quietly for long perilie ods, seldom move with their taiis, and push themselves about with their precocial hind legs. Perhaps other examples are in the literature but have not been evaiuated in this regard. Our understanding of the evolutionary interplay between the breeding ecology of most frogs and the feeding and locomotion of their tadpoles is elementary. A cursory review of the literature on more specialized patterns of breeding (Dueiiman and Tmeb 1986), especially on frogs that deposit eggs in relatively small and potentially ephemerai habitats such as phytotelmata, suggests fniitfd areas of investigation. The convergent evolution of similar morphology (e.g., body shape, taii length, oral stmctures), physiology, locomotion, and feeding in different lmeages in similar habitats is strikmg. What are the trade-offs and patterns of convergentes between feeding and locomotion in these difrent habitats in whch tadpoles find themselves? For example, what is the evolutionary hstory of the independent trends toward increasing tail length, dependency on pulmonary respiration and development of semiterrestriai/tailflipping locomotion, and changes toward macrophagy as exemplified by carnivory, cannibalism, and oophagy? Surely, more tadpole watching wi provide insight and answers to some of these questions and bring other morphologicai and behavioral correlates to light; we encourage such investigative .efforts. There is considerable information on the stmcture and funaion of hatching glands (chap. 3), but few observations have been made across species to identi5 potential differences in the timing and actual hatching process and behavior of the embryo (e.g., Bradford and Seyrnour 1985; Goiiman and Goiiman 1993a; Kothbauer and Schenkel-Brunner 1981; Lutz 1944a; Noble 1926a; Yoshzaki and Yamasaki 1991). Thrashing about by embryos may enhance hatching of other embryos in a clutch of eggs with h g h water turgor. Physicai features like pH (e.g., Dunson and Connell 1982) and oxygen tension (e.g., Petranka et ai. 1983) also influence hatching; one of us (R. Aitig, personal observation) has seen several cases where the surface-film eggs of Hyla cinerea became encased in a mat of aigae in a particuiarly eutrophic pool. These embryos died &er reaching Gosner 25 because they could not escape the jelly membranes. Behavior of hatchiings is largely unknown, but specific activities that have not been studied must occur in those frogs that have larvai transport or some other form of larvai brooding. Hatchlings of some endotrophs (e.g., G. J. Ingram et ai. 1975, Assa; Brauer 1898, Sooglossus secheliensis) and exotrophs with larvai transport (e.g., De Prez et ai. 1992a, dendrobatids; Inger and Voris 1988, Inger et ai. 1986b, Limnonectes microdtrca) must respond by negative

geotaxis, pheromones, or a combination of these and other cues to arrive at the proper place in or on the parent.

Dispersion and Social Behavior

Habitat Selectwn Numerous physical (e.g., distance from shore, oxygen concentration, substrate qualities, water depth and flow rate, site duration, and temperature; discussed by O'Hara 1981) and biological (e.g., presence and distribution of vegetation, other tadpoles, other organisms, and the phenology of ali organisms) factors influence the spatial and temporal distribution of tadpoles among microhabitats. Tadpoles may select habitats because of attraction to or avoidance of conspecifics and predators. Temperature often influences the behavior and ecology of tadpoles (review by D u e h a n and Tmeb 1986; chaps. 8 and 10). Temperature influences tadpole aaivity patterns, growth and development, metabolic rate, and the timing of metamorphosis (Kohos 1961; Srnith-Gill and Berven 1979; Wilbur and Collins 1973). Rapid growth and development are especially important for species that live in ephemerai habitats, such as in deserts where pools dry up quickly or at high elevations where ponds may freeze (A. R. Blaustein 1988; Hokit and Blaustein 1995; O'Hara 1981; Skelly 1996). In nature, the gathering of tadpoles (e.g., Bufo, Hyh, Pseudamis, Rana, and Scaphwpus) in the warmest thermai gradients suggests selection of favorable temperatures for rapid growth (e.g., Ashby 1969; Beiswenger 1972; Bragg 1968; Brattstrom 1962,1963; C. C. Carpenter 1953; Muiially 1953; O'Hara 1981; Tevis 1966). This assumption is supported by experimental studies (e.g., Beiswenger 1972; Beiswenger and Test 1966; De Vlaming and Bury 1970; Herreid and Kinney 1967). Few experimental tests have demonstrated that tadpoles selectively respond to key features of the substrate. Aitig and Brodie (1972) showed that tadpoles of Ascaphus tmei preferred smooth rocks above 55 mm in diameter. J. A. Wiens (1970.1972) showed that Rana aurora and R. cascada tadGoles could be conditioned to prefer certain substrates, although the two species displayed opposite preferences in identical experimental regimes. Rana aurora tadpoles reared over featureless or square-patterned substrates showed no preference for striped or square-pattemed substrates, but larvae reared on striped substrates preferred striped-patterned habitats. The preference for striped substrates was established during the first 14-17 days in the striped habitat. T h s preference was retained &er isolation from the substrate and could be reestablished in both young and old tadpoles. Rana cmcadae tadpoles reared in featureless or striped-pattern environments showed no preference for either square-patterned or striped substrates, but tadpoles raised in a squarepatterned environment showed a significant preference for that regime. J. A. Wiens (1972) suggested that the different responses were because of Werences in the substrates of the haGitats in whch these tad~oles found. Rana aurora m i are cally breeds in shallow ponds prone to surnrner drying and
,L

230

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

in areas where permanent ponds have overflowed (Storm 1960; J. A. Wiens 1970). Branches, and stems of emergent plants that are characteristic of these habitats cast a pattern of linear shadows on the bottom. The strioes that larval R. aurora res~onded mav resemble the linear substrates and to shadows {ound in their natural habitat. Methods similar to those of J. A. Wiens (1970, 1972) were used to test habitat selection in I*.loulapukhra ( P u m 1976) and Bufo americanus, Rana clamitans, and R. sylvatzca (O'Hara 1974). I*.loula tadpoles reared in striped habitats preferred stripes, those reared in square-patternedhabitats preferred squares, and those reared in featureless habitats displayed no preferences. Punzo (1976) suggested that tadpoles imprint on the habitat in which they were reared and that they select this type of habitat in nature. In contrast to studies by Wiens and Punzo, the rearing regime had no influente on the habitat choices of B. americanus, R. clamitans, and R. sylvatica (O'Hara 1974). Results obtained for Rana cascadae tadpoles by J. A. Wiens (1972) and by O'Hara (1981) are contradictory. O'Hara (1981) tested the responses of tadpoles reared on a smooth tank bottom versus natural substrates (e.g., sand, gravel, and rock). Tadpoles of Bufo boreas, Pseudacris regila, and R. cascadae preferred fine-grained over coarse-grained substrates. Tadpoles from different populations and wildcaught tadpoleS versus those reared &e laboratory environment behaved similarly. Rana cascadae tadpoles were influenced by the rearing regime in the study by Wiens, whereas in O'Hara's exoeriments thev were not. One must be cautious about interpreting results of experiments concerning habitat selection when the environment is quite artificial. Comparing experiments that used different techniques is difficult; theseresdts may reveai real ecoiogical ~ e r e n c e s in nature (see O'Hara 1974, 1981, and J. A. Wiens 1970 for discussions). Tadpoles may use visual, tacule, and olfactory cues to dist i n p s h among substrate types in field and laboratory tests (O'Hara 1981). Waringer-Loschenkohl(1988)showed that the presence of vegetation, as weli as other species, caused a significant shift in the distribution of European tadpoles from the bottom of aquaria toward the water surface. Experimental results (Pfennig 1990a, b) suggested that tadpoles of Spea multiplicata select their habitat by using diet-based environmend cues; tadpoles reared on similar diets aggregated with one another; he also argued that larvae have increased growth and survival when they were restricted to their natal environment. 1.A. Haii et al. (1995) found that larvae of Spea intemzontana did not prefer only cues of their natal environment but were able to react preferentially to nove1 stimuh (larvae raised on different diets) in forming associations. Social Interactwns Many aggregations of larval anurans form as a result of social attraction toward conspecifics or because of abiotic parameters. Through field observations and experimentation in the laboratory and field, the various moda of tadpole aggrega-

tive behavior have been assessed and classified (e.g. Bragg 1965; Caldweli1989; Downie 1990a; Dueliman and Trueb 1986; Lescure 1968; Wassersug 1973). Analyses of the geometric struaure of tadpole aggregations (e.g., L. C. Katz et ai. 1981; Pote1 and Wassersug 1981; Wassersug et ai. 1981b), the sensory bases by which individuals maintain contact with conspecifics (e.g., A. R. Blaustein and O'Hara 1982a; Waldman 1985a; Wassersug et ai. 1981b) and the selective pressures that may have influenced aggregation formation (e.g., A. R. Blaustein 1988; Waldman and Adler 1979; Wassersug 1973) have been investigated and discussed. Studies of the roles of competition and predation in aquatic communities in which tadpole aggregations are a significant component have provided important insight into how aquatic communities are structured (e.g., Morin 1981; Travis 198C)b;Wdbur 1972,1984; Woodward 1982,1983; chap. 10). Several authors (e.g., Beiswenger 1975; Bragg 1965; Caldweli 1989; Wassersug 1973) categorized tadpole aggregations qualitatively; Bragg (1965) described tadpole aggregations as being asocial or social and suggested that asocial groups form through association with environmental features (e.g., food patches or temperature gradients) that are attractive to individuals. Social aggregations form because of the attraction of individuals to conspecifics. Because of the difficulty in distinguishing between social versus asocial factors, Wassersug (1973) suggested a broad, functional classification of two types: a simple, taxic aggregation in response to some physical feature (e.g., temperature, iight, current) was equivalent to the asocial aggregation of Bragg (1965), and a biosocial aggregation (= school) was equivalent to the social aggregate of Bragg (1965). Either type of aggregate may be polarized (tadpoles oriented in same direction) or nonpolarized. Tadpole schools are biosocial aggregates that display some synchronized movement patterns, but the degree of polarization within a school varies. Also, some tadpoles form slow-moving, polarized aggregations within which individuals may iie on the bottom of the body of water or be suspended motionless in midwater. Beiswenger (1975) also proposed a functional classification of tadpole aggregations focused on moving (streams and schools) and stationary (feeding and metamorphic) patterns. Because of the difiiculty in assessing whether aggregations observed in nature were simple or biosocial, M. S. Foster and McDiarmid (1982), O'Hara (1981), Wassersug (1973) and Wassersug and Hessler (1971) conducted a series of controlied laboratory experiments to test rigorously how and if tadpoles of various species were sociaiiy attracted to one another. These studies consisted of placing one tadpole of a particular species within each of four sections of a glass or plastic tray (M. S. Foster and McDiarmid 1982; Wassersug 1973; Wassersug and Hessler 1971) or an aquarium (O'Hara 1981). Each section was separated from the others by water-tight, transparent partitions so that tadpoles in each section had visual but not chemical contact with one another (see M. S. Foster and McDiarmid 1982 for results of chemical contact with the same design). Each section

BEHAVLOR

23 1

was marked into four equai quadrants (fig. 1 of Wassersug and Hessler 1971). If visually mediated sociai interactions among test individuais were iandom, then it was assurned that tadpoles would not associate in one particular quadrant over another. If tadpoles were positively attracted toward one another, they would be found most ofien in the quadrants where they could be as close to one another as possible. If tadpoles were avoiding one another, tadpoles would be found more often in the quadrant farthest from other individuais. The tadpoles of a; least 10 species have been tested with these methods with variable results (see M. S. Foster and McDiarmid 1982 for a criticai evaiuation of these methods). Tadpoles known to aggregate in nature generdy preferred to associate in the quadrant closest to conspecifics (e.g., O'Hara 1981, Wassersug 1973, Bufo borem; O'Hara 1981, Rana cascadue; Wassersug 1973, Rhinophynus hrsalis and Xenupus laevts; Wassersug and Hessler 1971, Xenupus laevts). Tests of species that probably only occasiondy form aggregations in nature, such as th r ~ i d Rana boilii, R s catesbeiana, and R. pipiens, displayed a random distribution among the quadrants (Wassersug 1973). There was one discrepancy between the results obtained by O'Hara (1981) and Wassersug (1973). Wassersug (1973) found Pseudacri's regillu tadpoles exhibited no biomutuai attraction whereas O'Hara (1981) showed that they were mutuay attracted to one another. I?regilla tadpoles form relativelfloose, intermittent aggregations in nature and their aggregating tendencies seem to differ between populations (A. R. Blaustein 1988; O'Hara 1981). The different testing apparati used by Wassersug and O'Hara rnay have accentuated the apparently discrepant results. Of the ranids tested, only Rana cascad~ tad~oiesdis~aveda dear-cut biosocial attraction toward co&pecifics.L~/nature,R. cascadue tadpoles are one of the most social of the ranids and form small, cohesive aggregations in ponds and lakes (A. R. Blaustein 1988; Hokit and Blaustein 1997; O'Hara 1981). Tadpoles of Spea do not display an attraction for conspecifics (Wassersug 1973) even though they form aggregations in nature (aiso see J. A. Haii et ai. 1995). As Wassersug (1973) pointed out, these tadpoles rnay require more than visuicues in order to form aggregations or, as exemplified by Pseudacri's regilla tadpoles, the tendency to form aggregation rnay vary among populations. Based upon laboratory experiments and field observations, Wassersug (1973) described two basic modes of tadpole group formation (= schooling): the Xenupus mode and the Bufo mode, with comments on a t h r d mode made up of schooling tadpoles that occur in ephemeral breeding sites. H e judged that tadpoles within different famies generally fit into one of these basic modes of schooling behavior. In theXenupus mode, schools consist of clusters of strongly polarized individuais in midwater. and individuals avoid contact with one another. The B U . mode is characterized by schools being polarized or not and hundreds to thousands of relativelv slow swimrninp individuals in freauent contact 1 that "appear as dense black mats in shallow water or near the
V

bottom" (Wassersug 1973:285). The ephemeral group was characterized by highly motile, strong swimmers that occur in very active, closely packed schools; these tadpoles are often ornnivorous in terms of food particle size and quality. Severai species of pelobatid frogs (e.g., Scaphwpw and Spea), Rhinophlynus hrsalis, Pyxictphalus dpersus, and many Australian species fit this mode. Many schooling species remain to be categorized. Tadpoles of Rana hechcheri (Altig and Christensen 1981) move about in imrnense schools of severai cohorts; tadpoles in front are feediig and those behind are moving and coming to the surface to breathe. A large school can be heard by the crackling noise caused by so many tadpoles surfacing. In laboratory tests, there was evidence of sociai facitation in feediig. Tadpoles of severai species of Leptodactylus aggregate in nonpolarized schools that extend from the bottom to the top of the water, and these tadpoles also are stimulated to follw sma waves made bv the'ir attendant mother. Likewise, the nonfeedmg tadpoes of Nectophyne afra foliow vibrations caused by movements of the attendant maie (Scheel1970). Tadpoles ofAubriasubsigiluta form tight football-shaped clusters that appear to roli across the bottom (Schi0tz 1963). Tadpoles of Pyxicephalus adspersus promptly move through channels dug by the attendant maies from drying parts of the habitat to remaining water rnay school with (Kok et ai. 1989). Tadpoles of P~xicephalus those of SchZrmademza (R. Altig, personal observation; aiso see R. A. Griffiths and Denton 1993). Caldweii (1989) suggested that it rnay be difficult to place tadpole schooling behavior in one of Wassersug's modes because of considerable variation within a s~ecies and potentiaily different modes displayed by other siecies. Because similarities in the behaviors that result in groups rnay occur in species that are not closely related, Caidweii (1989) suggested that these behaviors should be designated by types rather than by taxonomic names. According to her scheme, Type I behavior is iustrated by tadpoles that form loosely aggregated schools in shaow areas or on the bottom of ponds but rarely in midwater. These aggregations rnay or rnay not be polarized, and movement of the group is amoeboid and slow (e.g., Bufo and possibly some Leptodactylus). Type I1 behavior (e.g.,Xenopus, some species of Phyllomedusa and some microhylids), is characterized by weii-organized schools in midwater. and individuais are not in contact with one another. Movement is slow in these strongly polarized schools. Type I11 behavior (e.g., Hylageogaphica, some ranids and rhacophorids) involves weii-organized, polarized schools in the shape of a sphere; and tadpoles are aiways in physicai contact with one another. These schools are found throughout the water column, near the edge or on the bottom. Individuais continuouslv movinp; toward the center of " the school creates a roiiing effect. Comparable data on the structurai organization and sensory bases of fish schooling behavior (e.g., Partridge 1982; Partridge and Pitcher 1980; review by Pitcher 1986 and E. Shaw 1978) are not available for anuran larvae. Anaiysis of the geometric structure and sensory bases of schooling (L. C. Katz et ai. 1981; Wassersug et ai. 1981b) suggested

232

KARIN vS. H O F F , ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

that schooling behavior of Bufo woodhousii and Xenopus l m ? than individuals in the early detection and avoidance of tadpoles is similar to that of fish. In the absence of other predators (O'Hara 1981). Theoretical, observationd, and cues, both Bufo and Xenupus larvae exhibited two features experimental evidence suggest that as prey group-size inof fish schools -oriented paraliel to and showed preferential creases early predator deteaion and probability of escape inbearing to their nearest neighbors. In spite of these similari- creases (e.g., S. J. Arnold and Wassersug 1978; Hamilton ties, most tadpole schools do not exhibit the mobiiity dis- 1971; Kenward 1978; G. V. N. Powell1974; P d i a m 1973; played by fish schools. Compared to those of fish, rnost tad- Siegfried and Underhill 1975; R. J. Taylor 1976, 1979; pole schools are relatively stationary and individuals in Treisman 1975; Vine 1973). Tadpole groups rnay efficiently schools are usualiy at random distantes fi-om their nearest swamp potential competitors. O'Hara (1981) and Olson neighbors. A preferred distance to neighbors displayed by (1988) occasionaliy observed aggregations of western toad fish is rarely observed in tadpole schools (Wassersug et al. (Bufo boreas) tadpoles displacing s m d groups of Pacific tree1981b). The mobile schooling behavior of Hylageographh frog (Pseuducris r@Lla) tadpoles. Intraspecific competition is an exception to the generalization that tadpole schools are rnay be mediated by aggressive behavior between individual relatively stationary (Caldwei 1989); the same is true for tadpoles; apparent aggressive "butting" has been observed schools of Rhznophynus donalis (Stuart 1961; R. W. McDi- in Bufo americanus tadpoles (Beiswenger 1975; D.C. Smith armid, personal observation). Tadpoles of certain species, 1990 and referentes therein; Waldman 1982a). Groups of tadpoles rnay thermoregulate more efficiently such as Phyllomedusa vaillanti, rnay be evenly spaced when schooiing in midwater (Branch 1983; J. P. Caldwell, per- than individuals. An aggregation of black tadpoles rnay sonal cornmunication). Bufo and Xnupus tadpoles differ in efficiently absorb solar radiation that raises the temperature some aspects of their schooiing behavior. The bearing rela- of the surrounding water (Caldwei 1989; Guilford 1988; tionship of one individual changed with iiiumination inXen- Wassersug 1973).O'Hara (1981) repeatedly found temperaopus but not in Bufo. Aithough Bufo tadpoles in schools seem ture differentials of 2-3C inside and outside of B. boreas agto be generaliy weakly polarized (Breden et ai. 1982), they gregations; he ascribed these to solar heat absorption and polarize more extensively in the field than in the laboratory possibly metaboiic activity. Aiso, tadpoles rnay accrue inclusive fitness by associating with related individuals. The con(Wassersug et ai. 1981b) L i e some fishes (e.g., Radakov 1973), tadpoles i cept of inclusive fitness (Hamilton 1964a, b, c) predicts, ali n schools rnay orient toward individuals of similar sizes (e.g., else being equal, that individuals wiii cooperate with or aid Alford and Crump 1982; Branch 1983; Breden et al. 1982). related individuals over unrelated individuals. Aithough The magnitude of and the factors involved in size sorting not required, Hamilton suggested that an ability to discrimivaries arnong species. Phyllomedusa vaillanti larvae in Brazil nate between kin and nonkin (= km recognition) couid fa(Branch 1983) formed temporary schools of rnixed size cilitate this so-calied kin selection (Maynard Smith 1964) classes that eventualiy segregated into schools cornposed of process. different size classes. Unequal locomotor capabilities of size classes, social attraction of like-sized individuals, and prefer- &n Rewpitwn and &n Association ences for specific water depths rnay contribute to sorting by Tadpoles of many species are socialiy attracted toward and size in these schools (Branch 1983).Ranasphenocephala tad- form schools with conspecifics in nature, and tadpoles in poles of different size classes rnay iive in different habitats, schools rnay orient toward individuals of similar sizes. Aiso, and both interferente and exploitative competition rnay based on numerous laboratory experiments that controlled . contribute to intraspecific size sorting (Alford and Crump for body size, tadpoles of several species seemingly prefer to 1982). S m d R. sphenocephala tadpoles rnay be displaced by associate with relatives over unrelated individuals regardlarge ones. Aithough Bufo woodhousii larvae seem to associate less of body size (A. R. Blaustein 1988; A. R. Blaustein with individuals of the same size, the magnitude of size sort- et al. 1987a; A. R. Blaustein and Waldman 1992; Waldman ing is sma (Breden et al. 1982). Both large and sma indi- 1991; also see Cornell et al. 1989; Dawson 1982; O'Hara and Blaustein 1985; Waldman 1982a, 1986a). It seems vidual~ rnay be found near one another. Group living has numerous benefits, including the possi- iikely, therefore, that tadpole schools of certain species are bhty of avoiding, detecting, and deterring predators, en- composed primarily of related individuals. Investigators have used variations of two basic rnethods, hancing foraging efficiency, accming benefits in competitive interactions, and increasing the efficiency of thermoreg- laboratory choice tests or laboratory open-field tests, to dec ulation (Aiexander 1974; Bertram 1978; Hamilton 1971; termine Ghether or not a tad~oie k discriminate between tests e Pitcher 1986; Pulliam and Caraco 1984; G. C. Williams kin and nonkin. ~ ~ ~ r e ~ a t i o n c h o i cof tadpoles of the 1964). Tadpoles that form aggregations rnay gain many of western toad (Bufo boreas), Pacific treefrog (PseudacrW rethese benefits (A. R. Blaustein 1988; Brodie and For- gilla), and the red-legged (R. aurma), Cascades (Rana casmanowicz 1987; O'Hara 1981; Waldman 1982a; Wassersug cadae), and spotted (R.prehosa) fi-ogs have been conducted 1973). For example, stirring of the substrate by groups of in the laboratory (review by A. R. Blaustein 1988; A. R. toad tadpoles results in suspending food that larvae can filter Blaustein and Waldman 1992; A. R. Blaustein and W d s from the water (e.g., Beiswenger 1972, 1975; Wdbur 1995) after test animals were reared in one of four regimes 197%). Tadpoles that form groups couid be more efficient (summarized in A. R. Blaustein and O'Hara 1986a): (1) km

BEHAVIOR

233

oniy, (2) kin and nonkin (mixed-rearing regime), (3) in colors, equai numbers of the two groups were placed in a isolation from an early embryonic stage, and (4) nonlun laboratory pool simuitaneously, and the distantes from each oniy. Association preferences were measured by allowing tadpole to its nearest s i b h g and nearest nonsibling were retest individuais to associate near members of two stimuius corded. Waidman (1985a, 1986a) also used choice tests in a groups within a rectangular tank (A. R. Blaustein and Y-maze to assess iun recognition in B. americanus tadpoles. O'Hara 1986a). Stimuius compartments that were separated In some of these tests, tadpoles were simultaneously prefiom the main portion of the tank by a screen were placed sented with water flowing from m o containers, each holdon opposite ends of the tank, and various numbers of tad- ing members of Merent s i b h g groups. In other tests, tadpoles that varied in genetic affinities were placed in each poles were simuitaneously presented with water flowing compartment. In most tests, the time out of 20 min that from a container holding siblings and one holding dechloritadkles spent in the h& nearest one stimuius group was nated tap water. Dawson (1982) used both choice tests and compared with the time that wouid be expected if associa- open-field tests to investigate lun recognition in B. ameritions were random. In other experiments, the time that tad- canus tadpoles. Like in experiments of Corneil et ai. (1989), ~ o l e sDent in the thrd of the tank closest to one stimuius the test apparatus used by Dawson was much smaller than s group, the third closest to the other stimuius group, or the the one used by A. R. Blaustein and O'Hara. The methods time spent in the middie third of the tank was compared and rearing regimes used by Dawson (1982) differed somenlth a random expectation. Resuits obtained by A. R. what fiom those used by A. R. Blaustein and O'Hara and Blaustein and O'Hara are difficuit to compare directly with Waldman, so again direct comparisons are difficult. similar tests of iun recognition done by Cornell et ai. (1989, Significant Merences in how tadpoles discriminated beR. ylvaticu) and Fishwild et ai. (1990, Pseudmris mcifer, R tween kin and nonlun exist among species (table 9.1), and pipns, and R. ylvatica) because rearing and testing proto- the different methods used by the various investigators cancols were different. not fuiiy explain the observed differences. DifTerences in disWaldman (1981,1984,1985b) and Waidman and Adier crimination between species were apparent even when iden(1979) conducted open-field laboratory experiments to test ticai rearing and testing regimes were used. For example, the kin recognition in B. americanus and R. sylvatica tadpoles three ranids tested and reared identically by A. R. Blaustein (alsosee Rautio et ai. 1991).After being reared under a vari- and his coiieagues displayed signhcant differences in how ety of regimes, members of two sibships were dyed different they discrirninated between iun and nonkin (table 9.1).
Table 9.1 A summary of the results of kin recognition experiments involving various anuran tadpoles. ENC = experiment not conducted. Discriminates Siblings Frorn Nonsiblings Yess Yes No No Early stages oniy Yesa No No Yes Identifies Half-Siblings Yesh Yesc ENC ENC ENC Yesd ENC ENC Yes' Dawson 1982; Waldman 1981 A. R. Blaustein et al. 1990; O'Hara and Blaustein 1982 Fishwid et al. 1990 O'Hara and Blaustein 1988

Species Bufonidae Bu@ americanus B. bweas Hylidae Pseuhris mtcijr l?r@ih Ranidae Rana aurora R. cuscadae
R. pipieru

A. R. Blaustein and O'Hara 1986b A. R. Blaustein and O'Hara 1981, 1982a Fishwid et al. 1990; O'Hara and Blaustein 1981 O'Hara and Blaustein 1988 Corneii et al. 1989; Fishwild et al. 1990; Waldman 1984

aCorroboratedin field experiments (O'Hara and Blaustein 1985; Waldman 1982a). bPreferencefor fuii siblings over paternal half-siblings; maternal half-siblings not distinguished from fi d sibiings (oniy half-sibling categories tested). <Aiicategories of fuii and half-siblings have been tested, but results indicate oniy preferences for f d sibiings over paternal half-siblings and for maternal half-siblings over nonsiblings. dCan discriminate between aii categories of fuii and haif-siblings and between haif-siblings and nonsiblings (details in text). =Preference paternal haif-siblings over nonsiblings; experirnents not conducted on other haif-sibling for categories.

234

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

Rana cascadae discriminated between kin and nonkin after exert a stronger influence than paternal cues. Rana c a s c h being reared under a variety of regimes. Rana aurora tad- tadpoles showed a positive attraction to kin rather than a poles discriminated oniy between kin and nonkin in early repuision to nonkin (A. R. Blaustein and O'Hara 1987). larva1 stages and oniy after rearing with fuli siblings. Rana Waldman (1984) conduaed open-field laboratory tests pretiosa tadpoles failed to discriminate between kin and non- of Rana sylvatica tadpoles that had been reared with sibhgs kin regardless of how they were reared (A. R. Blaustein or in mixed-rearing regimes. Tadpoles reared oniy with sib1988). Sidarly, R. sylvatica tested by Fishwild et al. (1990) lings displayed a smaiier, mean nearest-neighbor distance to discriminated between kin and nonkin whereas Pseudami3 siblings than to nonsiblings. The importante of famiiiarity is cmcifer and R. pipiens, tested and reared identicaiiy to R. 8- dificuit to interpret from these experiments. Mixed-reared vatica, did not. Using choice tests and a laboratory apparatw tadpoles distinguished between f a d a r siblings and familiar slightiy modhed from that used by O'Hara and Blaustein nonsiblings, but individuais reared in mixed regimes had (1981), J. A. Haii et ai. (1995) investigated the ability of closer contaa with siblings than nonsiblings; this couid have Spea intemntana tadpoles to discriminate between kin and influenced their preferences (Waldman 1984). The tadpoles diet-based (= environmental) cues in forming associations. showed a tendency to associate more closely with familiar They found no evidence of kin-based associations for indi- siblings than with unfimdiar sibhgs. Corneli et al. (1989) viduais raised on the same diet, and mixed resuits between showed that R. sylvatica did not discriminate between familkin and nonkin groups raised on different diets. They con- iar and unfamiliar siblinzs but did discriminate between uncluded that if kin discrimination occurs, environmentallly familiar paternal half-sivblings and unfamiliar nonsiblings. derived cues rnay be the basis for such discrimination and These resuits suggest that neither maternal cues nor prior that those cues rnay be external to the natal environment of association with individuals are necessary for recognition in R. sylaatica tadpoles. the tadpoles. Rana c m c h and R. sylaatica seem to have the most acute The discrimination system ofRana aurora tadpoles differs kin recognition abiiities. The rearing regime had little i d u - significantiy from those iiiustrated by R. c a s c h and R. sylence on the discrimination abiiities of R. c a s c h tadpoles. vatica. Oniy those R. aurora tadpoles reared with siblings Results of testing mixed-reared tadpoles show that individu- and tested in early developmental stages associated with kin als preferred to associate with unfamiliar sibluigs over un- (A. R. Blaustein and O'Hara 1986b; A. R. Blaustein et al. familiar nonsibhgs (O'Hara and Blaustein 1981).They also 1993). Ths tendency disappeared as tadpoles developed. preferred to be nearest an unfamiliar stimulus group comThe larvae of Bufo americanus and B. boreas showed both posed of sibhgs oniy, over a familiar group containing 50% sidarities and differences to each other and to the ranids. siblings and 50% nonsiblings (O'Hara and Blaustein 1981). The ontogeny ofB. boreas kin recognition was influenced by Also, tadpoles reared in isolation subsequentiy associated the rearing regime; tadpoles reared with kin preferentidy preferentidy nearest unfamiiiar siblings over unfamiliar associated with kin over nonkin in laboratory choice tests nonsibhgs (A. R. Blaustein and O'Hara 1981, 1982). (O'Hara and Blaustein 1982). but individuais reared in ,, These resuits suggest that familiarity with other individuals mixed-rearing regimes and with nonkm oniy displayed ranis not required for kin recognition, but siblings reared in dom association with respea to siblings and nonsiblings. mixed-rearing regimes failed to discriminate between famil- Even when preferences for siblings were fuiiy established iar sibiings and familiar nonsiblings. Tadpoles reared with after prolonged rearing with siblings, short-term exposure siblings could assess the relative composition of sibling to nonsiblings nulliied these preferences (O'Hara and Blaugroups to some degree; individuais discriminated between a stein 1982). stimulus group composed of 50% kin and 50% nonkm and The use of methods and sample sizes identical to those one containing 100% nonsiblings by preferring to associate used by A. R. Blaustein and O'Hara (1982a, Rana cas&) nearest the former group (A. R. Blaustein and O'Hara in their investigation of half-sibling recognition in Bufi bmem tadpoles (A. R. Blaustein et ai. 1990) makes comparisons 1983). Embryos rnay respond to cues that emanate from the jeliy between these studies especiaiiy meaningii. Bufo bmem tadderived from the mother's oviduct (A. R. Blaustein and poles preferentiaiiy associated with fuii siblings over paternal tad- half-siblings and with maternal half-siblings over nonsiblings O'Hara 1982a; Waldman 1981).Tests ofRana cas& poles that were reared after the jeiiy had been removed or (A. R. Blaustein et al. 1990). They did not discriminate beafter the jeliy had been removed and replaced with jeliy from tween fuii siblings and maternal half-siblings or between paa nonkin egg mass indicate that tadpoles were unaffeaed by ternal half-siblings over nonsiblings. These results suggest these manipulations (A. R. Blaustein and O'Hara 1982a). that a maternal component is necessary for discrimination, Rana c a s c h tadpoles preferred to associate with fuii sib- which contrasts with the data onRana cascadae and R. sylvatlings over either maternal or paternal half-siblings and with ica tadpoles. Unlike R. cmcadae tadpoles, Bufo boreas tadpoles half-siblings (either maternal or paternal) over nonsiblings did not discriminate between maternal and I oaternal half(A. R. Blaustein and O'Hara 1982a). Because fuli siblings siblings. The role of maternal cues in B. borem kin recogniwere chosen over maternal half-siblings, and paternal half- tion rnay be relatively complex (see A. R. Blaustein et al. siblings were chosen over nonsibhgs, maternal cues are not 1990). Bufo americanus tadpoles reared oniy with siblings or in necessary for kin recognition. Because maternal half-sibhgs were chosen over paternal half-siblings, maternal cues may isolation associated significantly nearer their siblings than

BEHAVIOR

235

nonsibiings (Waldman 1981; Waidman and Adler 1979). the laboratory rnay not have been motivated to display disTadpoles reared with sibiings in early development and then crimination behavior (A. R. Blaustein et ai. 198%) iun recexposed to siblings and nonsiblings in later development re- ognition rnay be m k e s t e d differentiy in differeit' popuiatained their dose association with sibhgs. However, tad- tions, rnay be a polyrnorphic trait within popuiations (A. R poles that were reared with both siblings and nonsibiings Blaustein 1988; A. R. Blaustein et ai. 1987b), and rnay in early development and exposed to sibhgs only in later change ontogeneticaiiy (A. R. Blaustein and O'Hara 1986b). development did not preferentiaiiy associate with familiar Studies of Bufo americanus (Dawson 1982; Waidman sibiings over familiar nonsiblings (Waidman 1981). This 1985a), B. borem, Rana aurora (A. R. Blaustein and O'Hara, suggests the possibility of a sensitive period in early develop- unpubiished data) and Rana mcadae (A. R. Blaustein and ment during which B. americanus tadpoles familiarize them- O'Hara 1982b) illustrate that h recognition in tadpoles is selves with other individuais (Waldman 1981). Maternal mediated by water-borne chemical cues. Kin recognition cues seem to be important in B. americanus h recognition persists after metamorphosis in at least R. cmcadue (A. R. because maternal haif-sibhgs were not distinguished from Blaustein et ai. 1984) and R. sylvatica (Cornell et ai. 1989; fii sibhgs, yet paternal h*-siblings were distinguished Waidman 1989); and based on the experimental regimes from fii sibhgs (Waldman 1981). used, it does not seem as if water-bome chemical cues were In laborato6 Y-maze experiments, Bufo americanus tad- used by metamorphs to discriminate between sibiings and poles preferred unfamiiiar sibhgs rather than familiar non- nonsiblings. siblings (Waldman 1986a). These data provide further eviPfennig (1990b) suggested that kin association in spadedenc;thit familiaritv is not the sole basi; for h recoenition foot toad tadvoles resuited from habitat selection based on " in B. americanus. In other Y-maze tests, tadpoles chose water diet-based environmental cues rather than from social preferl a c h g conspecific chemical cues over water containing non- ences. Tadpoles with similar diets aggregate in the laborasiblings (Waldman 1985a). Thus, avoidance of nonsiblings, tory. In nature, because of the spatial and temporal proximrather than attraction to siblings, codd be the basis for kin ity of siblings within a pond, individuals eating similar foods association in B. americanus tadpoles. Tests of B. americanus are more inciined to be related and therefore rnay form agkin recognition (Dawson 1982) showed that tadpoles gener- gregations. Work by Pfennig impiies that "h recognition" aiiy associated with siblings over nonsibhgs in choice tests behavior and lun association in tadpoles of some species rnay but f d e d to do so in open-field tests. be an artifact of habitat selection (also see arguments by J. A. Resuits of laboratory h recognition tests of similar Hail et al. 1995). O'Hara and Blaustein (1982) posed a simimethodologies with tadpoles of Bufo americanus (Waldman lar argument concerning habitat selection for kin association 1982a) and Rana cmcadue (O'Hara and Blaustein 1985) in B u . tadpoles, and Grafen (1990) suggested that aii cases have been corroborated in field experiments. Larvae of R. of vertebrate kin recognition can be more consistentiy excmcadae from six popuiations were reared in pure sibiing plained as species recognition. A. R. Blaustein et al. (1990) groups or in mixed-rearing regimes. Equal nurnbers of tad- addressed this point in detail; and for some species of anuran peles fiom two unrelated sibships were dyed diierent colors larvae, the proposal by Grafen (1990) rnay be correct (A. R and released together in ponds they naturaiiy inhabit. For Blaustein et al. 1990). several days, single aggregations were censused for sibship A. R. Blaustein and O'Hara (1986b) suggested at least composition. Aggregations consisted of tadpoles primarily two important parameters associated with the tendency for from one of the two color groups (controls revealed that a tadpole to display kin recognition behavior: dispersal charcolor did not d e c t aggregation behavior). Tadpoles raised acteristics and aggregation behavior. These parameters relate in pure sibiing and in mixed-rearing regimes associated in to the probabhty of whether a tadpole wiii have the opporgroups composed principaiiy of sibiings. Unhke the case of tunity in nature to interact with h . Species whose larvae laboratory tests, familiar sibiings were distinguished from fa- randomly disperse (see Augert and Joly 1994) from their miliar nonsibhgs. oviposition sites wouid have a relatively low probability of Resuits of 6eld exveriments on Bufo americanus tad- interacting with relatives; A. R. Blaustein and O'Hara poles generdy corrobo>ate the laborato& resuits (Waldman (1986b) predicted that larvae of such species wouid not ex1982a). As with Rana m , B. americanus tadpoles couid hibit kin-selected behaviors. Similarly, in species whose lar& discriminate between f a d a r siblinns and familiar nonsib- vae do not sociaily interact (form coheske aggregations) iings in the field, but they couid notudo so in the laboratory with conspecifics, h-selected behaviors rnay have iittle op(Waldman 1982a). These data suggest that field experiments portunity to evolve. In nature, the patterns of aggregation formation and disprobably are a more sensitive measure of kin recognition persal characteristics M e r among the anuran larvae tested than are laboratory experiments (A. R Blaustein 1988). Two ranids (Fishwild et al. 1990, Rana pipkns; O'Hara for kin recognition. Bufo americanus, B. borem, and Rana cmand Blaustein 1988, R. pretwsa) and two hylids (Fishwild cadae form persistent, compact, social aggregations (A. R. et al. 1990, Pseudacris mci$er; O'Hara and Blaustein 1988, Blaustein 1988). Rana cascadue aggregations are smaii and P s e u h s regilu) f d e d to discriminate between kin and usuaiiy composed of a number (less than 100) much smaiier nonkin in laboratory choice tests. These resuits do not ne- than the clutch size (A. R. Blaustein 1988). Rana sylaatua gate the possibility that these species rnay discriminate kin tadpoles rnay aggregate similarly in certain ponds but not from nonkin under different conditions. Tadpoles tested in in others (DeBenedictis 1974; Hassinger 1972; Waldman
" ,

236

KARIN vS. HOFF, ANDREW' BLAUSTEIN, ROY McDIARMID, RONALD ALTIG

1984), and their aggregations are usudy composed of hundreds of thousands of individuals from numerous sibships (Waidman 1984). Although little is known about the larvai ecology of Rana aurora, there is evidence that intermittent, temporary, aggregations rnay form in early larval stages (A. R. Blaustein, unpublished data; Caief 1973). Ranapretiosa aggregations probably form only sporadicdy (C. C. Carpenter 1953; R. K. O'Hara and Blaustein, unpublished data). Pseudacris regilha tadpoles occasiondy form s m d , loose, aggregations (O'Hara 1981); tadpoles of Pseuducris crtlcijir and Ranapipzens are not known to aggregate (Fishwild et ai. 1990). Bufo aggregations are generdy much larger than aggregations of Rana tadpoles, are often extremely dense (O'Hara and Blaustein 1982; Waldrnan 1982a; Wassersug 1973), and rnay be composed of hundreds of thousands of individuals from numerous sibships. S m d group size and low dispersal from sites of oviposition make it likely that aggregations of Rana cas& tadpoles are composed primarily of h.Because of these behaviors, R. cascadae tadpoles appear to have sufficient opportunity to interact with siblings and display a kin recognition system that would be efficient in their natural habitats (A. R. Blaustein et ai. 1987%; O'Hara 1981; O'Hara and Blaustein 1985). Potential exists for mixing with nonsiblings, especidy because egg masses may be laid commundy (A. R. Blaustein 1988; O'Hara and Blaustein 1981). If there are benefits to aggregating with kin, tadpoles of R. cascadae must be able to discriminate benveen siblings and nonsiblings; rhis abiiity must be resistant to modification after exposure to nonsiblings. Laboratory and field experiments indicate that the kin recognition system of R. cas& is established early in development and is generdy not altered by exposure to n o n h (A. R. Blaustein and O'Hara 1981, 1982a, b, 1983; O'Hara and Blaustein 1981, 1985). In certain ponds Rana sylvatica tadpoles rnay disperse rapidly and far from sites of oviposition, whereas in other ponds dispersai rnay be moreAlimitedand tadpoles mayavoid one another (DeBenedictis 1974; Hassinger 1972; Waidman 1984). In ponds where the tadpoles aggregate, there is some opportunity for the formation of kin groups. At times, large, single, polarized, schools break up into smder, feeding groups of severai hundred to severai thousand individuals (Waidman 1984). Althoueh most schools of the tad~oles of R. sylvatica conskt of in&iduals from many sibshiis, these smder units rnay represent sibling groups (Waldman 1984). Rana syloatica tadpoles display a kin recognition system that generdy is not influenced by exposure to nonsiblings. As in R. c s , a& a system that is resistant to modification after exDosure to nonsiblin~swould be reauired because of comrnunai egg laying and exposure to nonlun during larval stages (Waidman 1982b, 1984). Dispersal from sites of oviposition in both Bufo americanus &d B. boreas tadpoles is Ggh and members of different sibships intermingle (Beiswenger 1972; O'Hara 1981; O'Hara and Blaustein 1982; Waidman 1982b). The kin recognition system of B. americanus is generdy consistent with
I U I

its larvai ecology and behavior. Large aggregations rnay subdivide into smder schools (Beiswenger 1972, 1975; Waidman 1982a) that rnay represent sibling cohorts (Waidman 1982a). MWng with nonsiblings is frequent in active larvae (Beiswenger 1972; Waldman 1982a), and sibling preferences, once established, are resistant to modifications (Waidman 1981). The larvai behavior of B. boreas tadpoles may preclude kin association; extremely large aggregations form from individuals from numerous clutches, and the feeding behavior of the individuals in aggregation entails frequent moving, intermingling, and splitting o E The formation of discrete kin associations rnay be difficult even if such behavior conferred fitness benefits (O'Hara and Blaustein 1982). Maintaining cohesive sibling groups would be difficult even if tadpoles retained a capability to recognize siblings after encounters with nonsibiings. As in B. americanus, smder schools split off from the main aggregation, and these rnay be sibiing cohorts. The failure of Pseuducrir crtlcijir, E rgilla, Rana pipiens, and R. pretwsa to discriminate benveen h and nonkm is genera& consistent with their larvai ecology (discussed by A. R. Blaustein 1988). Because these species either do not aggregate or form aggregations only interrnittently, there is lide opportunity for siblings to interact. Too lide is known about the larvai ecology of R. aurora to speculate whether there is opportunity for kin to interact (A. R. Blaustein and O'Hara 1986b). These larvae are the onlv ones known to have a kin recognition system that changes ontogeneticdy. Ir is not necessary for a tadpole to aggregate with kin to gain the benefits of group living, but those individuals that preferentidy associate with kin rnay accrue additional benefits through an increase in inclusive fitness (Aiexander 1974; Hamilton 1964a, b). Kin recognition potentiay enhances . the formation of and maintenkce of kin e r o u ~ i which is important i there are benefits to associati& w i x h.Tadf poles living in groups composed primarily of kin could warn relatives if they release an aiarm substance when they are attacked (Hews and Blaustein 1985; Hrbek 1950; Waidman 1986b). This is a more likely benefit of kin association in R. cas& tad~oles(Hews and Blaustein 1985). Iniured R. cas& tadpoles release a chemical cue that causes others to elicit an escapelike response (Hews and Blaustein 1985). In t h s situation, kin would be warned because individuals in R. cascadae aggregations are presumably mostly kin, and kin recognition may be maintained by kin selection. Unpaiatabiiity and noxiousness are probably important antipredator adaptations in some species of tadpoles (Brodie et ai. 1978; Formanowicz and Brodie 1982; Wassersug 1973; Werschkul and Christensen 1977). The evolution of distasteh e s s or aposematic coloration has often been explained by a kin selection model (e.g., Benson 1971; Fisher 1930; Harvey and Greenwood 1978).If members of a single kin group aggregate with each other and a predator that sarnples one or more distastefd sibhes learns to avoid others in the group, then a gene for distastehess could increase in frequency through kin selection. This scenario has been invoked to explain the advantage of associating with kin
I

BEHAVIOR

in toad tadpoles (Waldman and Adler 1979; Wassersug Sensory Abilities and Learning 1973). Social behavior influentes growth and development. For Anurans frequently are used as model organisms for investiesample, in some insects, larval growth and development is gations of spinal reflexes, and considerable literature exists enhanced in groups composed of certain proportions and on behaviors with simple triggers (e.g., visuaiiy eiicited fight combinations of given genotypes (Bhalla and Sokal 1964; and feeding; see review by Ingle 1976). Comparatively lide Lewontin 1955; Sokal and Huber 1963; Sokal and Karten is known about the sensory capabilities of tadpoles (chap. 6) 1964). Social behavior rnay be especially important in the and how these relate to specific behaviors. This deficiency growth and development of tadpoles. Those that develop at simply rnay be because tadpoles are aquatic and small, and high elevations where the growing season is short or where thereby somewhat less traaable for many types of manipulahabitats are ephemeral may be under intense selection for tion. Work to date suggests that the sensory mechanisms of rapid growth. Some data suggest that tadpole growth and tadpoles differ from adults in ways that are consistent with development is influenced by the proportion of related indi- the large ecological Merences between the two life stages. The sensory abilities, especially pertaining to vision, of vidual~ unrelated individuals within a population or group to (Hokit and Blaustein 1994, 1997; Jasienski 1988; Shvarts tadpoles and their ability to mod* their behavior by learnand Pyastolova 1970; D.C. Smith 1990; Waldman 1986b). ing is a poorly understood field. How much they use vision Interferente competition has been documented in anuran during various behaviors (e.g., escape and feeding) is not larvae (chap. 10), and kin recognition rnay allow tadpoles to known. Aiso, oniy speculations can be made on the effects direct intraspecific competition toward nonkin when re- that the variations in eye positions (i.e., dorsal vs. lateral; see sources are iimited. Kin recognition could also direct canni- chaps. 3 and 12) and the amount of corneal and lenticular balism toward nonkin or, in some cases, toward kin (A. R. protrusion has on the vision field. The large, laterally placed Blaustein et al. 198%; Pfennig et al. 1993; Walls and Blau- eyes with curved corneas and prominent lenticular protrustein 1995). Because kin recognition persists for at least sion of a nektonic hylid tadpole must afFord visual abiiities some time after metamorphosis (A. R. Blaustein et ai. 1984; that difer greatly from the small, dorsal eyes with more flatCorneii et al. 1989; Waldman 1989), kin recognition rnay tened corneas and iittie lenticular ~rotrusionof a bufonid also play a role in adult anurans (e.g., balance inbreeding tadpole. with outbreeding; see Bateson 1983). When tadpoles form kin groups, the benefits rnay be di- viswn rected primarily toward kin even if the signals mediating the Tadpoles are nearsighted. Retinoscopic measurements of response are not kin-specific (see Waldman 1986b). For ex- resting refractive state of larval toads (Bufo americanus) show ample, alarm substances released by a tadpole injured by a a myopic defocus of about -275 Diopter (D) in air and -30 predator would become diluted as they difise away from D & &,ter for young tadpoles to abut -4 D in water for I the aggregation (Waldman 1986b). Those tadpoles in clos- tadpoles just before metamorphosis (Mathis et ai. 1988). est proximity to the injured individual would benefit most These measurements are similar to measurements on larval directly. In the case ofRana cascadm, kin would benefit most salamanders (myopic with a resting refractive state of about because they are in closest proximity to one another (O'Hara -3 D; Mantedel et al. 1977) and unlike measurements on and Blaustein 1985). Sidarly, chemicals afFecting growth adult amphibians (hyperopic with a defocus of about 8 D, and development rnay be favored by natural selection be- or nearly emrnetropic; Mantedel et al. 1977; Madus et al. cause they are behavioraiiy directed toward kin (Waldman 1988).It is therefore unlikely that tadpoles, even in the clear1986b). est water, use vision to detea any&ng at great distantes In summary, it is known that tadpoles of several species or with much precision. They are sensitive to light levels of anurans can discriminate between kin and nonkin. Kin (Laurens 1914), and some species (Duniap and Satteheld selection rnay have played an important role in the evolution 1985, Ranapzpzens) select the brightest background regardand maintenance of kin recognition in certain species. The less of preconditioning or track patches of sun across a pond tadpoles of those species that did not discriminate in labora- (although this could be thermal sensitivity; Brattstrom behavtory tests rnay discriminate under different conditions. Per- 1962). !hese resDonses mav be similar to ~hototactic haps, kin recognition and kin association were important in ior characteristic of many other animals including salamanthe evolutionary past under diferent environrnental and so- ders (Anderson 1972) and a few anurans (e.g., Altig and cial regimes, and what we now see are rernnants of behaviors Brodie 1972, Ascaphws trctei). Other species (e.g., Osteopilus that persisted. Finally, in iight of evidence that newly meta- brunneus) flee from intense iight presumably to avoid overmorphosed ranids discriminate between kin and nonkin heating (Lannoo et ai. 1987), and the phototactic prefer(A. R. Blaustein et al. 1984; Corneii et al. 1989), enhanced ences of tadpoles change ontogenetically (R. G. Jaeger and optimal outbreeding rnay be a potential function of kin rec- H a h a n 1976). ognition (sensu Bateson 1983). We need much more empiriVision also rnay be important in determining spatial arcal information to derive specific conclusions about the rangements in stationary schools in some species. Wassersug adaptive value of kin recognition and kin association in an- et al. (1981b) found that schools ofXenopus were more orgauran larvae (A. R. Blaustein et ai. 1990). nized (i.e., nearest neighbors were more nearly parallel-or

238

KARIN vS. HOFF, ANDREW BLAUSTEIN, ROY McDIARMID, RONALD ALTIG Differential predation was attributed to different swirnming speeds and tail flexion, activity pattems, and habitat use among species. Chovanec (1992) suggested that the laterai eyes and greater visuai field of Hyla accounted for lower preAzevedo-Ramos et ai. (1992'1 examdation in that s~ecies. ined predation by aeshmd naiads (Gynmantha membranalis) on tadpoles of the hylid frogs Hylagegraphica, Osteocephalus taunnus, Phyllomedusa tamius, and Scinax mbra from near Manaus, ~ r k i l They found a significant negative correla. tion between average distance moved per unit of time and survivai time, and a positive relationship between rapidity of movement inde~endentof distance covered and survivai time. Species (e.g., Scinax mbra) that were motionless for longer periods and then moved quickly, suffered less predation. Tadpole immobility is an effective deterrent to sight hunting predators but potentially results in reduced foraging efficiency. Skelly and Werner (1990) showed that tethered naiads of predatory dragonflies (Anaxjunius) caused a 41% decrease in activity of Bu@ americanus kept in small containers, and the tadpoles avoided the side of the container mhabited by the odonate naiads. Werner (1991) established that in the presente ofAnax naiads, small larvae ofRana catesbeiana and R. clamitans aiso reduced activity and sought retreats, but R. catesbeiana becarne the superior competitor because it reduced its foraging less t h k d i d R. clamitans. Werner and Anholt (1996) showed that such activity led to a decrease in growth rate, presumably because of reduced foraging activity in smaller tadpoles; they reported that these behavioral responses were mediated tl-kouih chemical modalities, with detection occurring either directiy from the presence of caged predators or indirectiy through exposure to water that previously held the predator. Apparently, tadpole responses are size dependent because larger R. catesbeiana tadpoles continued to forage in the presence of caged Anax naiads and had higher growth rates because they had access to the resources made available by the depression of foraging activity of smaller tadpoles (compensatory reduction in competition) as a result of the naiad presence. In the presence of Anax naiads, growth rates of the larger builfrog tadpoles increased 1.26-fold and a positive effect was detected both on the fraction of the popdation that metamorphosed and the mass of the metamorphs. Such behaviorally indirect effects acting through bodyAsizeand activity levels of the species can have a major and variable effect on community dynamics (chap. 10). Such results suggest that maintenance activity (e.g., feeding and swimming) as weii as temporal and spatiai use of microhabitats in the presence of predators may be mediated by sensitivity to chemicals emitted by the predator (aiso see Kiseleva 1984, 1992, 1997; Kiseleva and Manteifel 1982; Lefcort 1996; Manteifel 1995; Manteifel and Zhushev 1996; see D. J. Wdson and Lefcort 1993 for d e c t s of predator diet on the aiarm response). The presence of speciesspechc aiarm substances & suggested by the occurrence of aiarm responses in Bu@ bmeas to conspecifics damaged by giant water bugs but no such response to damaged heterospecific tadpoles (Hews 1988).

antiparallel) in the light than in the dark. Tadpoles of Rana catesbeiana can move their eyes (Stehouwer 1988), andXenopus tadpoles track movements with their eyes and turn their bodies to foiiow moving visual patterns (i.e., optokinetic response, P. Grobstein, personal communication). Aubum and Taylor (1979) showed that R. catesbeiana larvae oriented to polarized light cues indoors if they had been trained outdoors. Tadpoles were acclimated in an outdoor tank with a deep end at a particular compass heading. Presented with indoor patterns of polarized light, these animais oriented to the compass heading (deep end, outside) rnferred from the pattern of polarized light expected at that time of day. Goodyear and Altig (1971) studied orientation of bullfrogs during metamorphosis, and Justis and Taylor (1976) demonstrated that extraocular photoreception could be used in compass orientation of larval builfrogs. Many amphibians, including anurans, have extraoptic photoreception via the pineai body that may have some funaion in orientation (review by Adler 1970); we know of no studies of pineai-mediated behavior in tadpoles. Leucht (1987) found that Xenupus tadpoles changed pigmentation to match the background differently when weak magnetic field fluctuation was added to the normal light cue. Cutaneous Senses and Chemical Communicatwns Tadpoles have weii developed lateral line systems (Lannoo 1987; chap. 6) that presumably detect disturbances in the water. Lum et ai. (1982) noted that stationary schools of Bu@ americanus and Xenupus are significantly nonrandom even in the dark. It is likely that laterai h e information plays an important part in school structure, but the critical experiment (e.g., severing the lateral line nerves) has not been done. Tadpoles are extremely sensitive to touch, and light stroking or prodding of the skin elicit strugghg or escape response from both embryos and larvae. These responses may be mediated by Rohon-Beard ceiis, the terminais of which are in the epidermis (Fox et ai. 1978) and rapidly adapt to light stroking (A. Roberts and Hayes 1977).No studies have addressed the details of this behavior. The specific activity of cold-sensitive neurons in the s h is correlated with behavioral temperature selection (Dupr et ai. 1986) Tadpoles of severai species preferentially associate with sibiings and the mechanism for sibling recognition is presumably chemical (review in A. R. Blaustein et ai. 1993; see Aggregations above). Aversive reactions to predators and apparently conspecifics under some conditions are aiso weii documented. The presence of predators often results in tadpoles spending more time in refugia (Petranka et ai. 1987; Sernlitsch and Gavasso 1992) or othenvise aiters tadpole biology (e.g., R. A. Griffiths et ai. 1998; Kupferberg 1998; Rode1 and Linsemair 1997), and controiied predation studies have hghlighted the survivai vaiue of different patterns of swimming and activity among tadpoles. Chovanec (1992) compared the susceptibility of tadpoles of four European frogs (Bombina bombina, Bu@ buJo, Hyla arbmea, and Ranu dalmatina) to predation by dragonfly naiads (Aeshna cyanea).

BEHAVIOR

239

Chemical cornmunication is implicated in intraspecific and possibly interspeciic interference competition. Steinwascher (1978b, 1981) documented changes in growth rate and survivorship in mixed-size (one species) and two-species assemblages. One species can mhibit the growth of other species, and large tadpoles may inhibit the growth of smaiier conspecifics (chap. 10). Behavioral changes such as reduced feeding activity were not documented in these studies. Most behavioral work on anuran larvae has concentrated on immediate responses to a variety of sensory stimuli (e.g., iMunn 1940a, b). Other studies considered the effects of early experience on habitat preference (Punzo 1976; see Aggregations above) and simple avoidance learning (W. J. Hoyer et al. 1971; Punzo 1991). Both types of studies illustrate the abiiity of anuran larvae to m o d e somewhat their behavior by assimilating new information, an ability previously thought to be the exclusive province of long-iived animais in heterogeneous environments. Punzo (1991) noted that learning of simple, active, dark-cued, shock avoidance (i.e., hght turned off before electric shock) in some tadpoles that school (e.g., Rana heckrcheri) was increasingly enhanced as group size increases to about 10 individuais. Nonschooling tadpoles have not been tested, but other studies suggest that schooling and nonschooling species may have d8erent learning abiiities. Hoff (1988a) showed that escape directions relative to the positions of nearest neighbors d8ered between schooling and nonschoohg species. In particular, shock-induced startle responses of nonschooling Ranaseptentrionalis f d e d to reflect the changing position of the conspecifics around them (i.e., there were many cohsions), whde schoohg tadpoles of Bufo americanus never coiiided with their conspecifics during startle behavior. The pattern of avoidance directions of the R. septentrionalis tadpoles was identical to that of addt anurans avoiding present or recently seen obstacles (Hoff 1989, R. catesbeiana; Hoff and Ingle 1988, R. pipiens). That is, they avoided the positions previously held by their nearest neighbors but not those positions currently held by those neighbors. T h s observation suggests that the short-tem spatial memory used by addt anurans to form a map of potential escape directions whde their eyes are focused on near-field activities such as feeding (Hoff and Ingle 1988) may be present in nonschooling tadpoles (which deai only with stationary objects in their

environment). T h s abhty may be absent or suppressed in schooling tadpoles that must update spatial information continuously during avoidance behavior. Strickler-Shaw and Taylor (1990, 1991) and D. H . Taylor et ai. (1990) showed that sublethal doses of lead inhibited learning in tadpoles of Rana chmitans. Steele et ai. (1991) showed that tadpoles do not avoid lead-rich water. Escape behavior of tadpoles ofRana ylvatica changes ontogeneticaiiy, and R. M. Brown and Taylor (1995) suggested that these changes are correlated with poorer swimrning performances of younger and older tadpoles relative to those of intermediate stages. During stages 29-37, tadpoles had a low probabiiity of deviating from a straight escape trajectory, while younger (< 29) and older (> 37) tadpoles showed a greater propensity to turn and maneuver at sharp angles from the initial trajectory.

Summary
Aithough there is much that is not known about the mechanicai aspects of the behavior of anuran larvae, the mdiments of behaviors associated with feeding, respiration, thermoregulation, and locomotion are understood and documented. The same cannot be said of the social behavior of these animals. Empirical investigations suggest that there is as much or more variety in the details of social behavior as there is in any aspect of tadpole behavior. Aithough the ecological context of some behaviors, such as preferential km association, is understood for some species, the understanding of general patterns must await more experimentation.

ACKNOWLEDGMENTS
Kate Spencer, whose artistic skds are considerable, designed and produced the two drawings for this chapter; we greatly appreciate her willingness to work with us. Karl-Heinz Jungfer loaned siides and provided information on the tadpoles of Anotheca spinosa (fig. 9.3A), Osteocephalus oophagus (fig. 9.3B), and Mgophys mntana (fig. 9.5A). Mark-Oliver Rode1 and Karsten Grabow shared their knowledge of Hemisus mamratus biology with us and K. Grabow l o k e d the slide reproduced in figure 9.3C. Jonathan A. Campbel loaned the slide of Petropedetesparkeri shown in figure 9.5. To each of these colleages weexpress our thanks.

ECOLOGY
Resource Use, Competition, and Predation
Ross A. M o r d

Introduction
As noted in the quote above, R. M. Savage (1962) recognized that the ecology of tadpoles is a cornplex and variable reflection of rnost of the processes that govern the dynamics of populations and cornrnunities. During the 1960s and 1970s severai pioneering studies took advantage of this idea by using larval amphbians as rnodels for examining ecologicai and evolutionary processes (see fig. 10.1). Continuation of this trend into the 1980s and 1990s led to a rapid expansion of the tadpole iiterature. Anuran larvae were used in studies ranging from the genetics of growth (aiso see Rist et ai. 1997) and development to the organization of complex communities and ecosysterns. Sede's (1987) synthesis of the literature on amphbian energetics is appiicable to various subjects below, and Wibur (1980) summarized the state of tadpole ecology at the start of the 1980s. In this chapter, I provide a similar summary, and point out new areas of inquiry that have arisen from work done primarily since 1979. Ali staging notations are from the systern of Gosner (1960; aiso see table 2.1).

An adequate picture of the instability of life in a

pond, and of the ups and downs of tadpole life, cannot be gained by compressing the events into one colorless summary. The risks, the successes, the disasters overtaking whole populations are fundamental to the ecology of the animal, and for their understanding detailed descriptions of specific instances are essential.
R. M. Savage 1962:24

Tadpole Habitats and Communities

General Features
The generalized anuran life history, as defined by Wassersug (1974, 1975), includes a free-iiving, primarily herbivorous larvai stage. Because anuran larvae are specialized for growth and rnust transform before reproduction (Slade and Wassersug 1975; Wassersug 1974; Wilbur 1980), they cannot forrn stable long-term ecologicai associations. Some larvae exploit relatively sirnple, - predictable habitats, such as water-holdmg epiphytes, large nut husks, and large snail shells (~aldwell 1993; Grandison 1980; Inger 1985; Lannoo et ai. 1987; Orton 1953; Starrett 1973) in tropicai areas. Others develop in more complex perrnanent aquatic habitats as temporary invaders in established communities (e.g., Faragher and.Jaeger 1998; Sinsch 1997; Werner and McPeek 1994). These situations can lead to increased cornpetition and predation pressure frorn perrnanently resident species (~radfrd 1989; Heyer et ai. 1975; Kats et al. 1988; Wassersug 1974; Werner 1986). Most anuran larvae inhabit ternDorarv habitats that rang.e I a frorn depressions in the trunks of fallen trees in tropicai forests (Starrett 1973) to large pools in boreai habitats (Koskela 1973).Unpredictable temporai and spatiai distributions and cvciic vattirns of nutrien; availabili6 are common features of thek habitats. New patches of habitat rnay appear and disappear on a tirnescale of a few years. Existing patches ofi

ECOLOGY

241

ten iil and dry in response to broadly seasonal patterns that vary between sites and years. Mineral nutrients and hghquality detritus are likely to be most available early in each episode of fiing and drying (Biklcher et al. 1978; McLachlan 198la, b; Osborne and McLachlan 1985; Wassersug 1975). The cycle of W n g and drying, combined with fluctuations in resources, leads to complex changes in the quality and quantity of available habitat (Wilbur 1987). These Processes Primary faciors

changes preclude the estabiishment of stable ciimax cornmunities and may fachtate invasion by transient organisms such as anuran larvae.

EcOh~iicd.L Interaccions It is iikely that the interaction of biotic and abiotic factors influentes tadpole ecology (Dunson and Travis 1991; fig. 10.1). During their lives, tadpoles of most anuran species
Components Secondary faciors
Population structure

Reproduction 4

Mate selection Site seledion Phenology

Site history Habitat and microhabitatselection Long-term climatic history Reeent weather history

Physical

I
Survival
Biotic

Water movemenis pHlnutrienislcontaminants Dissolved O, Desiccation risk

I
Antipredator behavior Phenotypic plasticity

Interspecific competition Intraspecific competition Interspecific predation Intraspecific predation Diseaselparasites

(U

2 m

Historicallpriority effects Microhabitat selection Temperature fluduations

Development

Fig. 10.1. Summary of factors affeaing the ecology of adults, eggs, and hatchlings (combined because of similar processes), and tadpoles of anuran amphibians. Four major processes control transitions through and between life history stages, and these processes are iduenced by primary factors in or of the biotic environment. While secondaq factors are reflected primarily in the reproductive ecology and behablor of adults, they can affect direcdy the biotic environment of a larva (e.g., microhabitat selection).Arrows indicate interactions and their directions, which need not be directly causal. For example, developmentai differences are larva1 responses to components of the environment and iduence the duration of larvai period and thus the degree and type of exposure to the physicai environment.

242

R O S S A. A L F O R D

increase in mass by factors of 50 or more, They may experience local densities of several hundred inhviduals/m2 or 25+/liter (table 10.1). Most of the habitats occupied by tadpoles are used by more than one species and contain herbivores of a variety of other taxa, many of which may compete with tadpoles (Morin ct al. 1988). Most also contain predators (Cooke 1974; C. L. Rowe et al. 1994), and the roles of predator and competitor may switch through time (L. Blaustein and Kotler 1993; L. Blaustein and Margaht 1994; Hearnden 1992; Petranka et al. 1998). The phenology and reproductive site preferences of adult &ogs control the exposure of tadpoles to con~petitors predators. and The general type of reproductive habitat chosen (temporary or permanent water, surroundrng habitat type) has a strong influence on the entire developmental strategy followed by many species. Reproduction in permanent water decreases the risk of drying before development is complete, generally increases the diversity of competitors and prcdators prescnt, and decreases the availability of food. Species spawning in permanent water tend to have prolonged larval periods and metamorphose at sizes larger than temporarypond spawners (Patterson and McLachlan 1989). In part this may be influenced by adult reproductive parameters; egg size and clutch size can depend on the type of reproductive habitat used by a species (Tejedo and Reques 1994a). Woodward (1987a) found that species breeding in tempo-

rary ponds had relatively smaller numbers of relatively larger eggs than permanent pond species. Also, differences in relative egg and clutch sizes among breeding aggregations suggested that inexperienced females might produce fewer, smaller eggs and visit less optimal sites (Woodward 198%). Larvae of smcies that sDawn earlv in the historv of a temporary pool experience different ecological conditions from those that reproduce later (Wilbur and Alford 1985). Species that reproduce explosively produce a single cohort of larvae that initially are very similar ecologically. Members of such a cohort may diverge into a variety of size classes as growth and development proceed (Wilbur 1984). Species with extended reproductive periods are likely to spawn several times durinea season and produce distinit larval cohorts with different bGdy sizes that Lay function as separate 'ecological species" (Alford 1986a; Alford and Crump 1982; Polis 1984). In some species, this ecological separation between merent-six conspecifics reaches the point where larger individuals become predators on smaller ones (Crump 1983). Larvae of early cohorts may also prey on conspecific eggs of late cohorts, and large larvae may also prey on small larvae or eggs of other species (Banks and Beebee 1987; Hearnden 1992). Tadpoles are thus likely to encounter a variety of predators and both intra- and interspecific competitors. t heir ecological interactions often will depend not simply on the identity of the species they interact with but
I
i i

Table 10.1 The mean, standard deviation, minimum, and maximum numbers (listed in order vertically) of tadpoles per liter and tadpoles per square meter in field samples from a temporary pond in northern Florida in 1981.Number of samples and pond area (mZ)are given below each date. Expanded from Mord 1986b.
.

1 March 7 1400
Liter m2

8 Mar& 12 1470
liter m2

23 march 6 612
liter m2 liter

5 April 11 264
m2

13 April 10 80 liter m2

17 April 10 50 liter ma

19 April 12 20 liter m2

ECOLOGY

243

on their own body size and that of others (Aiford 1986a; Werner 1986; Wilbur 1988). The direa ecological interaaions experienced by tadpoles occur through a variety of mechanisms. Intra- and interspecfic competition can be caused by exploitation: the simple depletion of resources (Kuzmin 1995). The intensity of exploitative competition depends on the degree of resource overlap between species, and the degree of overlap can change evolutionarily or behavioraiiy in response to local conditions (Aiford 1986a). Negative effeas of density can aiso be caused by interferente -some tadpoles either reduce the access of others to preferred resources (R. M. Savage 1952; Steinwascher 1978b) or release aiielopathic chemicals or syrnbionts (Goater 1994; R. A. Grfiths et al. 1993; Licht 1967; Steinwascher 1979b; also see Tejedo and Reques 1994b). In most popuiations, the negative effeas of density are probably produced by a combination of interferente and exploitation mechanisms (Alford 1994; Berven and Chadra 1988; Semiitsch and Caldweil 1982; Steinwascher 1979b; this chapter). The effects of either of these mechanisms can be mediated by larvai genotype (Aiford 1989b; Newman 1988b, 1995; Travis 1980a, b, 1983a, b; Travis et ai. 1985a), body size (Aiford and Wilbur 1985; Ebenman 1988; Kupferberg 1997a; Werner and Gilliam 1984), species-specificcompetitive a b i i i ~ environmental heterogeneity and microhabitat differentiation (ALford 1986a; DiazPaniagua 1987; Waringer-Loschenkohi1988), and behavior (Aiford 1986a; Morin 1986). Tadpoles aiso interaa with other types of herbivores. A variety of aduits and larvae of herbivorous, aquatic inseas, crustaceans, and zooplankton co-occur with tadpoles and use similar resources. Few studies have examined either the impaa of tadpoles and other aquatic herbivores on their shared resources or possible interphyletic competitive interactions involving tadpoles. Aiford (1989~) demonstrated experimentaiiy that tadpole densities dected the growth rates of larvai newts, which prey on zooplankton, and suggested that this was probably an indirect effea of competition between tadpoles and zooplankton. Morin et ai. (1988) demonstrated that interphyletic competition can rnfluence the growth of tadpoles. Bronmark et ai. (1991) found that larvai Rana temporaria negatively d e a e d the growth of snails, whle snails apparentiy faciiitated the growth of Rana, possibly by aitering the composition of the aigai assemblage. Interphyletic interaaions have &o been demonstrated by L. Blaustein and Margalit (1994) and Kupferberg (1997b). Interactions of tadpoles with predators have recentiy been studied in some detail. These interaaions are generdy complex and not described weil by simple species-specific coefficients of predation rate. Many predators of tadpoles, such as fish and salamanders, are gape iimited. Larger tadpoles swim faster (Wassersug and Hoff 1985) and rnay be Risk of more capable of escaping predators (Feder 1983~). predation is thus often size-specific, either decreasing monotonically with increasing tadpole body size (S. J. Ridiards and Bull 1990a, b) or increasing to a maxirnum and then decreasing (Brodie and Formanowicz 1983; Wilbur 1988). Tadpoles rnay &o aiter predation risk by behaving differ-

endy when predators are present. Behavioral changes often involve alterations in microhabitat seleaion (Formanowicz and Bobka 1989; Hews 1995; Horat and Semiitsch 1994; Petranka et ai. 1987; Semiitsch and Reyer 1992a) or total aaivity leve1 (Chovanec 1992; Hews 1988; S. I? Lawler 1989; Skeily and Werner 1990). The uitimate effeas of ecological conditions on tadpoles are often diKcuit to determine. Changes in tadpole body size relative to age, stage, or rnetarn~r~hosis ~ e w m & (cg.; and Dunham 1994) and survival and developmental rates are relatively easy to measure. Dficulties arise in interpret. . ing how these responses d e a the popuiation biology of species throughout their life cycle. Although it is possible to detea some patterns of relationship between aduit body size and size at metarnomhosis (Patterson and McLachlan 1989: Pough and Kamel 1984), most anuran larvae have considerable flexibility in their rates of growth and development (Wilbur and Colhs 1973). This means that single measures of tadpole performance such as metamorphic b6dy mass can be misleading; tadpoles from two popuiations rnay metamorphose at the same size after very different larvai periods. Alternatively, two popdations rnay have similar larvai periods but very different masses at metamorphosis. Berven (1990) and D.C. Smith (1987) showed that increased metarnorphic mass can increase survivai to first reproduaion. John-Alder and Morin (1990) dernonstrated that jumping abiiity in metamorphic Bu@ woodhousii is positively correlated with body mass. Blouin (1992b) demonstrated genetic correlations between larvai life-historv charaaers and asDeCtS of terrestrial morphology. Egg deposition in environments with relatively high predation risk rnay be advantageous if predation reduces competition sufficientiy to increase the bodv size and thus the survival of metamor~hs. These complidtions do not d e a the validity of tadpiles as model organisms for studies of ecological processes, but it is difiicult to relate studies of tadpole ecology to the popuiation biology of frogs throughout their iife cycles.
I

Resozlrce Use and Efects on Resozlrces Kenny (1969b, c) and Wassersug (1972) described the basic morphology and mechanics of feeding in anuran larvae (chaps. 2, 3, and 4). Most tadpoles have keratinized jaw sheaths surrounded to some extent by rows of labial teeth, which in turn are surrounded by labial papdlae. The jaw sheaths and teeth remove material from substrates, and the jaw sheaths aiso chop larger pieces of material into sizes that fit into the mouth. Food and respiratory water are transported through the mouth, buccal cavi$ and gills by buccal pumping. Food scraped into suspension from a substrate or ingested diredy from the water column is removed from suspension by branchial filaments and entrapped on branchial mucus. Species differ in their iitration rates and abilities to ingest particles of different sizes, but most are very efficient at extracting a wide variety of particle sizes from the incoming water. Many kinds of tadpoles can ingest particles that rnay range from below 1ri, to as large as hundreds of micrometers (Kenny 1969b, c; Seaie et ai. 1982). Seaie and coworkers

244

ROSS A. ALFORD

(Seale 1982; Seale and Beckvar 1980; Scale et al. 1982; Seale and Wassersug 1979) investigated the feeding dynamics of several species of tadpoles. Laboratory trials (Seale and Beckvar 1980) compared the abihties of Bufo americanus, B. wmdhwiifowled, Pseuidacris cmc$?er, Rana catesbeiana, and R. <vlvaticato ingest suspended blue-green algae. All five species ingested cultured algal cells (Anabaena sphazca) at similar maximum rates. Ingestion rates of B. v.fmf& and R. catesbeiana were constant when the density of algal cells exceeded about 1 x lo7 pm3/d3.Pseudacris crucifer reached its peak ingestion rate at a concentration about 2 times greater. Filtration rate decreased as food concentration increased but at a rate that kept ingestion rate constant. Rana catesbeiana larvae ingested monocultures of six species of algae of widely varying sizes and morphoiogies at similar maximum rates, suggesting that these tadpoles are relatively unselective feeders. Seale (1982) compared the filtration ability of an obligate suspension feeder (Xenquslamis) to those of facultative filter feeders (Ranapqkens and R. -'phe~~)cephala). found She that X e ~ q g s tadpoles regulated their filtration and ingestion rates by altering pumping rate, while tadpoles of Rang altered their pumping volume. & n o p were able to reach maximum ingestion rates at a lower partide concentration then Rana. Her results suggest that there may be some trade-off between pure filter-feedng efficiency and the ability to feed by scraping particles from the substrate. Viertel (1990) examined suspension feeding at low algal concentrations by larvae of Bufa bufo, B. cahmica,&na temporaria, and Xenopus hvis. He found that Rana filtered algae at very low concentrations, while Bufo filter-fed at low concentrations only in later developmental stages. All species were more efficientat higher concentrations of algae, and all but Xenopw scraped food from the substrate in addition to filter feeding. Viertel (1992) found that the feeding rates of four species responded differently to difFerent particle sizes and that responses also changed ontogenetically within species. In general, feeding was initiated at lower particle concentrations when particles were larger (see Kupferberg et al. 1994). Some authors have used oral and body morphology to suggest likely foods and feeding behaviors for tadpoles. Orton (1953) suggested that tadpoles fell into seven major ecological groups: arboreal, carnivorous, direct development, generalized lentic, mountain stream, nektonic, and surfacefeeding. Orton's work was expanded by Starrett (1973). Inger (1986, discussed below) recognized five ecological types based on morphology and diet. Altig and Johnston (1989) synthesized the available information on feeding morphology and behavior of tadpoles. They proposed the terms endotroplic, to describe species that gain their nutrition entirely from initial oogenic processes of the mother, and exotrophic, to describe species that consume food during development. They also proposed an ecomorphological classification that includes 6 types of endotrophic larvae and 18 types of exotrophic larvae. This scheme should be of considerable use in developing hypotheses about ecological interactions among tadpoles. At least one species of tadpole (Crump 1989b; Bufapen2lenes) seems to switch facultatively between exotrophic and endotrophic nutrition.

The contributions of Kenny, Seale, and Wassersug greatly advanced our knowledge of tile mechanics and rates of tadpole feedmg, and the -classification of Altig and Johnston (1989) allows more specific suggestions about likely feeding mechanisms based on morphology; data on food usage and conversion efticiencv are relativelv scarce. Several studies have examined tlre ibility of tadpdlcs to use different food types. Some researchers used morphological features to predict d e t composition, while others directly examitled growth and devekpmental rates on a variety of foods. Most tadpoles are primarily herbivorous, but many are also capable of feeding on carrion and a few are carnivorous. R. M. Savage (1952) exarnined the gut contents and clearance rates of tadpoles throughout development. The guts of Bufa bufo and Kana tempmaria contained a wide variety of food ranging from nektonic algae to detritus. He concluded that longterm observations of feeding preferences and growth efficiencies would be necessary to establish the value of each food source. Digestive clearance times for R. temporaria averaged 6.25 hr at 17"C, while B. bufa cleared their guts in 4.75 hr at the same temperature. Some algal cells and fragments of higher plants apparently passed intact through the guts of both species. Pavignano (1989) also found identifiable material in feces and suggested that fecal samples could be used instead of gut contents to study tadpole diets. Tadpoles must lack the ability to digest cellulose and rely on mechanical rupture of plant cell wails to extract the contents (R M. Savage 1952). Steinwascher (1979a) raised Rana clamitans tadpoles under a variety of condtions in the laboratory and fed them powdered rabbit pellets. In some treatments, this food was incorporated intogelatin-agar pellets, whch forced tadpoles to scrape their food into suspension before ingestion. The concentration of food in pellets was either high or low; this presumably varied the energy required to extract a similar amount of food from the pellets. The tadpoles exhibited two basic feeding modes: sc;aping food frgm the pellets and filtering suspended food and fecal material from the water column. At low food levels, the proportion of time spent filtering food from the water column increased. The concentration of food in pellets had lit& effect on tadpole growth rates (fig. 10.2). At high food levels, tadpoles spent more time scraping and grew more rapidly on more concentrated pellets. Under these conditions, scraping was a more efficient feeding mode for R. clamitans than filtering, and scraping from concentrated pellets was more efficient than scraping from more dilute pdlets. Altig and Kelly (1974) measured gut lengths and areas and examined gut organic contents of 13 species of tadpoles. Relative gut lengths varied from 1.43 times body length in Ascapbus true; to 8-08times body length in Scaphiopuscouchii. They suggested that species feeding primarily on animal material should have relatively shorter guts, while those feedmg on plant material and detritus should have relatively longer guts. Reexamining their data shows no correlation between the ashed weight/m3 of gut contents (a measure of inorganic material, and thus detritus, in the d e t ) and relative gut length (rZ= 0.018; p = 0.63). Diet is not well known for

ECOLOGY

TREATMENT
Fig. 10.2. Growth of tadpoles ofRanu clantitans reared (see x-axis) at nvo densities, two food levels, and two degrees of food dispersion. Treatments are 2 or 8 individuals per container, high (H) or low (L) food level, and concentrated (C) or dispersed (D) food. Vertical bars = 95% confidencc limjts of the means. Data frorn Steinwascher (1979a).

most of the species they examined, and they presented no tive efficiencies of 7.8% at 22C to 24.9% at 30C and the data on the composition of gut contents, so drawing firm fast clearance times ranned from 29 to 101 min at 22C. " conclusions from their data is ciiffcult. Digestive efficiencies may have been higher in t h s study An alternative approach to understanding the significance than they are in nature because processing of rabbit peiiets of differences in relative intestine length appeared in Horiu- may rupture ceii walis and expose a greater proportion of chi and Koshida (1989), Nodzenski et al. (1989), and Yung their contents for easy absorption than wouid be true for (1904). Horiuchi and Koshida (1989) showed that the in- intact algal ceiis. Dash and Mishra (1989) examined digestestines of Rhacuphoms arboreus tadpoles that were fed plant tive efficiency as part of a study on the growth and rnetamores material from haiching were about-1.5 times as long o; day phosis of ~ & p e & z tmacu&s. ~ a d i o l e smay sometimes 15 as larvae fed an animal diet. Adding ceilulose to the ani- feed on cryptic items; animais apparently scraping the surmal diet produced longer intestines. Intestines of tadpoles face of macrophytes, for example, may actudiy be feeding switched from an animal diet to a plant diet partway through on epiphytic diatoms (Kupferberg et al. 1994). Even though the larval period responded by becoming longer and reach- larval diets and other rearing faaors have been studied by ing maximum lengths similar to those of tadpoles always fed L. E. Brown and Rosati (1997), Carmona-Osaide et ai. plant material. This study showed that relative intestine (1996), Cuiiey (1991), Culley et al. (1977), and Cuiiey and iength can be related to &e proportion of plant material in Sotairidis (1983) in the laboratory and by Castanet and Caethe diet. It also indicates that interspecific comparisons of tano (1995) in the field, dietary requirements and metabolic tadpoles from field coiiections shouid be made cautiously be- efficiency of tadpoles need further attention. cause there may be considerable phenotypic plasticity for relA few studie; have used nrowth and develovmental rates ative intestine length. Nodzenski et al. (1989) examined the of laboratory popuiations to compare the relative vaiue of development of the viscera of pelobatid tadpoles and found feeds. Li and Lin (1935) raised larval Kalouu borealzs at a that Merent organs grow at different rates after tadpoles density of l/liter (274/m2).They found that tadpoles fed anihatch; in thi gut and pancreas experiente gowth mal material (egg yolk and macerated pork) gr& more rapidly and survived at a greater rate than those fed algae or a spurts shortly after hatching. Percent assimilation and clearance times were measured hay infusion. Nagai et al. (1971) examined the relative valby Altig and McDearman (1975) for tadpoles of five species ues of severai diets for larval ~ h f bufo. Food rnixtures in, o. , fed rabbit peilets. Mean percent assimilation of organic mat- cluded vegetable material oniy, vegetable and animal mateter for four species ranged from 53.8% to 85.7%. The fifth rial, animal material oniy, and a cannibal diet of ground species, Rana catesbeiana, showed considerably lower diges- conspecifics. Unfortunately, the cannibal tadpoles were fed
U

246

ROSS A. ALFORD

only one-fifih as much dry weight per day. Tadpoles on the Crump 1992). Three of these are Spea, which develop speanimal and mixed animal/vegetable material ciiets reached ciaiized cannibal morphotypes under appropriate environthe greatest metarnorphic sizes, whle cannibals showed the mental conditions (Pfennig 1990a). Larvae of at least 15 greatest conversion efficiency. This experiment established species eat normal conspecific eggs, whiie m o eat unfertilthat Bufo tadpoles benefit considerably from the addition of ized nutritive eggs provided directly by the mother. At least animal material to their diet and that cannibaiism or scav- 12 species have unspecialized tadpoles that are known to enging from dead conspecifics is an efficient feeding mode. consume conspecific tadpoles, and one species consumes Larval Bufo stomaticus and Hoplobatrachus tberinus respond conspecific metamorphs. At least three species are known to with optirnal growth rates and size at metamorpliosis on a cannibalize both eggs and smaller conspecific larvae, whde diet combining animal and vegetable material (Dash and some species (e.g., Heusser 1970b, Rana temporaria) prey Mahapatro 1990; Hota and Dash 1986; Mohanty and Dash on eggs of both tiieir own and other species. Interspecific predation among anurans, particularly pre1986), as do larval Xenupus lua+> (L. E. Brown and Rosati 1997). Ahlgren and Bowen (1991) found that rates of sur- dation by tadpoles on the eggs and hatchlings of other spevival and growth of larval Rufn americanas fed on deuitus cies, may prove to be very common. The predators in this depended on the source and particle size of the material. Pol- situation gain nutrition whle reducing or eliminating polei1 was not a good source of nutrition for larval Pseuddtris tential competition or predation. Hota and Dash (1983) attempted an experiment on competition between B u . melacriseriataferiarum (Britson and Kisseii 1996). The effect of cannibaiism o11 the growth of larval H y h nostictus and Hoplobatrachus tgerinus, but the experiment pseudopuma was investigated experirnentaiiy by Crump failed because the Rana larvae consumed many of the Buf (1990). These oppoministic cannibals and predators of con- in most of their treatments. Tejedo (1991) found that predaspecific and heterospecific eggs and hatchlings in the field tion by larval Pelobates culcripes and Pelodpes punctatus were raised inciividually in laboratory containers on a 1:l strongly affected rates of embryoriic survival in Bufo calam(dry weight) mixture of ground tadpoles and artificial feed. ita. Magnusson and Hero (1991) tested 14 species of tadThe tadpoles used as food were either H. pseudopuma, Osteo- poles with eggs of 6 species of frogs; every species tested pilus sepientrionalis, or Scinax stauferi. Tn one-experiment, consumed at least some heterospecific eggs. Crossland and each H. pseudopuma tadpole was fed 20 mg of one of the Aiford (1998) found that larvae of all six species of Austratypes of food every 2 days, and in the other experiment, the iian frogs they tested attempted to consume eggs of Bufo quantity of food was adjusted so that each tadpole received marinw when they were available, despite the fact that Bufo food with equal energy content. Iii both experiments, tad- eggs were highly toxic to them. Crossland (1997) found that poles given food containing conspecifics reached signi- larval Limnodynastes omatus consumed eggs of all five other ficantly greater mass at metamorphosis than tadpoles fed species they were tested with, as weii as cannibalizing conheterospecifics. Crump suggested that cannibalism is a par- specific eggs. He also found that the competitive effects of ticularlv valuable source of animal food for tad~oles dis- larval Limnodynmes ornams on B. marinus were reduced and cussed the implications of cannibalism for species popula- when B. marinus entered ponds as eggs after L. omatzts. T h s tion biology and evolution; she pointed out that species that occurred because consumption of Bufo eggs kiiied larval enzazc i cannibaiism should have mechanisms that reduce L. omatus, reducing their numbers and their competitive n the probability of consuming close relativa. effeas on Buf. Larval Rana amurensis preyed on eggs and In experiments with artificial feeds, Steinwascher and hatchlings of Rombina orientalis (R. H . Kaplan 1992). PeTravis (1983'1demonstrated that arowth rates oflarvalRana uanka et ai. (1998) found that larval Rana sylvatica consume chmitans do not respond to changes in the ratio of protein eggs of the salamander Ambystoma maculatum, which is a to carbohydrate, whiie larval Hyla chrysoscelis grow much predator on Rana tadpoles. An alternative method for feedbetter at h i ~ ~rotein:carbohvdrate ratios. Test and McCann ing on animal matter is that of Rbacophomsyiridis, which eat " h (1976) noted that larval Buf americanus aggregate and their own foam nests after hatching (Tanaka and Nishinara switch from scraping substrates to ater feeding in response 1987). Foam nests may also serve to protect g g s from canto high concentrations of protozoans and suggested that nibalism or interspecific predation in some species (Downie protozoans are a higher quality resource than periphyton. In 1990b; Magnusson and Hero 1991; personal observation). general, the results of these studies suggest that tadpoles Estimates of the rate of occurrence of oppoministic can&w more rapidly when animal mate&l occurs in their nibalism and cannibal morphotypes are avaiiable for a few species, but in general iittle is known about the importante diets. Many tadpoles may supplement their diets with animal of cannibaiism and carnivory in the diets of generalist tadprotein through predation or scavenging on conspecific and poles. In a review of the evolution of cannibaiism and canniheterospecific eggs and tadpoles (Crump 1983, 1990,1992; bal morphotypes in amphbians, Polis and Myers (1985) Magnusson and Hero 1991; Tejedo 1991). Crump (1983) concluded that there was little evidence for genetic polymorsuggested that cannibalism rnight be particularly common in phisms and much evidence for switches to cannibaiism in species that reproduce in ephemeral ponds where it can serve response to environmental cues. Pfennig (1990a) demonas a means of rapidly consoiidating resources in a high- strated that the development of cannibal morphotypes in aualitv form. ~ a b a e - o at least 30s~ecies k n o G t o Spea multiplicata is controlled by the avaiiability of animal f are I eat conspecific eggs, tadpoles, or metamorphs (table 10.2; food and (Pfennig 1992b) that diet switches can cause
U
V

ECOLOGY
Table 10.2 Cannibalism in anuran larvae. Abbreviations: CL = cannibal-morph larvae, E M = metarnorphs prior to tail resorption, and NE = special nuuitive eggs. Species Predator Prey NE, L L
=

eggs, L

larvae,

Reference Jungfer 1995 Nagai et ai. 1971 Hearnden 1992 Polis and Myers 1985 Brust 1993 Polis and Myers 1985 Polis and Myers 1985 Heusser 1970b Magnusson and Hero 1991 Magnusson and Hero 1991 Crump 1983,1990 Polis and Myers 1985 Polis and Myers 1985 Polis and Myers 1985 Heyer et al. 1975 Crossland 1997 Crossland 1997 Jungfer and Schiesari 1995 Magnusson and Hero 1991 Poiis and Myers 1985 Crump 1986 Magnusson and Hero 1991 Poiis and Myers 1985 Bleakney 1958 Heusser 1970b Poiis and Myers 1985 Polis and Myers 1985 Poiis and Myers 1985 Polis and Myers 1985 Pfennig 1990a

Anotheca pinosa Buf buf" B. marinw Chacophyspierott+ Dendvobatespumili@ Euphiyctis cyanophiyct& Hopiobatrachusti&erinw H y h arboveaa H. boaw H.8eo8vaphica3 H. pseudoprrma H. zetek+ Hymenachirus boettg& Lechriodusfitcheri Leptodactylus pentadactylus L i m n o d w e s ornatup Litwia rubeUaa Osteocephdusoopha8us 0.taurinup Osteopilusbrunneup 0.septenhonali~ P h y h e d w a vaiUantP Pyxicephalus adpenus Rana sylvaticaa R. tempovanaa Rhinophynw donalis Scaphiopw holbmokii Spea bombtf7orrc S. hammondii S. multiplicata
=Noti table 12.1 of Crump 1992. n

S. multiplicata to change morphotypes. Pfennig suggested (1992b) that the proximate cause of switches may be thyroxine obtained from the bodies of prey. Pfennig (1992a) showed that the freciuencv of cannibal and noncannibal morphotypes aiso appears to be regulated by pond duration and that different ponds have Merent frequencies of the two morphotypes.-~hisresult was extended by Pfennig and Frankino (1997), who showed that both S. multiplicata and S. bombzfrons are less likely to develop cannibai morphotypes when they are raised with groups of siblings. The occurrence of cannibd morphotypes ~pa may be threshold character mediated by underlying genetic variation in liabiiity. Predisposition to cannibalism in species lacking polyphenisms is also Irkely to be underlain by quantitative genetic variation. These questions deserve % d e r investigauon using quantitative genetic techniques similar to those recently applied to salamander paedomorphosis (Harris 1987; Semlitsch and Wilbur 1989). Relatively little information is avadable on the diets of tadpoles; C d e y and Sotairidis (1983), Horseman et ai. (1976), Leibovitz et ai. (1982), and G. A. Marshd et ai. (1980) supply various data concerning tadpole nutrition derived from culturing of Rana catesbeiana. Field studies of tadpole diets are not common (e.g., Belova 1964; Costa and Balasubramanian 1965; Farlowe 1928; Lajmanovich and Faivovich 1998; Sekar 1992; Sin and Gavrila 1977; aiso see
1 ,

Ahlgren and Bowen 1991; Plytycz and Klimek 1992; Syuzyarnova and Iranova 1988) and often not very informative. After examining gut contents of Rana ciumitans before and during metamorphosis, Jenssen (1967) concluded that the larvae are unselective continuous flter feeders. The larvae ceased feeding at stage 42 (forelimb emption) and did not feed again until tail resorption was complete. Hendricks (1973) found that 97% of the food items in the gut contents of Rana pipiens larvae were free-floating forms. By examining the diets of larvai Acris crepitans from two ponds with differing phytoplankton assemblages, L. M. Johnston (1991) concluded that the abundance of aigae in tadpole guts were related to their abundance in the environment and that tadpoles from the pond with more abundant phytoplankton grew more rapidly. Johnston's data aiso suggest that A. crepitans are selective feeders; comparisons of proportionai similarity showed that the diets of animais were 25% and 67% similar to the composition of the aigai assemblage from the two ponds but were 82% similar to each other. C. L. Taylor et al. (1995) demonstrated experimentdy that larvai Rana sphenocephaiu and Bufo woodhousii selected agar blocks with lower agar concentrations and containing food with more anirnai material. Dietary information may suggest ecologicai relationships among species. The diets of seven species living in Spain were related to morphology and microhabitat by Diaz-

248

ROSS A. ALFORD

Paniagua (1985, 1989). Proportional similarities (Feinsinger et al. 1981) between pairs of five species varied from 0.4 to 0.7. Relative intestine length varied among species and was short in detritivores. Relative intestine length generaiiy increased with development to about Gosner 36 and then remained constant or decreased. Pavignano (1990) concluded that the diets of larval Hyla arborea and Pelobates fscus showed large overlaps consistent with a high degree of competition. Twelve of 16 species in a Bornean rain forest (Inger 1986) occupied stream microhabitats. Inger divided these inhabitants into 3 groups based on food particle size: macrophages-2 species that fed on relatively large particles (mean size > 0.12 mm) of mostly vascular plant fragments; mesophages - 3 species that fed primarily on intermediate sized particles (0.069 mm < mean size < 0.08 mm); rnicrophages -7 species feeding on small particles (mean size < 0.05 mm). He found broad overlaps in types of food eaten among species within feeding types. Some of this apparent lack of dietary differentiation is probably because of the low statistical power of tests caused by a small sample size (maximum N = 3 per species). Inger suggested that much of the dietarv differentiation he observed was because of ~hvietic dzerences in feeding behavior, rnicrohabitat preference and morphology rather than to immediate competitive pressures. Trenerry (1988) examined the diets of four species of tadpoles from a rain forest stream in northern Queensland, Australia, during wet and dry seasons. The suctorial tadpoles of Litoria nannotis and Nvctimvstes davi inhabit riffles and other , , fast-moving water in the same areas of the creek. Their diets were 89% similar in the wet season and 85% similar in the dry season (proportional s d a r i t y index; Feinsinger et al. 1981). The large dietary overlap suggests that they compete or occupy dZerent rnicrohabitats, and food rnay not be a limiting resource. Tadpoles of two pool species, Litoriagenimaculata and Mwophyes schevilli, had diets that were 61% similar in the wet season and 68% in the dry season. The diets of the pool and riffie species were 65% similar on average. Parasites and commensals of tad~oles have not been studied extensively, and this field of research would seem to be worthwhile in light of the current interest in the declining and deformed amphibian phenomena. I present only exarnples of citations on the subjea. The numbers, developmental stages, and species of enteric protozoans vary arnong tadpoles and depend on the developmental stage of the host (Affa'a 1979, 1983; Affa'a and Amiet 1994; Hegner 1922; Oku et al. 1979; Schorr et al. 1991; Wessenberg 1978), but the significance of these presumed commensals to the welfare of the host has not been determined (see Beebee and Wong 1992a, b). Affa'a and Arniet (1994) presented an engaging sumrnary of the biology of nyctotherans w i t h the frog fauna of Carneroon. Studies of other parasites include Desser et al. (1986, myxozoan), Hird et al. (1981, Aeromunas), Nyman (1986,-~eromunas), Papernal and Lainson (1995, rnicrosporidian), and Poinar (1988, nematodes). Studies such as Aho (1990) on parasite communities relative to the ecology and phenology of the tadpoles would be of
I
i

interest. M. L. Adamson (1981a, b) discussed the epiderniology and seasonal changes in populations of a heiminth in wild tadpoles, and Goater (1994) and Goater and Vandenbos (1997) discussed the effects of parasites on larval life history of tadpoles; see Eberhard and Brandt (1995) and Schrnutzer (1977) for other information on the interactions of tadpoles and parasites. Thraii (1972) reported that there were no bacteria in the guts of tadpoles of Rana catesbeiana, while others (e.g., Battaglini and Boni 1967; R. Altig, personal cornmunication) report numerous bacteria. Because tadpoles often occur at relatively h g h densities, usuaiiy have broad diets, and have high rates of ingestion n and clearance, they rnay be the major primary consumers i some ecosystems, particularly temporary ponds. Tadpole grazing can strongly alter the standmg crop and species composition of algae. Dickrnan (1968) showed that grazing by caged Rana aurora tadpoles reduced the standing crop of filamentous green algae to 1%-2% of levels found in ungrazed controls. This altered the species composition of the algal community. Lamberti et a. (1992) found that tadpoles ofrlrcaphw truei (5/m2) could reduce periphyton biomass by 98% and chlorophyll a by 82% in streams in Washington, a response that was about equal to the total effeas of grazing by smaii invertebrates. Seale (1980) showed that tadpoles, feeding primarily by filtration from the water column, significantly altered the distribution of nitrogen from suspended to dissolved compounds. Tadpoles also shifted the community composition of algae in a pond from blue-green algae to green algae. Morin et al. (1990) demonstrated that Hyla andersonii tadpoles reduced the standing crop of periphyton in artificial ponds. In some habitats, tadpoles ma? alter algal species composition through feeding and by promoting nutrient cycling (Kupferberg 1997a; Osborne and McLachlan 1985; Weigmann 1982). Tadpoles rnay also simply decrease algal biomass without affecting species composition (Holomuzki 1998). The feces of tadpoles can have greater nutritive value than the food the tadpoles ingest (Steinwascher 1978a). Tadpole feces can form a nearly continuous layer several centimeters deep in a Carolina bay (Harris et al. 1988) at the end of summer (personal observation). If the tadpole populations observed by Trenerry (1988) in a northern Queensland creek were processing food at rates similar to those measured by Altig and McDearman (1975), in late summer his pool 2 would receive roughiy 1000 g of tadpole feces per day. It seems likely that in some systems tadpole grazing rnay strongly &ect the composition of the plant community and that tadpole feces rnay be an important source of fine particulate organic matter for detritivores of many taxa. This possibility deserves further investigation.

Growth, Development, and Survival

A host of factors has been suggested as potential influentes on the rates of growth, development, and survival of tadpoles in nature. Many species rnay grow exponentiaiiy (Alford and Jackson 1993). Growth rates rnay be influenced by competition, environmental temperature, and other aspeas

ECOLOGY

249

of environmental quality such as dissolved oxygen and p H (see Bradford et al. 1994; Cummins 1989; Dunson and Conneli1982: Picker et al. 1993: Saber and Dunson 1978). Competition couid be of several types, including intra- and intercohort competition within species, interspecific competition within the tadpole "guild" (cf. Root 1967), competition with other vertebrates such as larval fish. and comDetition with members of other phyla. The kechanism'of competition couid be exploitation (usudy defined as a resource demand:supply ratio greater than 1 : l ) or interferente. Interference competition rnay be behavioral or d e lopathic, which couid be caused by chemicals, parasita, or pathogens. Rates of development rnay respond directly to environmental influences such as dissolved oxygen, pH, and temperature (Hayes et al. 1993; chap. 8). Endocrine feedback systems rnay also lead to responses t o extrinsic factors such as photoperiod (e.g., M. L. Wright et ai. 1990) or to intrinsic cues such as growth rate or body size thresholds (chap. 11,and below). Survival is directiy affected by predation and rnay be iduenced strongly by growth rate; mortality rates of tadpoles rnay decrease with increasing sue, and faster-growing tadpoles often metamorphose earlier than slower-growing individuals and thus experiente lower cumulative mortality rates as larvae. Survival is also dependent on biotic and abktic environmental factors- the cm~osition of the assemblage of potential predators and the p;obability of environmental changes such as drying of ephemerd ~ o n d s . A model of influences on development, growth, and survival of tadpoles proposed by Wilbur and Collins (1973) stimuiated a great deal of work. They noted that the develo~mental rates and metarnor~hc suis of individuals of the same species often vary widely within a given physical environment. They suggested that this might be caused by an evolutionary advantage conferred on species that are able t o adjust their developmental rate in response to their growth rate. Adjustment wouid d o w individuals t o alter the proportion of total growth between egg and first reproduction that occurs in the aauatic environment. If aauatic conditions were favorable, do&ing rapid larval gowth: tadpoles wouid slow their developmental rate and metamorphose at a large body size. If aquatic conditions were unfavorable, tadpoles w o d d acceleraie their developmental rate, metamorphse at a species-specificminimum body size, 6, and continue growing in the terrestrial stage. Coiiins (1979) modified their suggested feedback process slightly (fig. 10.3). One ~ r o b l e m with the Wilbur-Collins model is the lack of a spehfic mechanism for the feedback of growth rate on developmental rate. Crump (198 1) found that larval Pseudacri5 &.U$~Y raised at high density and fed ad libitum had lower mass specific energy contents and metamorphosed at smder sizes than tadpoles raised at low density. She suggested that the accumuiation of energy reserves, in the form of body fat, might affect differentiation rate (McCarrison 1921). A direct mechanism that might d o w feeding rate, and thus growth rate, to control differentiation rate was suggested by Wassersug (1986). H e found that a hormone in the prostaglandin E group (PGE,) is secreted into oral mu-

cus by some tadpoles and transported into the stomach with entrapped food particles. Inhibition of the development of aduit stomach features is caused by PGE, (Wassersug 1986). Higher quantities of PGE, are ingested at higher feeding rates. Wassersug suggested this shouid slow the development of the stomach and could also act to inhibit the operation of the hormonal mechanisms promoting develophent of other features. Mobbs et al. (1988) investigated t h s hypothesis with isolated tadpole t d s in vitro and found no evidente for PGE, inhibition of thyroid-hormone induced metamorphosis.-They pointed out, however, that PGE, might not act directly on tail tissue or that some other component of oral mucus rnay regulate metamorphosis. Delidow (1989) examined the effects of a variety of growth hormones on larval Rana and found differential effects on tail and body growth. H e suggested that differentiation rate was not simply regulated by thyroxine. A number of studies have examined ~redictions the of Wilbur-Collins model, while others haveLproposedalternative models for the control of growth and development in tadpoles. In their original paper, Wilbur and Coliins made several predictions about the responses of growth and developmental rates of individual tadpoles to ecological conditions and the responses of tadpole and metamorph populations t o ecological conditions. They predicted that popuiations in stable, predictable environments wouid tend to maximize the proportion of growth accomplished in the aauatic environment. Most individuals wouid metamor1 phose at or near the species-specific maximum body sue, c, after widely varying larval periods. Conversely, they suggested that popuiations in unstable, unpredictable environments such as temporary ponds shouid show greater variante in both larval period and metamorph body size as individuals adjusted their differentiation rates in response to local conditions. Within populations, they suggested that at low-density/high-resource availability, most individuals wouid grow at rates greater thang and thus wouid delay metamorphosis until reaching maximum size, c. At higher densitiesAower resource avaiiabilities. the first individuals t o reach minimum metamorphic size wouid be growing slowly and wouid metamorphose at or near 6. Their departure wouid make resources avaiiable t o the remaining tadpoles and d o w them to grow more rapidiy and delay their metamorphosis until reaching slightiy larger body sizes. This process wouid continue until the growth rates of remaining individual~ were suKcient to d o w them to metamorphose at maximum size. Wilbur and C o h s (1973) did a preliminary test of one prediction of their model: individuals shouid grow more rapidly when released from density stress. They selected larval Rana sylvatica i 3 size classes (< 100, 101-200, 201n 400 mg body mass) from popuiations that had been growing at high densities (806/m2; 5.33lliter) and placed them in pens at low density (13.4/m2; 0.089/liter). A of the new popuiations began growing rapidiy. Wilbur and C o h s interpreted this as support for their model. Coilins (1979) how the size-freauencv distribuused the model to ~redict tion of metarnorphs emerging from a popuiation of Rana
L ,

250

ROSS A. ALFORD

catesbeiana larvae should change over time. He suggested that the avadability of resources present after fast-developing individuais metamorphosed should lead to populations that obey the Wilbur-Cohs model by showing a pattern of increasing metamorph mass with increasing larvai period. The R. catesbeiana populations he studied appeared to foliow this pattern. Pfennig et ai. (1991) detected the opposite pattern in four natural populations of Spea multiplicata. They varied food levels and the duration of artificial ponds to examine the causes of this correlation. Age and size at metamorphosis were negatively related. Larvai period and size at metarnorphosis were positively correlated in long duration, low-food ponds and were negatively correlated in high-food ponds. They suggested that negative correlations are likely to be common in nature because as ponds age, the animais in them are likely to experiente diniinishing resource availabilities. Kehr and Adema (1990), Reques and Tejedo (1995), and Tejedo and Reques (1992) ais0 discuss correlations between size at metamorphosis and length of larvai period in the context of the Wdbur-Cohs model. Severai snidies examined the effects of decreases in population density late in the larvai period on rates of growth and Merentiation in tadpoles. Semlitsch and Caldwell (1982) reared ScaphUIpus holbrookii tadpoles at three densities in aquaria. On days 7, 15,28, and 56 of their experiment, they

(HATC)

I
NO

4
GROW

t -

YES

W>b+c

Fig. 10.3. The Wilbur-Coiiins model relating growth and development rates of tadpoles. Abbreviations: b = minimum size for metarnorphosis; c = dzerence between minimum and maximum body size for metamorphosis; g = si=-specific growth rate below which metamorphosis is initiated, above which metamorphosis is delayed; and W = body mass. After C o h s (1979).

removed 10 individuals from each density treatment and reared them aione. They found that length of the larvai period was affected by density and by the timing of release from density stress, but the effect of timing of release was independent of density. This suggests that after density relesse, ai indwiduals shified to similar developmentai proi grams. Their experiment demonstrated that differentiation rate can respond to changes in ecological conditions during the larvai period. Travis (1984) used l a ~ Hylagratwsa to d determine whether low food level aione could produce a positive correlation between mass at metamorphosis and length of larvai period within a cohort and to examine the effects of competitive release. He reared individuai tadpoles in 500 rnl of water. Two sets of 40 individuals were fed at constant rates of 1 and 2X per individuai. In 2 other sets of 40 (the "release" treatments), the food rations of metamorphosed animais were divided among those remaining. Larvae fed at the 1X rate had more variable larvai periods than larvae fed at 2X. Larvai periods of animais in the release treatments were no more variable than those animais fed at constant relative rates per individual. Metamorphic masses of individuals in the low-food, release treatment were greater on average than those of animais in the low, constant food treatment and were aiso more variable. Because the relesse treatments did not appear to &ea length of larvai period, Travis proposed that differentiation rate might be fixed early in the larvai period. The laboratoty experiments discussed thus far examined responses to competitive release but did not address the prediction of the Wdbur-Cohs model that animais in deteriorating environments should increase their differentiation rates and metamorphose more rapidly than animais in constant environments. Alford and Harris (1988) used larvai Bufo woodhousii to simultaneouslyexamine the effects of constant, increasing, and decreasing rates of food supply on growth and differentiation rates. We raised tadpoles aione and supplied food at two relative rates. On days 12, 18, and 30 of the experiment, we switched one treatment from the low food regimen to the high food regimen and one treatment from high to low food. We found that differentiation rate was not fixed early in development; individuais switched from high to low food levels as late as day 18 had larvai periods nearly identical to individuais raised at constant low food. Individuais switched from low food level to high food level at days 12 and 18 reached metamorphic masses nearly identicai to individuais raised at constant high food after vety similar larvai periods. Indwiduais switched from low to high food on day 30 reached metamorphic masses very similar to constant high-food individuais but took longer to do so. Differentiation rate thus responded to both increases and decreases in growth rate in a manner consistent with the Wdbur-Collins model. Hensley (1993) performed a series of experiments with Pseudami cvucifer larvae that were designed similarly to that of Alford and Harris (1988). Hensley found that Merentiation rate in this species responded to growth rate in accordance with the Wilbur-Coiiins model until about Gosner 35, at which time it became fixed. Beck (1997) and Leips and Travis (1994) have continued this line

ECOLOGY

251

of research by focusing on when in the lama1 period developmental rate loses its ability to adjust to different growth conditions. Thev demonstrated that the stage at which fixation " occurs varies among species. Beck (1997) s~unmarized adaptive explanations that have been proposed to explain this variation. Although it has also been suggested that this variation may be nonadaptive, to my knowledge no studies have examined the possibility that it may be explained by simple functional morphological constraints. That is, ifthe development of the fore1imG constrains the filteriie and food colleaing apparatus in any way, the potential increase in growth rate achievable if resource levels increase must decrease. Perhaps developmental rates become fixed at or after the point at which thk trade-off between forelimb size and filierine a rate becomes unfavorable for additional growth. This possibility would be fairly easy to examine experimentally. Crump (1989a) examined whether tadpoles of Hylapseuhums increased their develonmental rates when water level dicreased. She raised tadpoles' at a single density under constant high water levels, constant low water levels, and decreasing water levels. Tadpoles raised in decreasingwater levels had the shortest larval periods and metamorphosed at sizes similar to tadpoles raised a t constant low water levels. Tadpoles raised at constant high water levels were significantly larger at metamorphosis than either of the other groups and metamorphosed after intermediate larval periods. Tejedo and Reques (1995) showed that Bufb c a h i t a larvae metamorphosed more rapidly from ponds with shorter duration-~hese studies a&in demonstrated that de" velopmental rate can respond to ecological deterioration in the direction predicted by the Wilbur-Collinshypothesis. Although most tadpoles appear to have at least some ability to respond as predicted by Wilbur and Collins, larval salamanders may not. Beachy (1995) found that developmental rate of larval Desmopathw ochrophaew did not respond to changes in growth rate at any point during the larval period. The Wilbur-Collins model is an abstract model concerned with suggesting how relative rates of growth and development should be controlled within a population experiencing a common physical environment. Smith-Gill and Berven (1979) developed a predictive model for larval period and size at metamorphosis in populations of Ranapipiens exposed to different thermal regimes. They found that differentiation, as expressed by along the Taylor/Kollros staging series (A. C . Taylor and Kollros 1946), was strongly correlated with size and time. This result might have been predicted; the Taylor/Kollros staging series was developed by choosing the morphological benchmarks for each stage deliberately so that stage increased linearly with time in a laboratory population of Ranapipiens. Smith-Gill and Bcrven 11979) also found that the relationshiv between size and developmental stage depended on environmental temperature; developmental rate decreased faster than growth rate as temperature decreased so that larvae grown at lower temperatures grew more in each stage and metamorphosed at larger body sizes than those reared at higher temperatures. Berven et al. (1979) applied these concepts to explain differences between montane and lowland populations o f h
U

clamitans. Pandian and Marian (1985a) developed a similar model that precficts metamorphic size based on the feeding rates and stage-specific growth rates of HopEoba@mhus ti~erinus larvae raised under a variety of regimens involving density, food quality and quantity, and temperature. Werner (1986) developed a model (see chap. 11) that suggests that the timing of metamorphosis should be determined by trade-offs between growth and mortality rates in the aquatic and terrestrial environments. In a stable population. this would lead to metamorvhosis a t a bodv size that minimized the ratio of mortality rate to growth rate at each size. Further development of this type of model by Ludwig and Rowe (1990) and L. Rowe and Ludwig (1991) has increased the model's realism bv allowing fo; constraints on " the timing of first reproduction. These models are not incompatible with the Wilbur-Collins model or any other model thus far proposed for the proximate control of growth and met&~;~hosisin anurGs. These models provide a framework for broader interspecific comparisons (c.g., Patterson and McLachlan 1989; Pough and Kame1 1984). The remainder of this section concentrateson studies that have examined aspects of the growth and survival of single species of tadpoles in a variety of natural and artificial systems. Several of these studies originally suggested that tadpole mortality rates decreased as tadpoles grew. Petranka (1985) examined data from several stucfies and showed that the mortality rate (proportion of the population dyingitime) of tadpoles is usually constant over the larval period. Herreid and Kinney (1966) studied populations of Rana sylvatica tadpoles at fow small ponds in central Alaska. All populations experienced nearly constant proportional mortality from o&position through metamorpho&. They noted considerable predation by larval diving beetles (@tiscw sp.), whose growth patterns may enable them to prey at nearly constant rates on growing tadpoles (Brodie and Formanowicz 1983). Augert and Joly (1993) found that larval Rana tempora.p.iafrom two adjacent sites differed in size at metamorphosis. They suggested that this difference was caused by environmental rather than genetic effects. Calef (1973; Rana aurora larvae in Marion Lake, British Columbia, Canada, 1969-1970) counted the numbers of egg masses deposited and followed the resulting tadpole populations through metamomhosis. In both vears, the tadvoles died at a rela, , tively constant rate, and about 5% survived to metamorphosis. Mean body size at each census was strongly correlated to the number of degree-days since hatching of the cohort. Caging experiments established that food was not limiting. Most mortality appeared to be caused by a variety of invertebrate (e.g., odonate naiads) and vertebrate (e.g., rainbow trout, Oncmhynchus mykiss) predators. This wide range of predator sizes may explain the apparent lack of a size refuge from predation. Relatively constant rates of mortality were also found bv h i s (1991) in - populations of larval Ram dalmatina and A. tem~oraria. Predation also ippcared to be the major cause of mortality in two populations of Rana cmtesbeiana larvae (Cecil and Just 1979) near Lexington, Kentucky. For tulo years, mortal-

252

ROSS A. ALFORD

ity rates appeared t o be constant but growth was exponential. Absolute popdation densities ranged from 7/m2 shortly after hatching t o less than l/m2 nearly 1 year later. Many invertebrate and vertebrate predators were present. Licht (1974) measured growth and survival of R. aurora and R. pretwsa in a temporary pond and a river in British Columbia, Canada. Differences in percent survival t o hatching, an average of 71% of R.pvetwsa and 92% of R. aurora, were caused by choice of oviposition sites; R.pretiosa deposited eggs near the margins of water bodies where they were vulnerable to desiccation caused by s m d fluctuations in water level. Both species showed relatively constant mortality rates throughout development. Cumulative survival from egg to metamorph was less than 1% in the pond and about 5% in the river site. Licht suggested that most tadpole mortality was caused by predation. When predation is absent, tadpoles can survive to metamorphosis at high rates. Seigel (1983) estimated that R. sylvatica in a highly ephemeral pond experienced only 37.5% mortality from egg to metamorph. D. C. Srnith (1983) presented one of the most comprehensive studies of growth and survival of tadpoles under natural conditions. He studied populations of the c h o m frog, Pseudacvis trzjeriata, which breed in ephemeral rocky pools on the shorelines of two s m d islands in Lake Superior, just off the coast of Isle Royale National Park, Michigan. PseudaE1.i~ triseriata required ephemeral pools with intermediate size and duration for development. Larger, more permanent pools harbor popdations of dragonfly naiads (Anaxjunius) that eliminate tadpoles from them, while very s m d , ephemerai pools were either scoured out by r a i d d or dried before tadpoles could metamorphose. Removai of A.junius increased tadpole survivai in large pools. Variation in the persistence of intermediate-size pools caused tadpole survivai rates to vary from O to 100%. Tadpole densities in these pools were high (mean initiai density = 0.95lliter) and ofien increased as tadpoles grew and pool size decreased. There was strong evidence for effects of density on the growth rates and size at metamorphosis of tadpoles. Experimentai manipulations of density and food level showed that competition was occurring. Schmuck et al. (1994) studied the effects of density on growth and metamorphic performance as weii as associated physiologicai responses. Tadpoles of H@emlius viridz@vus responded to increases in nitrogenous waste from metabolic processes or evaporation by increases in the ornithine cycle (increased urea synthesis; also Shoemaker and McClanahan 1973). In contrast, tadpoles of H. mamratus taeniatus tolerated high arnmonia levels (aiso Candelas and Gomez 1963; Harpur 1968). The deterioration of natal pond conditions also increased the postrnetamorphic responses of H. v. nitidulus and H. v. ommatostictus. Newman (1987) conducted a study that used natural ponds but d o w e d a rigorous anaiysis of patterns. He monitored the growth and survival of Scaphiupus cwchii tadpoles in 82 ephemerai desert ponds over a 3-year period. In 16 ponds, predation was the major cause of S. couchii mortality, and desiccation was the major cause of death in 49 ponds; only eight ponds produced metamorphs. High initial densi-

ties of eggs led to slow growth and mortality through desiccation. Supplementing the available food in high-density ponds increased individual growth rates, d o w e d some animais to metamorphose, and demonstrated that resources were in limited supply. Berven (1990) studied the popdation biology of terrestriai and aquatic stages of Rana sylvatica at two ponds i n Maryland over a 7-year period. Enclosure of the ponds with drifi fences d o w e d complete censuses of frogs as they arrived at and departed from the ponds. Survivai from egg to departing metamorph varied from 0.36% to 8.0% (n = 4.5%) at one pond and 0.09% to 1.7% (E = 0.95%) in the other. There was strong correlational evidence that larval growth and survival in each pond depended on larval density. Spearman rank correlations for the two ponds were -0.77 and -0.90 between the proportion surviving to metamorphosis and number of eggs deposited, -0.77 and -0.94 between mean snout-ischium length of metamorphs and number of eggs, and 0.77 and 0.83 between mean larval period and number of eggs. Cohorts developing in years of high adult reproductive effort survived at lower rates, reached smder metamorphic sizes, and spent longer in the aquatic environment than cohorts in years with lower egg input. Juveniles produced from larger, faster developing larvae were more likely to s w i v e to reproductive age, reproduced earlier, and were larger and more fecund as adults. Berven suggested that the consistently lower larvai survivai rates in one pond may have been caused by larger numbers of predators in the pond that dried less ofien. Larvai density, predation, and pond duration appeared to control the growth and survivai rates of larvai Bombina variegata at 46 Swiss ponds studied by Barandun and Reyer (1997). Skeiiy (1996) showed that larvae of Pseudacrij mcifer and l trism'? ata are affected by density, predators, and pond permanente, while L. Blaustein and Margaiit (1995) demonstrated that the distribution of larvai BuJ viridis arnong pools was correlated with the availability of iiamentous aigae. While the studies cited above have I ~roduced useM estimates of growth rate and survivai of tadpoles under natural conditions, most d o w many possible interpretations of the observed Datterns. Growth and survivai of tad~oeshave been studied rigorously in nearly natural systems by two additionai approaches. The first approach is to subdivide natural populations in pens and manipulate factors such as density and predators. The second approach is to construa artificial cornmunities in large outdoor ponds or tanks. Brockelman (1969) placed pens in two ponds to examine the responses of larvai Bufo americanus to conspecific densiq food level, and the presence of predators. Larvai period increased and metamorph body size decreased with increasing density, and the addition of supplemental food reduced but did not eliminate the effects of increased densitv. In one pond, his experimental pens were heavily colonized by predators, and low survivai was not related to initiai density. In the other pond, mortality was lower and strongly related to density; about 45% s w i v e d to metamorphosis at an initiai density of 6 7 tadpoles/m2 (= 0.4/liter), and swivai decreased to about 32% at 178 tadpoles/m2 (= 1.18/liter) and
I

ECOLOGY

253

14% at 471 tadpoles/m2 (= 3.12/liter). The naturai tadpole density in the second pond was higher, and his estimate of curnulative larvai survivai was aiso higher - about 5% of tadpoles surviving to metamorphosis. The pen technique was aiso used by Wilbur (1976) in his study of density effeas on the survivai, growth, and development ofRanaylvatica tadpoles. He reared tadpoles at initiai densities ranging from 100 (= 0.66/liter) to 1612/m2 (= 10.66/liter) in pens placed in a natural pond. Proportion surviving to metamorphosis decreased exponentidy with increasing density. Mean body mass at metamorphosis aiso decreased exponentidy with density, with a lower asyrnptote of about 150 mg reached at an initiai density of 403 tadpoles/m2 (= 2.66/liter). The distribution of body mass at metamorphosis was not significantly different from a normal distribution. Wilbur suggested that this arose as a result of individual larvae foliowing the Wiibur-Collins model of growth and differentiation. Skeliy (1995b) used pen experiments to show that larvai Pseudacrir crctcifr are not excluded i-om permanent ponds by competitio; with Rana clamitans larvae. Travis (1983b), Travis et ai. (1985a), and Travis and Trexler 1986) used ~ o ~ u l a t i o raised in field enclosures to exns amine how ecologicai and genetic faaors d e a the growth and development of severai species of tadpoles. Hylagratwsa larvae of 7 sibships reared in enclosures at an initiai density of 1.6 animais/liter (235/m2) survived to 5 weeks of age at rates varying from 1% to 22% (Travis 1983b). The enclosures excluded larger predators but were colonized by a variety of invertebrate predators. In each enclosure, percent survi"ing was positiv~ly correlated with mean bod; size. This suggests that faster growing popuiations suffered less mortality from predation than slower growing popuiations. Travis et ai. (1985b) further examined the influences of sizelimited predation on H. gratwsa. They raised tadpoles of 2 sibships at initiai densities of 21, 42, and 84/m2 (0.055, 0.11, and 0.22/liter) with predators (naiads of Tmmea h&a, larvaiAmbystoma opacum, or both) present or not. The experiment started 2 weeks &er egg deposition. Survivai to day 10 of the experiment varied from O to 78% with a median of 3 1%. Initiai tadpole density did not d e survivai to ~ ~ day 10. On average, 47% of tadpoles survived to day 10 when both predators were absent. Both predators d e a e d survivai equdy by each reducing survivai by 24% compared to treatments where both I ~redators were absent. The ~ r e d a tors did not influence each other: survivai when both predators were present was exady as prediaed if there was no interaaion between the sources of mortality. The two sibships grew at different rates, and the faster growing sibship had consistently greater survivai than the slower growing sibship. Using an ecologicdy different species, Travis and Trexler 1986) examined the interaction between influences of nenral environmental harshness and density on the growth and survivai of tadpoles of Bufo tev-rer~vzis three initiai size of classes. In 1 experiment, they placed tadpoles belonging to 3 size classes in the same enclosures used by Travis et ai. (1985a) at densities of 42, 84, and 168/m2 (O.11, 0.22, and
I I I U

0.44/liter). Initiai body size explained a significant amount of the variabiiity in survivai to metamorphosis; tadpoles with the iargest initiai sizes survived better than medium-sized tadpoles, which did better than the smdest ones. The initiai me&um-sized group metamorphosed at the same body size as the initidy large group. Petranka (1989) used manipulations of a seminaturai pond, experiments in field enclosures and laboratory experiments to examine whether chemicai interference competition among tadpoles occurs outside the laboratory. His study focused primarily on larvai Rana sphenocephah. He introduced hatchling tadpoles to a 10 x 19 m pond devoid of tadpoles (initiai density about 416/m2; 1.4/liter). He assayed local crowding effeas by growing hatchling tadpoles in water from locdy crowded and uncrowded areas of the pond. Early in the hjstory of the population, when local densities were high, water from aggregations significantly inhibited the growth of assay tadpoles. D&e inhibitory effeas were assayed by raising individual tadpoles in situ in enclosures that dorded various degrees of exposure to possible inhibitory substances. These tadpoles showed no evidence for growth inhibition over 10 days of growth. He aiso coliected 13 water samples from 11breeding ponds and assayed them for inhibitory effects on the growth of R. sphenocephah. Two of the samples inhibited Rana growth: one from a pond with high densities of Gastrophlyne carolinensis and Hyla chlysoscelis tadpoles and one i-om a pond with a relatively high biomass/unit area of ovenvintering R. sphenocephala tadpoles. Petranka concluded that chemical interference may be relatively uncommon in naturai populations but may be important &hen aggregation or po6d &ying produces high local densities. Much remains to be learned about chemicai interference; for recent perspectives on t h s subject and the role that the unpigmented alga Prototheca might play, see papers by G. C. Baker and Beebee (1997), Beebee (1995), R. A. Griffiths (1995), and Petranka (1995). Field enclosures were used by Skeliy (1991) to demonstrate that anti~redator behavior bv larvai Hvh venicohr has a cost in reduced growth rate. He placed larvai Ambystoma t&&rinum mesh enclosures inside pens in a pond. Hyh verin sicohr larvae in pens with Ambystoma larvae grew more slowly than thos in pens with empty predator-enclosures. Laboratory experiments showed that the overd activity levels of tadpoles decreased when predators were present. Tadpoles of many species decrease activity when predators are detected (Hokit and Blaustein 1995; Lefcort 1996, 1998). Changes in activity levels are often accompanied by shifts in microhabitat use (Anholt et ai. 1996; Hews 1995; Kupferberg 1998; Manteifel 1995). Antipredator responses can be depend upon the size of the tadpole (Anholt et ai. 1996; Bridges and Gutzke 1997), the relatedness of tadpoles in groups (Hokit and Blaustein 1995), and the previous diet of the ~redator (Laurda et ai. 1997). Plastic bags floating in ponds were used as field enclosues by R. H . Kaplan (1992) for experiments that demonstrated that changes in maternal investment as reflected by egg size can aiter hatchiing morphology and vulnerability to predation in Bombina orientalis.

254

ROSS A. ALFORD

Field enclosure experiments offer the benefits of in- Four replicates were destructively sampled on days 5 and 6 creased control but ofien suffer from high variability among of the experiment, and those that remained were destmcrepiicates. The use of artificiay assembled experimental tively sampled on day 30. In ponds with no predators, an communities is the next leve1 of increased control and de- average of 90% of Hyla tadpoles survived the first 5-6 days creased realism. Morin (1981, 1986) first used this tech- of the larval period, and 43% survived to day 30. When N. nique to study tadpole ecology. The artificial ponds used in viridesmns only were present 21% of Hyh survived to day his studies, like most others who have employed t h s ap- 516, and none survived to day 30. Predation by C. bartonii proach, are cattle watering tanks that contain about 1,000 foiiowed a different pattern; 78% of Hyla exposed oniy to iiters of water. Tanks usuav are covered with iids made from crayfish survived to day 516, and 14% survived to day 30. screening to prevent colonization by aquatic insects and in- When both predators were present, an average of 29% of sect larvae. In a typical experiment, tanks are fiied with wa- Hyh survived to day 516 and 1% survived to day 30. Despite ter one to two weeks before hatchling tadpoles are added. their lower density, Hyh tadpoles exposed to either predator Leaf iitter, supplementary nutrients, and a standardized set or to both together were smaer on day 516 than tadpoles ofzooplankton and phytoplankton are added sometime after not exposed to predators. Tadpoles survived to day 30 in the ponds are fiiied. This technique establishes a set of ex- reasonable numbers oniy when predators were absent or C. tremely similar but totay independent communities that bartonii was the only predator; data (table 1of Fauth 1990) can be manipuiated by selective additions or removals of indicate that mean tadpole size at day 30 in these treatrnents larval anurans and other species. The advantages and draw- did not differ significantly. These results can be explained by backs of this approach to experimental ecology are discussed antipredator behavior by H. chiysoscelis tadpoles. Presence of by Hairston (1989), R. G. Jaeger and Was (1989), Morin either predator elicited avoidance behavior that reduced the growth rates of tadpoles in the first days of life. Avoidance (1989), and Wdbur (1989). Morin (1986) examined the effects of intraspecific com- behavior elicited by the presence of both predators simuitapetition and predation by the newt Notophthalmw viridescens neously may have reduced mortality caused by either predaon the growth, larval period, and survival of PseudanZs cm- tor more than the avoidance elicited by oniy one species of cifer tadpoles. Three densities of hatchling tadpoles (31,62, predator. Fauth concluded that the effects of the predators and 124/m2; 0.1,0.2, and 0.4/liter) were combined with 0, on the growth and survival ofH. chiysosceliswere not additive 2, or 4 addt newts per pond. Oniy newt density dected and that their final effects at the end of the larval period survival of tadpoles; it deciined from 73% with newts absent couid not be predicted from short-term studies near &e beto 61% and 47% at medium and high newt densities, respec- g i ~ i n of the iarvd period. g tively. The growth rate, length of larval period, and mass at The artificial pond technique aows for controiied experimetamorphosis of tadpoles were primarily controiied by ini- ments under seminaturai conditions but stdl leaves some tial tadpole density; tadpoles at hgher densities grew more variation uncontroiied. For example, because artificial ponds slowly and emerged at s m d e r sizes after longer larval peri- are largely self-sustainingsystems, most tadpoles feed on priods. Tadpoles avoided newts by switching from foraging in mary production that occurs in the ponds. Rates of primary exposed sites to foraging in hidden, benthic sites. This ap- production, and thus of food supply, are likely to difer beparently did not affect their growth rates. tween ponds given different experimental treatments. AddiArtificial ponds were used by A. H . Roth and Jackson tionay, unbiased intermediate censuses and measurements (1987) to examine the effects of pond size on survival and are difticult to conduct; artificial ponds present the same predation rates of larval Hyh cinevea. They established pools sampling problems that arise in natural ponds. It is difficult of about 0.09, 1.5, and 3 m2 and aowed insects to colonize to foliow individuals, because that wouid require that they some of them for 3 months. Over the remainder of the vear. be individudy marked. Fine control of factors like tempera, they performed 3 experiments in which larval H. cinevea ture and photoperiod is also difficuit. To look in more detail were stocked at a constant density of 43/m2 (0.43/liter) and at one or a few controiied factors, many studies have been foiiowed survival of the tadpoles through metamorphosis. conducted under laboratorv conditions. Most predation occurred in the first week of larval life, and Early studies of tadpofe popuiations in the laboratory predation rates increased significantly as pool size increased. usuay were concerned with examining factors controlhg They suggested that greater predation pressure in large pools growth and differentiation rates from a mechanistic rather might be one factor causing selection of sma pools as repro- than an ecological perspective. R. M. Savage (1952) obductive sites. served that large Bufo tadpoles behavioray kept smaer inFauth (1990) manipuiated the occurrence of aduit No- dividual~away from preferred food sources. H e suggested tophthalmus vindes~ns the crayiish Cambams bartonii in that this might lead to inhibition of the growth of smaer and 375-liter artificial ponds and examined the effects on the individuals. Adolph (1931) showed that growth rates of at growth and development of larvai Hyh chiysos~lis a den- Ranapipiens and R. sylaatica declined with increasing density sity of 200 individuals per pond (130/m2; 0.53/liter). The and that the growth rates of tadpoles raised aione decreased experimental treatments included neither predator present, as the size of the rearing container decreased (see Guyetant oniy N.viridescens present (two per pond), oniy C. bartonii 1977a, b; Hourdry and Guyetant 1979; Lametschwandtner present (two per pond), and two predators of each species et ai. 1982: Streb 1967). H e attributed the effects of crowdpresent per pond. Each treatrnent was replicated eight times. ing to "stress," which increased metabolic rate and decreased

ECOLOGY

255

growth efficiency of tadpoles. An experiment by Lynn and Edelman (1936) suggested that stress caused by physical encounters between individuals might decrease growth rates under crowded condtions. They reared Rana sylvatica tadpoles in an aquarium separated into a series of compartments by mesh screen that allowed water circulation and vision between compartments. They used a wide range of densities (7.8-12 1.O/hter) and supplied food ad libitum. Tadpoles in the more crowded compartments grew more slowly, experienced higher rates of mortality, and metamorphosed later than tadpoles in less crowded compartments. The feeding regimen eliminated food as a possible cause of these effects. Water circulation and visual communication between compartments should have reduced any effects of hfisible substances or visual disturbance.The effects of water-borne substances may not have been eliminated, as after initial mixing the circulation of any substances produced would have been by I f h i o n , and appreciable concentration gradents could be maintained. Rugh (1934) reached a similar conclusion with a similar experiment on R. pipiens. Grornko et al. (1973) suggested that reduced growth in larval R. pipiens reared under crowded conditions might be caused by adrenocortical responses to social stress. Although perhaps a different phenomenon is involved, Downie (1994a, b) investigated the developmental arrest ofleptodactylus tadpoles in a foam nest caused by the failure of the nest to be inundated. John and Fenster (1975) raised R. pipiens tadpoles in aquaria containing 20 inlviduals (3.51liter). Four aquaria were bare, and four contained partitions that partially subdivided them into 10 connected chambers. Mean body volume at metamorphosis in the undivided aquaria (1.88 cm3) was significantly lower than mean size in divided aquaria (2.34 cm3).They suggested that this was caused by stress induced when animals come into physical contact with others and that spatial complexity might be important in controlling tadpole growth rates. Increasing spatial complexity would decrease encounter rates and thus the effects of stress. John and Fusaro (1981) showed that the size of growth chambers could dramatically affect growth rates of larval R. pipiens fed ad libitum; species that normally inhabit small volumes of water (e.g., Caldwell 1993) need to be studied in this regard. Other work on growth rates of individuals in laboratory populations has concentrated on biological or chemical factors that may inhibit tadpole growth. These mechanisms fall into three general categories: simple fouling of water with metabolic wastes, production of or infestation with parasites that increases with density, and production of inhibitory chemical substances. The first mechanism usually is accepted as potentially important in every study but is seldom measured in detail. Some studies have eliminated fouling by using flow-through systems. Most studies reduce fouling by changing all or part of the water in rearing containers on a regular schedule. Several authors have suggested that growth is idubited by living cells or parasites. Water conltioned by large R. pipiens tadpoles reduced the growth rates of other R. pipiens (C. M. Richards 1958). Richards concluded that the inhibi-

tory effects of conditioned water were caused by cells of an unknown nature that were present in the feces of the tadpoles that conditioned the water. One large Ranapipiens tadpole growing in 3 liters of water, replaced daily,codd prevent smaller R. pipiens tadpoles in the same container from growing (S. M. Rose 1959). This suggested that tadpoles whch gain a size advantage early in growth can maintain and enlarge that advantage through i h b i t i o n of slower growing individuals. In a later study, S. M. Rose (1960) cultured groups of 14, 16, 37, and 53 tadpoles in 12 liters of water %thAfoodprovided ad libitum. he mean mass attained bv all tadpoles in a container declined as density increased, but the 14 largest indviduals from each container had the same mean mass at all densities. This suggests that ithbitory relations between large and small individuals are asymmetrical. Several other experiments by S. M. Rose produced similar results. Water condtioned by large tadpoles inhibited the growth of small tadpoles even after the large animals were removed. This ithbitory effect was removed by heating the water to 60C or greater. Rose d d not present detailed evidence regardmg the nature of the inhibitory substance but agreed with C. M. Richards (1958) that it was cellular. Nakata et al. (1982) proposed a mathematical model of tadpole growth Inhibition that explains results obtained by Rose in terms of growth rates and inhibitor substance concentrations. Steinwascher (1978b, 1979b) identified cells interacting with larval Rana clamitans and R. sphevweephala as a yeast, Candida humicola. He demonstrated-that low concentrations of C. hunzicola cells enhanced the growth of all tadpoles, but higher concentrations decreased the growth rates of smaller tadpoles without affecting growth rates of larger individuals. Steinwascher suggested that h s host-parasite interaction might be selectively advantageous to larger tadpoles because it increased their growth rates by reducing the growth rates of smaller individuals and thus their demands on resources. More recently, Pelaz (1987) found cells similar to the ones found by Richards and Rose in water conditioned by larval Rana dibunda. He identified the cells as green algae of the genus Chlmella but was unable to show that they exerted any inhibitory effect on tadpole growth. Beebee (1991) and R. A. Grfiths et al. (1993) reported the results of experiments on interference between anuran larvae that is mediated by a recently described, nonpigmented alga Pratotheca richardsi (A.L.-C. Wong and Beebee 1994); for further discussion of this topic, see recent exchanges among Beebee (1995), Petranka (1995), and R. A. Griffiths (1995). A case for chemical rather than cellular inhibitory substances was made by Alun (1966), who experimentally reared R. pipiens tadpoles with and without access to their own feces to reduce the rate of ingestion of any Inhibitory cells. Both groups of tadpoles grew at the same rate, which suggested that the inhibitory substance was not contained in the feces and thus was not cellular. Alun also assaved the effects of water conltioned by other aquatic species on the growth and mortality rates of R. pipiens tadpoles. Inhibitory effects decreased with increasing phylogenetic distance between R. pipiens and the conditioning species. Akin pro-

256

ROSS A. ALFORD

posed that inhibitors are chemical compounds that act primarily to inhibit the growth of conspecifics with aiielotoxic effeas being largely incidental. This interpretation was contradiaed by Licht (1967), who examined the effeas of water conditioned by Bufo woodbousii and R.pipiens tadpoles on the growth of many other species. Licht found no evidence for decreasing effects with increasing phylogenetic distance and attributed the inhibitorv effects of conditioned water to oarasitic cels shed in the feces of tadpoles. To summarize, growth inhibition is common in laboratory populations of tadpoles. It can have several causes, including exchange of parasitic organisms and release of chemical compounds. It probably does not occur on a large spatial scale in natural populations but may be important where tadpoles are locaiiy dense, in hatchling aggregations, and in habitats such as smaii pools and larger pools in the final stages of drying. Manv recent studies have used tadooles as models of the operation of intraspecific competition in general. These usuaiiy have concentrated on exploitation competition where th rate of resource suoolv i; not sufficient i o meet maxi, mum demand. In exploitation competition, individuals prevent others from using the resource by consuming it rather than by defending it. Many studies have concentrated on competition among individuals reared at a variety of densities on identical food suppiies. This varies the rate of food supply per individual and hence the degree of exploitation competition. Unfortunately, without controi treatments that vary density while holding rate of supply per individual constant, such experiments cannot distinguish effects caused by resource limitation from effects caused by interference. In general, interference competition, which may be caused by any of the inhibitory effects just discussed, can increase the e&s of density observed & such experiments. Wiibur (1977a) investigated the effeas of density on laboratory populations of larvai Bufi americanus. The variance of bodv mass arnong individuals reared in isolation and fed u ad libitum increased as they grew. There was no correlation between the rates of growth and development of individuals. Other Rufo were raised at densities of 10, 20, 40, 80, 160, and 320 per 0.08 m2 in 2-liter pans with the same ration of food supplied to aii densities. Instantaneous mortality rates were not related to initial density, and larval growth and development rates deciined with increasing density. This led to the apparently paradoxical result that despite densityindependent instantaneous mortality rates, the probability of survival to metamorphosis was strongly correlated to initial density. At high initial density, low growth and differentiation rates mean that individuals are exposed to equivalent instantaneous chances of mortality for ionger perids, and thus a greater proportion of individuals die before metamorphosis. Dash and Hota (1980, Hoplobatracbus therinus) and A. Sokol (1984, Litoria ewin~z) found similar patterns in comparable experiments. In order to separate the effeas of exploitative and interference competition, Wibur (197%) raised larvai Rana sylvatica in the laboratory at 1 , 2 , 4 , and 810.0110 mZ(250-ml dish) and 1, 2, 4, and 8X then 6X multiples of a ration of
L L

ground rabbit food. This aiiowed a set of comparisons of growth and development rates between tadpoles raised at equivalent rates of food supply per individual over a range of population densities. If the effeas of density were caused o& by exploitation, populations given the sa&e rate of supply per individual should have grown and developed at the same rate. Figure 10.4 summarizes his results for mean body mass at day 29-30 at densities of 1,2, and 4 individuals and feeding rates of 1, 2, and 4X (highest density and feeding regime omitted because the highest food level changed during the experiment). His results are difficult to interpret. When a single individual was raised alone, its mass deciined with increasing food level. At density 2, mass increased when food went from 1 to 2X. then deciined. Onlv at density 4 did mass increase consistently with increasing food level. At density 2, mass increased when food went from 1 to 2X and then declined. Only at density 4 did mass increase consistentlv with increasing-food level. The broken lines " connecting treatments with equivalent rates of food supply per individual suggest that there was no consistent effect of density independent of food supply. The hypothesis most consistent with these results is that at lower densities and higher feeding rates, growth was depressed by f o u h g effects caused by excess food. A series of papers by Steinwascher (1978a, b, 1979a, b) explored the ecological impiications of competition among tadpoles in laboratory systems. H e found that tadpoles can shift feeding modes as resource avaiiability changes, that larger individuals may monopolize higher quality resources, mav and that some of the effects of interference com~etition actuaiiy be caused by host-parasite interactions involving a yeast, Candida bumicola. Steinwascher (1978a) demonstrated that aiiowing access t o feces increased the growth rate of larval Rana catesbezana. H e suggested that the primary function of coprophagy was to increase the length of time food was resident in the digestive traa and that it might also d o w some microbial digestion. The food value of feces was u lower than the value of a similar quantity of original food, but filtering feces from suspension required less energy than scraping soiid food from a substrate. When he fed R. catesbeiana solid food at low rates, larger individuals fed primariiy by filtration of feces and forced smaiier individuals to expend additional energy in scraping food from the substrate. Semlitsch and Caldwell (1982) examined the effects of density on the growth and development of larval Scapbwpw bolbrookii. In one experiment, total food ration was held constant, while individual density varied from 3 t o 30 individuals per container (1.2-12.01liter; 0.01-0.1/cm2). This experiment produced a complex set of results; growth rate decreased and length of larval period increased as density increased. Ln combination, these two trends led a reduction in mass at metamorphosis as density increased from 1.O t o 4.8 individuals per liter and then increased with increasing density. This increase was caused by developmental rate slowing faster than growth rate as density increased. They suggested that this was caused bv earlv metamorohosis at s m d sizes by faster growing individuals in the high-density treatments, which allowed remaining individuals to take advantage of
i

ECOLOGY

257

the increase in resource avadability by growing more rapidly while reducing their rates of development. The remaining individuals emerged at large sizes after an extended larval period as proposed by Wilbur and Collins (1973). This experiment did not separate effects caused by exploitation from effects of interference. A complex experiment performed by Berven and Chadra (1988) evaluated the influentes of egg size, densis; and food level on tadpoles of Rana sylvatica. They raised Rana tadpoles hatching from large and s m d eggs at densities of 2, 4, or 8 individuals per pan (size unspecified). These egg size and density treatrnents were completely crossed with three food supply rates. The absolute quantities of food supplied were altered over the experiment but remained in the ratio 1:2:4. They found that size at metamorphosis decreased and length of larval period increased with decreasing food per tadpole. They found no effect of density independent of food level; interference competition was apparently minimal or absent. It is difficult to interpret this lack of interference competition because they presented no information about the absolute densities per unit area or volume used in their experiment. Egg size dected metamorphic parameters only at higher food levels and lower initial densities. Kehr (1989) experimentdy manipulated food level and density of Hylapulchella tadpoles in unreplicated aquaria. The tadpoles exhibited effects of both interference and exploitation competition; at constant low feedmg rates per animal, larval period increased with increasing density while body length

remained constant. When feedmg rate per individual decreased with increasing density, length at metamorphosis decreased and larval period increased. I recently conduged a laboratory experiment designed to separate the effects of interference and exploitation competition on the growth of larval Limnodymxites ornatus. I raised tadpoles at densities of 1, 2, and 4 individuals in 10 x 7 cm-circular plastic containers holding 250 rnl of water and completely crossed these densities with feedmg rates of 1,2, and 4X multiples of a basic food ration consisting of a 1:3 mixture by weight of finely ground tropical fish flakes and alfalfa pellets. I weighed d tadpoles in the experiment on day 10 of development (fig. 10.5).The design of this experiment was similar to that of Wilbur (1977b), but its outcome was different. In L. ornatus the effeas of interference and exploitation competition appear to be nearly equivalent, particularly at lower densities. For example, the performance of isolated individuais raised at a 1X ration is nearlv identical to the performance of individuals raised in pairs with a ration of 4X. The effects of interference competition increased less when density increased from 2 to 4 individuais per container. These results suggest that the intensity of interference competition may increase noniinearly with density and perhaps reaches an asymptote beyond which adding individuals does not increase the intensitv of com~etition. This could be true if interference were caused by stress, and the asymptote would be the point at which hormonal systems were maxim d y perturbed by disturbance. This could also be true if

2X

FEEDING RATE
Fig. 10.4. Responses in body mass at day 30 of lawae ofRanasylvatica raised at three densities (1,2, and 4 per container) and three food levels (lX, 2X, and 4X). Dashed lines connect means of treaunents expected to have the same mean mass ifinterference competition were absent, and solid lines connect means of replicates raised at the same density. Data from Wiibur (197%).

ROSS A. ALFORD

FEEDING RATE
Fig. 10.5. Responses of body mass at day 1 1 of larvae of Limm~dyrz~esomatus raised at three densities (1, 2, and 4 per contwier) and three food levels (lX, 2X, and 4X). Dashed lines connea means of ueatments expeaed t o have the same mean mass ifinterference competition were absent, and soid lines connect means of replicates raised at the same density. Vertical lines are Gabriel simultaneous 95% comparison limits for means.

interferente was caused by physicai interactions bctween in- egg size; these effects were of about the same magnitude as dividuais, and a maximum of aggressive encounters might the effects that were related to egg size. Newrnan (1988a, b) investigated genetic and environbe reached at intermediate dcnsities bevond which individual activity could decline. I obtained similar results in an mental infiuences on the growth and development of Scaphiexperiment with larvai Bufo marinus (Aiford 1994). Other opus couchii. In a laboratory experiment, Newman (1988b) rate laboratory experiments that have examined intraspecific demonstrated that develo~mental has a h e h heritabiiitv competition include Hota and Dash (1981), Mahapatro and and that the ranking of pirformance of families can depend Dash (1987), Mishra and Dash (1984), and Murray (1990). on the competitive regime. In another set of experiments, Parichv and Ka~lan 11992b) examined effects of maternal Newman 11988a) showed that five d sibshi~s differed in investrnent on growth and survivai of larvai Bombina orien- developmental rate in the laboratory. He raised the same sibtalis in the laboratory and found that tadpoles from smaiier ships in experimental ponds with long and short durations eggs performed worse than those from larger eggs in low- and showed that similar differences ~ersisted:the sibshi~ quality environments, but egg size had no consistent effects with the slowest rate of development experienced greater in higher quality environrnents. They suggested that these mortality in short-duration ponds but reached greater sizes results are consistent with the hypothesis that frogs produce in long-duration ponds. Blouin (1992a) raised tadpoles at different food levcls eggs of a variety of sizes as a form of "bet hedging." In a second paper (Parichy and Kaplan 1995), they showed that (Hyhcinmea and H.gatwsa) and temperatures (H. cinerea, larvae from larger eggs have higher sprint swimrning speeds H. ~ratiosa,and H. squirella). H c compared species mean and may thus be more capable of avoiding predation. In norms of reaction of mass at metamorphosis and lenmh of " contrast, Tejedo (1992) found no trade-off between egg size larvai period across levels of food and temperature and and number in Bufo cahmita and suggested that increases in found that the mean reaction norms of H. cinerea and H. both the size and number of eggs with increasing femalc ~ratwsawere paraiiel across levels of food. Evidence of body size must be caused by some other factor, perhaps differences in slope suggested species-specific differences in "phylogenetic constraints." Egg size may not aiways be re- responses to temperature variation. Blouin suggested that lated in a sirnple way to ecological properties of tadpoles; selection may have favored different plastic responses to Wfiamson arid Bu (1989) found that variation in egg size temperature in the three species. Blouin (1992b) showed within clutches of Crinia sgnifera was related to size at that some aspects of metamorph morphology in H. cinerea hatching but not to time to-hitching or feeding. Clutches are geneticdy correlated with larva1 life history characters, with larger average egg sizes both developed more rapidly so that selection on length of larvai period, for exarnple, and produced larger hatchlings. They also found effects of might also d e c t size-adjusted head width at metamorphosis. clutch on size and hatching time that were unrelated to mean Breden and Keiiy (1982) were interested in the extent to
I
\

ECOLOGY

259

which social interaction affected developmental synchrony in Bufi americanus tadpoles. They hypothesized that the aggregative behavior often shown by B. americanus might faciiitate metamorphic synchrony and perhaps aow them to swamp terrestrial predators upon emergente. They held the rate of food supply and the density of tadpoles per unit volume constant. Tadpoles were raised in isolation in individual compartments in aquaria that aowed visual and chemical but not physical interaction and in open aquaria that d o w e d fu interactions. They found that tadpoles raised alone metamorphosed at the smaest sizes, partidy interaaing tadpoles metamorphosed at intermediate sizes, and fully interaaing tadpoles metamorphosed at the largest sizes. Tadpoles in the partiay interacting treatment had the shortest mean length of larval period (33.8 days), followed by the f d y interacting (36.6 days) and the isolated treatment (37.3 days). Variance of larval period among repiicates was greatest in the fuy interactive followed by the partiay interactive and isolated treatments. This is the reverse of the order predicted by their hypothesis. They suggested that their results were caused by competition; faster growing tadpoles inhibited the growth of smaer individuals until the larger animals metamorphosed, which aowed smaer individuals to grow more rapidly later in their larval periods. If faster growing individuals tend to delay metamorphosis until they reach larger sizes, this might explain their resuits. The hypothesis that female mate choice shouid lead to superior larval performance was evaluated by Woodward (1987d; see Woodward and Travis 1991). He captured 12 female and 16 male Scaphwpus couchii and mated each female with a large and a sma male. Withm females, tadpoles sired by larger males survived to day 34 of the experiment at a higher rate than tadpoles sired by smaer males. N o other larval or metamorph characters were dected by paternal size; there was no evidence that females wouid have chosen large males if aowed their choice of mates, so his resuits must be interpreted cautiously. Woodward et al. (1988) carried out a more elaborate set of experiments that aowed females to mate with the male they had amplexed with in nature. Six female Pseudacris ci.ltcifer were captured in amplexus. Each female was mated with her original mate, a male which had entered amplexus with another female, one unmated male ("bachelor") from the upper haif of the male body size distribution, and one bachelor from the lower haif of this distribution. Tadpoles were raised individudy. Offspring of naturay mated males grew slightly but significantly faster than offspring of bachelors. Offspring of large bachelors grew significantly faster than offspring of s m d bachelors. Male attributes other than body size had stronger effects on growth rate. They concluded that under some ecological conditions, females may enhance their reproductive success by choosing larger males as mates. ) Woodward ( 1 9 8 7 ~ investigated the possible effects of interactions benveen cohorts of larval Bufi woodhousii in laboratoty containers. H e raised hatchlings in 1.5-liter containers alone, in pairs, or in pairs involving tadpoles from a cohort about 20 days older. Food supply rate did not appear to be iimiting for hatchlings raised aione or in pairs; both

treatments foilowed indistinguishable growth trajectories. Tadpoles reared with larger individuais grew more slowly size and reached minimum metamor~hic later than tad~oles reared alone or with members of the same cohort. The presence of a larger tadpole also reduced survivorship of hatchlings. This suggested that females reproducing later in the season face selective disadvantages caused by competition from earlier cohorts. How does intraspecific competition among close relatives compare with competition among u ~ e i a t e d individuds? In general, simple ecological theoty wouid suggest that the intensity of exploitative competition shouid increase with increasing relatedness among competitors because closely related individuals shouid have more nearlv identical niches than distantly related individuals. Interference competition through aelopathy might be less intense arnong relatives than nonrelatives because an effective delotoxin shouid not negatively d e a the individual producing it and thus might be expected to have weaker effects on close relatives of the producer. At least five studies have examined some aspect of this question in tadpoles. Shvarts and Pyastolova (1970) found that larval Rana amalis grew better when raised in mixed-sibship groups than when raised in single-sibship groups. Travis ( 1980b) included a mixed-sibship treatment in an experiment examining interspecific competition between Hylafemoralis and H.gratwsa (also see Goiiman and G o h a n 1993b; Semlitsch and Schmiedehausen 1994). Jasienski (1988) examined competition within and among sibships of Bombina variegata. Tadpoles from eight sibships were raised in two experimental treatments: groups of eight individuals from one'sibshi~with one re~iicate sibshiv ver and eight mixed groups of eight individuals with one from each sibship. Density was 50/iiter, and food was supplied ad iibiturn. This feeding regimen shouid resuit in any competition being caused by interference rather than exploitation. Mean body mass at day 43 of single-sibship groups was greater than mixed groups. Variation in body mass was Luch greater withn &ed groups than w&n singlesibship groups. The largest individuals in mixed groups were larger, and the smdest individuals were smder than in single-sibship groups. Jasienski's resuits are consistent with the hypothesis that interference competition occurred in his beakers and that it acted less strongly among sibs than among nonsibs. D.C. Smith (1986, 1990) experimentally manipulated the genetic composition of poplations of~seudacris triseriata in sma natural pools. He found that in s m d ponds with high initial densities, groups of animals from Single ~ i b s h i ~ s - ~ rmore quicklfand survived to metaew morphosis at higher rates than mixed-sibship groups and suggested that these resuits were probably caused by relatedness-mediated interference. Hokit and Blaustein f 1997) showed that the mass ofRana mcaae larvae reared in mixed groups was skewed toward smder body sizes compared to tadpoles reared in single sibships, whch suggests that competition may have been more intense. In summaq the experiments reported to date suggest that interference competition tends to be weaker withn groups of siblings than withn groups of unrelated individuA

260

ROSS A. ALFORD

als. It is not clear whether exploitation competition among siblings is stronger than exploitation competition among nonrelatives. This should be explored further with experiments designed to examine bothAtypes competition s k u l of taneously. Three additional topics on the ecology of tadpole growth and development deserve mention: responses to environmental temperature, dissolved oxygen anh pH (see chap. 8). Three general themes are apparent in the literature on responses to temperature and dissolved oxygen. One set of studies (e.g., Floyd 1985; Harkey and Semlitsch 1988; Herreid and Kinney 1967; Licht 1971; Zweifel 1977) examined the etiects of temperature on growth and development and related these to conditions encountered in nature; thermal tolerances as limitations on the reproductive biology and range of specics are emphasized. Another group of studies examines how tadpoles behaviorally respond to small-scale thermal or oxygen concentration variation by orienting themselves in gradients (e.g., Beiswenger and Test 1966; Bradford 1984b; Lucas and Reynolds 1967; Noland and Ultsch 1981; Wassersug and Seibert 1975). The third set of studies iooked at how thermal preferenda change over time through acclimation (Hutchison and Hill 1978; E. Marshall and Grigg 1980a, b) or during development (Duprk and Petranka 1985; Wollmuth and Crawshaw 1988; Wollmuth ct al. 19871. A of these studies are usem as examinations U of the etieas of thermal and oxygen constraints on tadpole growth and development. The etiects of environmental pH on growth and development of tadpoles are of interest becauseof the natural acid& of some wetlands (Gosner and Black 1957b), and because acidification by human activities may have a major impact on frog populations (e.g., Henrickson 1990). Recent studies concentrated on North Temperate areas where the problem of acidification is perceived as most severe (e.g., Andrkn et al. 1988; Cumrnins 1989; Gascon and Bider 1985; Padhye and Ghate 1988; Pierce and Harvey 1987; Piercc et al. 1984). Pierce (1985) reviewed acid tolerance studies and concluded that amphibians in general were relatively acid tolerant, with larval mortality usually increasing only below o H 4. Some soecies were vew sensitive and would thus be ~articularly &rable to an&ropogenic acidification. Warner et al. (1991, 1993, discussed below) demonstrated that bealterations in OHhad comolcx eff'ects on com~etition I tween Hylafemoralis and H.8ratwsa tadpoles, and photoperiod (M. L. Wrtght et al. 1990) may act similarly. Changes in environmental acidity may also affect the interactions of species with their predators, sometimes reducing the vulnerability of tadpoles to size-limited predators by decreasing predator growth rates, as Kiesecker (1996) demonstrated for t&viata and the interaction between larvae of Pser~daai~ Ambystumu tz;4rinum. Responses to increased acidity can depend on larval size or developmental stage (Rosenberg and Pierce 1995) so that the etiects of transient increases may depend on their timing relative to the larval life span of affected spccies.

Interactions with Predators

Field data on survival rates of tadpoles suggest that predation is the major source of mortality. Predation begins at the egg stage (Villa et al. 1982) and continues through metamorphosis (S. 1. Arnold and Wassersug 1978; Wassersug and Sperry 1977). The presence of predators has been suggested as a limiting factor in pond use by many anurans (Bradford 1989; Kats et al. 1988; Sexron and Phillips 1986; Woodward 1983). Predation may affect the outcome of other interspecific interactions in tadpole communit~es (Morin 1981). Tadpoles are vulnerable to a wide range of invertebrateand vertebrate predators. Their vulnerability can be influenced by tadpole behavior, body size, coloration, genotype, habitat preferences, and palatability. Wassersug (1971) persuaded several human volunteers to taste eight species of tadpoles from Costa Rica and documented considerable variation in the palatability of tadpoies. The tail of most species was least disagreeable to the volunteers, the skin second, and the body most disagreeable. Bufo murirtus tadpoles were rated the least palatable. Wassersug suggested that this is in accord with the known toxicity and relatively low predation rate on bufonid tadpoles. This result agrees with experiments performed by myself and students on 3. marinas tadpoles in Queensland, Australia; all specics of freshwater fish we have tested either will not taste them or immediately spit them out after tasting them (personal observation). These tadpoles appear to be distasteful or harn~ful some (e.g., larval l)yttrcus sp. and the odonates to fIemta72axpapue~sis Orth-m and caledonkum) hut not all invertebrate predators; naiads of the cosmopolitan d o n a t e Pastala jkpescess consume them avidly (personal obscrvation). Licht (1969) found that ovarian eggs of four &na species were palatable to larval Ambystoma~racile, Gasterosteus acuEeatus, and Salmo clarkii. Ovarian eggs of Bufo valliqs were toxic to most potential predators (Licht 1968). Walters (1975) documented variation in feeding rates of salarnanders on anuran eggs, hatchlings, and swimming larvae. She suggested that Rasa chitans eggs and hatchhngs were distastefut to salamanders. Kruse and Francis (1977) also suggested that palatability varied among tadpoles of ditierent species. They found that tadpoles of four species were eaten bv three soecies of fish but that larval Rana clamitans were taken on$ by Mimopterus salmoida and only when alternatives were not abundant. They suggested that tadpoles from more permanent habitats should be less palatable. A similar result was obtained by Grubb (1972), who found that Gambwia ate more eggs of temporary-pond frog species than of permanent-pond species. Formanowicz and B r d i e (1982) showed that tadpoles of Bufo a&nus, Pseudks mnyq Rana chmitans, R.palumi, and R. sylaatica in Gosner 38-42 were vulnerable to predation by larvae of the diving beetle Dpbcus ~evticalb. Vulnerability decreased at metamorphosis (Gosner 45-46), probably bccause of the development of glands secreting distasteful substances. Brodie and Forrnanowicz (1987) demonstrated that hatchling (22-25) stages of B. ameP-icanus

ECOLOGY

261

were unpalatable to newts (Notophthalmus viridescens) and limited predators like fish and salamanders. Size limitation sucking (Belostoma sp.) and chewing (Anmjunius) inverte- of predators may mean that tadpoles are exposed to differing brates. Predators exposed to unpalatable hatchlings eventu- predation risks from d ~ e r i n g suites of predators as they ally avoided more palatable intermediate stage (Gosner 30- grow. The ecological implications of size-specfic predator33) tadpoles. Brodie and Formanowicz suggested that this prey interactions were dscussed in detad by Persson (1988), learned avoidance protects later stages from predation with- Polis (1988), and Wilbur (1988). M. K. Moore and Townout continued production of defensive compounds. Sheratt send (1998) discussed the effects of abiotic factors on these and Harvey (1989) investigated the effects of absolute and interactions. Nearlv all ~redators have been examined in that relative density of tadpoles of Phyllomedusa trinitatir and Phy- the laboratory show some evidence for size-selectivity (table salaemuspustulosus on rates of predation and prey selectivity 10.3). For many predators, decreasing predation rate with by larvae of the odonate Pantalaj-lavescens. The consistent increasing prey size may be caused by a combination of sevpreference of Physalaenzus over Phyllomedusa independent of eral factors. The operation of the mouth or mouthparts may absolute or relative density might reflect a dfference in palat- limit prey taken to sizes less than the maximum gape, faster ability between the species. Denton and Beebee (1991), burst swimming rates may enable large prey to escape more Henrickson (1990), Peterson and Blaustein (1991), and effectively (Huey 1980; S. J. Richards and Bull 1990b), and Readmg (1990) found that Bufo eggs, embryos, or larvae larger and later stages may develop glands that secrete protective substances (Formanowicz and Brodie 1982). Laborawere less palatable than other species. The effect of tail coloration on tadpole predation by odo- tory experiments on prey selection may or may not be realisnate naiads (Anm junius) was investigated by Caldwell tic simulations of nature; prey may be forced into closer (1982). Tadpoles ofAGms crepitans andA.glyllus have black, contact with predators than they would normally allow, and mottled, or clear tads. The clear morphotype was more com- outcomes may depend on exact numbers of prey exposed to mon in permanent water containing fish, while the black tail oredators and on when data are taken. pattern was most common in temporary pools where A.junEncounter rates between tadpoles and predators are ini was the dominant predator. She suggested that this re- fluenced by the behavior of both predator and prey. Tadpole m sulted from natural selection for crypsis in the presence of behavior may change in response to extrinsic cues having no fish and for deflection of predator strikes away from the relation to predatory interactions and may also respond to body when odonates were the dominant predators. The hy- cues related to predation. Some behaviors of tadpoles may pothesis that tadpole tails may be lost to predators with little actually facihtate predation. Ideker (1976) suggested that loss in fitness was supported by an experiment conducted by aggregation of larval Rana berlandieri near shore increased Wilbur and Semlitsch (1990). They experimentally damaged the rate of predation by grackles. Feder (1983~) demonthe tails of Rana catesbeiana and R. sphenocephala tadpoles strated that both swimming and air-breahg increased the and raised them in pond enclosures and in the laboratory. distance from which painted turtles (Chlysemyspicta) would Removing any part of the tail of R. utesbeiana tadpoles attack Rana be~landiem' tadpoles. Moving tadpoles were atslowed their rates of growth and development in field enclo- tacked from &stances as great as 175 cm, whlle the maxisures. Growth rate over a 36-day period was reduced by mum &stance for attack on nonmoving tadpoles was 30 cm. about 33%. A second experiment with R. catesbeiana in Air breathing increased rates of attack on tadpoles but did whlch tadpoles were reared in the laboratory showed no not increase their escaoe abilitv and thus increased their effect of any degree of tail damage on growth or develop- overall rate of loss to turtle predation. When behavior inment. There was no effect of tail damage on survival rates of creases predation risk, this increase must be offset by gains tadpoles in either experiment. Tail damage also did not affect in fitness from other sources. Such trade-offs might prove to survival rates of R. sphenocephala exposed to predation by be a fertile field for M e r investigation. " adult Notophthalmus virt'descens. Figiel and Semlitsch (1991) Many species of tadpoles also show behavioral responses demonstrated that tail damage needed to be severe (75% to predators that decrease their vulnerability to predation. lost) to increase rates of crayfish predation on larval Hyla ~ a d ~ o l ofsAscaphus @mi altered their feeding behavior in e chlysoscelis in 20-liter containers. Semlitsch (1990) also response to nonvisual cues from four predators (Feminella found that 75% tad damage increased the rate of predation and Hawkins 1994). Caldwell (1989) suggested that Hyla by dragonfly naiads (Tmmea lmerta) on H. chlysoscelis larvae. ~eographicatadpoles form schools to reduce individual vulIn some species tail damage may have greater effects. Parichy nerability to predation, and Rodel and Linsenmair (1997) and Kaplan (1992a) showed that Bombina orientalis that demonstrated that swarms of larval Phlynomantis microps hatch from larger eggs are less affected by tail injury, as are form in response to visual or olfactory cues of predator preslarvae raised at higher food levels. Tad injury tended to in- ence. Morin (1986) suggested that Psedumis C~ZGC@Y tadcrease the larval period and decrease survival to and size at poles altered their microhabitat use in response to predation metamorphosis in this species. by Notophthalmus videscens. This change may have been a The influence of body size on tadpole vulnerability to response to chemicals released by newts or other tadpoles predation has received considerable attention. In general, or might have been a reaction to other sensory cues. Some mechanical limitations should cause most predators to pre- chemical cues eliciting avoidance behavior may originate in fer prey within a limited range of body sizes. This is less true the bodies of tadpoles. Hews and Blaustein (1985) demonfor some predators, such as odonate naiads, than for gape- strated that larval Bufo boreas responded to an extract of
2

262

R O S S A. A L F O R D

Table 10.3 Summary of the effects of predator and prey body size on predation rates on tadpoles. C = constant, D = decrease, I = increase, - = not measured. Predator taxa: A = bird, B = beetle adult, b = beetle larva, F = fish, H = hemipteran or heteropteran adult, N = odonate naiad, S = salamander adult, and s = salamander larva. Effect of Prey Size Increase D Db I then Db Db I then Dh Db I then Db D D D D D D D D C C D D D D IorD' I D D Dd D D I then D D Db I then Db Db I then Dh Db I then Dh D D D D D D D Effect of Predator Size Increase

Prey

Predator

Reference Kehr and Schnack 1991 Henrikson 1990 Henrikson 1990 Henrikson 1990 Henrikson 1990 Henrikson 1990 Hellrikson 1990 Heyer et al. 1975 K. L. Lawler and Hero 1997 S. J. Richards and Bull 1990b Semlitsch and Gibbons 1988 Semlitsch 1990 Caldwell et al. 1980 Crump 1984 Crump 1984 Crump 1984 K. L. Lawler and Hero 1997 S. J. Richards and Bull 1990b Gascon 1989b Gascon 1989b Heyer et al. 1975 Hews 1995 Crump and Vaira 1991 Brodie and Formanowicz 1983 Brodie and Formanowicz 1983 Brodie and Formanowicz 1983 S. J. Richards and Bull 1990b Travis et al. 1985b Cronin and Travis 1986 Cronin and Travis 1986 Henrikson 1990 Henrikson 1990 Henrikson 1990 Henrikson 1990 Henrikson 1990 Henrikson 1990 Formanowicz 1986 Brodie and Formanowicz 1983 Brodle and Formanowicz 1983 Brodie and Formanowicz 1983 Caldwell1994 Caldwell1994 Heyer et al. 1975

Bufo arenarum B. bufo

Crinia s&n@a Hyh chryoscelis

H.gratiosa H. pseudopuma

Pleurodema borellii Pseudaois wcifer Pseudupblyne bibronii Rana areolata

R. clanzitam R sylvatica R. warschm'tschii Smiliscaphaeota

H , Belostoma oxyuwm S, Trituwsvubaris B, Dytiscus lapponicus B, Bantus exoletus N, Aeshna sp. N , Leucowhinea dubia H , Notonectaglauca N, Pantahfivescens F, Lates calcamj2r N , Hemicordulia tau F, Lepomis macrocbiws N, T m a lazerta s, Ambystoma talpoideum B, Rhantusguticollis N , Sympetrum n&rocentrum N , Aeshina sp. F, Lates calcanj2r N, Hemicordulia tau F, Pyrrhulina sp. N, Aeshnidae N, Pantalafivescens F, Lebidcinapanamensis A, Pitanqus sulphuratus b, Dytiscus verticalis s, Ambystomajeffersonianurn H , Lethocerus americanus N, Hemicordulia tau N , T m a lacerta H, Notonecta i&a H , Notonecta undulata S, Triturw vubaris N, Dytiscus lapponicus B, Bantus exoletus N, Aeschna sp. N , Leucorrbinea dubia H , Notonectaglauca B, Dyticus verticalis N , Anax junzus b, Dytiscus verticalis H , Belostoma sp. N, Libelluh hercztlea N , Megaluprepus caeruhtus N , Pantalaflavescens

*Increasingpredation concentrated on larger prey bstatistical significance not determined. cDependedon microhabitat. dNot statistically significant, P > 0.05.

chemicals from injured conspecifics by increasing their over- iads of Aeshna umbrosa. Other chemical cues may originate al activity level and avoiding areas containing the substance. in predators. Petranka et al. (1987) demonstrated that Hyla l Bufo boreas tadpoles captured by the water bug Lethocems ch~soscelis tadpoles exposed to water conditioned by the fish americanus release chemicals that act as alarm substances Leponzis cyanellus spent more time in refuges than tadpoles in (Hews 1988). Other B. boreas tadpoles increased activity unconltioned water. Petranka and Hayes (1998) showed rates and avoided areas where tadpoles had been injured. that larval Buj%americanus and Rana sylvatica responded to Bufo tadpoles l d not respond to feeding by L. americanus water condtioned by the predatory odonate Anax junius by on tadpoles of other species. In another experiment, Hews reducing activity levels and moving away from the source (1988) showed that thls avoidance behavior was effective at of the odor. Larval B. americanus responded identically to decreasing the vulnerability of B. boreas to predation by na- conspecific tissue extracts, whde R. sylvatica larvae ignored

264

ROSS A. ALFORD

cies and B. americanus were unpalatable. The only species that regularly occur in permanent ponds and are palatable were A. glyllus and H. chlysoscelis. None of the temporarypond species or the unpalatable species spent significantly more time in refuges when exposed to water conditioned by fish. Tadpoles ofA.gryllusspent 86% of their time in refuges when fish were absent and 91% when fish-conditioned water was present, but the increase was not statistically significant. Larval H. chlysoscelis significantly increased time spent in refuges from 49% when water was not fish-conditioned to 82% when fish chemicals were present. Kats et al. (1988) suggested that the patterns were consistent with evolution of defenses against fish predation by species regularly exposed to it in nature. They also suggested that smd-scale geographic comparisons between popuiations exposed to differing predators might provide interesting information on the evolution of defenses in tadooles. I demonstrated that genetic variation for factors affecting vulnerability to predation exists in Hyla chlysoscelis (Alford 1986b). I reared 4 sibships of tadpoles, the offspring of a factorial cross of 2 males with 2 females. for 23 davs. I then exposed groups of 9 tadpoles to predation by male and female broken-lined newts (Notophthalmusv. dorsalis) in 24 x 30 x 12 cm plastic trays. Male and female parents had significant effects on tadpole growth at day 23. Male parent, and the three-way interaction of male parent, female parent, and newt sex had significant effects on the numbers of tadpoles remaining in the trays after 30, 60, and 120 min of exposure to predation. The least vulnerable families were not those with the largest bodies; this suggested that some geneticdv determined factor other than size had affected medation ;ate. This couid have been a difference in behkior or in levels of distastefd secretions. Semlitsch and Gibbons (1988) induced bodv size variation in larval H. chmsoscelis by rearing five sibships at three densities. When these were exposed to predation by bluegill sunfish (Lepomis mamochias), both tadpole body size and sibship controlied vulnerability to predation. In contrast, Semiitsch (1990) found no effects of sibship on the vulnerability of H. chlysoscelis tadpoles to predation by larval Tramea lacerta. D. C. Smith and Van Buskirk (1995) demonstrated that competition and predation interacted with the phenotypes of larval Pseudmi mtcifer and l? trileriata in predictable ways. Both species showed some alterations to their phenotyie when &sed in the presence of predators. l?crucifer, which normdy occurs with predators, responded less than l? mieriata, which does not. McColiuni and Leimberger (1997) and McColium and Van Buskirk (1996) showed that larval Hyla chlysoscelts and H. vmicolor raised in the presence of predators developed different morphology and color patterns than those raised alone. The changes were such that predation rates wouid be likely to decrease. Van Buskirk et al. (1997) examined the selective pressures exerted by predatory dragonfly (Anm)naiads on larval Pseudacris seriata and phenotypic responses by tadpoles to that pressure. They found that predation favored larvae with shdow, narrow bodies, high tail fins, and a wide tad muscle. Tadpoles exposed to Anax during development acquired higher tail
--~

fins and wider tail muscles despite not showing greater activity, which suggests that this response may reflect morphological plasticity triggered by exposure to predator cues. Warkentin (1995) demonstrated that late-embryonic &alychnis callidryas respond to snake attack by hatching early, dropping into water, and avoiding the predator. The evolution of tadpole defenses against predation is an area that deserves greater attention in future work. The influente of predation on species interactions among tadpoles wdi be considered later in this chapter.

Composition and Ecology of Species Assemblages

Conzpositwn and Ecology Species composition and habitat use in tadpole assemblages can be looked at in two ways. One approach compares ponds by examining which species co-occur and which do not and the factors that explain patterns of species co-occurrence. The resuits of this approach are often difficult to interpret in terms of tadpole ecology; the spatial distribution of tadpoles among ponds and their temporal patterns of occurrence resuit from the spatial and temporal distribution of reproductive effort by adult frogs, and aduits may respond to many factors in addition to the ecological requirements of their larvae. For exarnple, spawning site use may be constrained by aduit predation risk, predation or other mortaiity risk to eggs, or the quaiity of surrounding habitat for postrnetamorphic juveniles. Any of these factors, or many others, may cause tadpoles to occur in habitats that are not necessarily "optimal" for their growth and development but reflect compromises between the ecological requirements of tadpoles and other stages in the life history. Bradfield (1995) undertook a series of experiments and observations to examine this hypothesis with larvae of two species of rainforest tree frogs. Larval Litmiagenimaculata naturdy occur only in stream pools, and larval L. xanthomera occur in pools on the forest floor. Using transplant experiments, Bradield showed that both species grew faster and survived at higher rates in forest pools than in stream pools. This suggests that L.genzmaculata are constrained to occur in streams by the requirements of other life history stages or by phylogenetic history. The second approach to the study of tadpole assemblages is to examine and attempt to explain the spatial and temporal distribution of tadpoles within ponds. This corresponds to the usual notion of a community study in which ecological relations among two or more co-occurring species are examined. Both approaches depend to some extent on examining patterns of reproductive habitat use by aduit frogs. Tadpoles have little control over the general habitat type (permanent or temporary, still or running water) they occupy. This is determined by the choice of spawning sites by aduits, which determines the range of species with which tadpoles may potentidy interact. The timing of frog reproduction determines the temporal distribution of tadpoles w i h sites. This affects the interactions each cohort of tadpoles experientes with other cohorts of the same species,

ECOLOGY

265

with other species of anurans, and with other vertebrates and invertebrates. The most diverse tadpole faunas occur in wet tropical regions. Detaded studies of spawning habitat use in the tropics include that of Cmmp (1974), who examined the spatiai and temporal use of reproductive habitats by 53 species of aquatic-breedmg anurans in a relatively aseasonai area at Santa Cecilia, Ecuador. She studied 8 breeding sites used by 26 species. The maximurn number of species breeding at any site was 16 and most sites were used by fewer species. Six of the 8 sites she studied were used by 7-13 species. The 2 most similar sites shared 56% of their species. Duellman (1978) further anaiyzed the tadpole communities at Santa Cecilia and found that no more than 10 species occurred at any site simultaneouslv. H e exarnined microhabitat use and iden&ed three broad habitat types: vegetation-choked areas, pelagic areas, and pond bottoms. A maximum of 7 species occurred in any microhabitat. Aichinger (1987) examined temporal and spatial use of seven reproductive sites by frogs in the seasonai wet tropics of Peru. A total of 34 anurans bred at his aquatic sites during 1 year of monitoring. Five of his sites were temporary and dried for long periods during the winter months. Twenty or more species of males caiied at four of the sites, but the maximurn nurnber observed at any time was 12 at a permanent pond site and eight at a temporary puddle site. Most species were reproductively active m a d y or exclusively in the wet season. No species reproduced at aii seven sites. Magnusson and Hero (1991) examined the distribution of tadpoles of 18 species arnong water bodies in a rain forest area in Amazonian Brazil; they found that site use appeared to be dictated by predation pressure exerted by anuran egg predators. Gascon (1991b) examined the phenology and reproductive habitat use of 25 species of frogs at 53 aquatic sites in rain forest in Brazhan Amazonia. He found considerable variation in the degree of reproductive seasonality shown by species and that activity by more seasonal species tended to be strongly alected by rainfaii. H e used discriminant function anaiysis to determine the environmental factors that appeared to be responsible for determining the spatiai distribution of the 11most common tadpoles. A variety of factors were implicated, including the size and depth of water bodies, vegetation cover, and dissolved oxygen. Gascon did not find evidence for discrete structured assemblatzes " of co-occurring species; he concluded that each species responds individuaiiy to the environment in selecting sites for reproduction and subsequent larvai development. Gascon (1991a) examined the detailed responses of Leptoh$us knudreni to seasonal and annuai variation in rainfaii. Moreira and Lima (1991) aiso found that rainfaii was an important factor controlling reproduction and recruitment in f6ur species of Amazonian leaf-litter frogs, and Donnelly and Guyer (1994) reached a similar conclusion for an eight-species assemblage of hylids in Costa Rica. In the seasonaiiy dry tropics, Heyer (1973b) studied the tadpole communities of a series of ponds in Thaiiand. Ten species caiied at these sites and larvae of eight species were found. Six species started caiiing near the onset of the wet

season. wMe two s~ecies delaved for 1-2 months after the onset of heavy rains. Species composition of the larval assemblages in his pools ranged from three to eight and was not correlated in any obvious way with diierences among the pools. He concluded that the species composition of the larvai assemblage at most ponds was largely caused by chance. In a similar study, Heyer et ai. (1975) studied larvai species assemblages in Costa Rican tropical wet forest and concluded that the tirning of oviposition and site selection by many species appeared to rninimize larval exposure to predation. Severai students and I are currently studying the use of reproductive habitats by frogs in seasonaiiy dry eucaiypt woodland in the Australian tropics near Townsvdle, Queensland. We have observed that adult frog density near ponds sometimes decreases temporarily after the first light wetseason rains, probably because frogs disperse to forage. Adult density increases dramaticaiiy, and most species begin &n after the first heaw rains of the wet season. A few species are attracted to very temporary habitats, such as flooded paddocks, creek overflows, and roadside ditches. These species also reproduce at less ephemerai sites such as dams and permanent water holes. For aii but the most ephemerai habitats, the species composition of larval assemblages early in the wet season does not appear to foiiow a strong pattern. The most reproductively catholic species is the introduced Bufo marinus, which reproduces at temporary or permanent water in any month of the year when there is rainfaii. Larvai B. marinus are distasteful to most fishes and to some dragonfly naiads (notably Orthetrum caledonuum, the most abundant species in many habitats) and larvai dytiscid beetles. This low vulnerability to predators may mini. mize selection for s~ecialization site-or season. b; Reproductive tirning and site choice have also been studied in detail in temperate anurans. Detailed analyses of reproductive timing and sites are avadable for many temperate species but fewer entire species assemblages. Some of these are at least as diverse as the seasonai tropical communities; H . M. Wilbur (personai communication) observed 18 species of anurans calling at a single pond in the Sandhillsregion of North Carolina. Most of these species breed in the two months after the first heavy rains in the spring. Utsunomiya et ai. (1983) examined the timing and spatial distribution of reproduction by five species of Rana in a wooded mountain stream in Okinawa, Japan. They observed a greater than normal degree of pattern; none of the species overlapped in reproductive habitat use. Four used mutuaiiy exclusive habitats. The fifih overlapped with one of the first four in habitat but differed in season. Strijbosch (1979) documented reproductive habitat selection by six frogs in fen ~ o o l s The Netherlands. H e found that ranids were the in Lost selective species and bufonids the least selective. Most species avoided very oligotrophic or acidic sites. Thuty-three ~ o n d in desert habitats in New Mexico (Woodward 1983) s had six anuran species present. Four species (one Bufo, three Scaphwpus) occurred primarily in temporary ponds, while two Rana used only permanent ponds. H e suggested that this dichotomy was caused prirnarily by differing vulnerabiliV

266

ROSS A. ALFORD
=

Table 10.4 Records of frogs calling in a seasonal tropical eucalypt w o d a n d near Townsviile, Queensland, Australia. Perm Temp = temporary water.
--

permanent water,

T i g in Wet Season Early Taon Temp Perm Temp MxdLate

Post-

Perm

Temp

Perm

Temp

Perm
X

Bufo marinus c 1i nz ' a desertzcola GYCI~RIUZ brkpes mmhoUandtm Limnudyastes omatzrs t~sm~niensis Litm aibo~uttata biufior rarmlea fdfrw: griuhta

X X

X X

X X X
X X

X
X X

X X X X X X X X X X

X X X X X X X X X X X X X X X X X X X X X

X X X X X

X X X X X X X X X X

inem
aaszrta rothi rubella Upemha lithmda mimula

ties of larvae to predation; the temporary-pond species were more vulnerable to permanent-pond predators than were the avo Xana species. Sernlitsch et al. (1993) examined the facthe reproductive phenology of salamanders tors in South Carolina. Their study is remarkable for the large number of marked individuals used and the availability of data collected over an extended ~eriod. More detailed studies of the cues governing reproductive habitat selection have been performed for many indvidual species. Of 75 rock pools on the west coast of Sweden (Andrin and Nilson 1985), Bufo wtlamitu reproduced in 31 of them. The rate of use by toads was positively correlated with flatness of shorehe, height above sea level, and distance from the sea. Use by toads was negatively correlated with maximum water denth and nool size. The correlation with shoreline flatness may primarily reflect adult toad preferences; flatter shorehes make better calling sites and simplify entrv and exit from the water. The other correlations nrobably primarily reflect larval habitat requirements; increasing &stance from and elevation above sea level reduce water salinity, whlle snlall, shallow pools reach greater temperatures which allow faster growth and development. They suggested that choice of pool size was a compromise between avoidance of drying and attainment of maximum growth and develo~ment rates. Some species appear to be capable of evaluating the state of the aquatic community in small pools and adjusting their oviposition behavior accordingly. Adult HyIa ch?ysoscelk in North Carolina (Resetarits and Wilbur 1989: see 13etranka et al. 1994) discriminated among experimental pools for

calling and oviposition based on the presence of aquatic predators and conlpetitors, including H; chysosceEis belonging to earlier cohorts. The experinlent by Resetarits and Wilbur (1989) was criticized by Ritke and Mumme (1993), who suggested (among other things) that their experimental ponds were too small and close together, and that salamanders in one of their predator treatments may have consumed large numbers of eggs before they cotlld be counted. Even so, R. W. McDiarmid (personal communication) observed a similar pattern of oviposition by Smiliscaphaeota on the Osa Peninsula, Costa Rica, and Hylapseuhpunza also appear to be able to choose oviposition sites based on the condition of the habitat. Crump (1991) showed rhat they are capable of discriminating based on water depth and the presence of potential egg predators. Frogs may select reproductive habitats based on the presence or absence of other frog species. Gascon (l992b) showed that adult Pipa arrabali greatly decrease the survival of eggs and larvae of OsteocephaEw taurinus; he suggested that the species may coexist because 0,taurinus choose to spawn in sites in whichl?arrabati are absent. Pearman and Wilbur (1990) used computer simulations to show that increased patchiness of oviposition may increase population stability. Anurans are not always successful in choosing breeding habitats; Tevis (1966) documented repeated reproductivefailures, most caused by drying of aquatic habitats, in a population of Bu@pu72~tatus, Laurila and Aho and (1997) showed that R m temporuria d ~ d avoid deposa not iting eggs in rock pools containing predatory fish. Most of the studies cited so far concentrated on choices of calling and spawning sites by adult frogs. Characteriistics

ECOLOGY

267

of the chosen sites and spawning phenology affect the ecology of tadpoles by defining whch species of competitors and predators are likely to be present and whether the habitat rnay dry before the end of the larval period. Species that spawn very early in the history of a temporary pool rnay gain an advantage for their larvae from the flush of primary productivity that arises from the high levels of dochthonous nutrients available early in the hstory of such pools. Their larvae rnay also gain size refuges from gape-limitedpredators and gain a size-mediated advantage in competition with other herbivores. Larvae of early spawners in permanent water bodies rnay gain simdar advantages against the larvae of fish and insects (e.g., Majecki and Majecka 1996) that use these sites. Several disadvantages are balanced against the advantages; spawning before temporary pools have completely filied rnay increase the risk of desiccation for tadpoles, and spawning early in the summer season in permanent water rnay lead to slow growth and development caused by low water temperatures. Such slow growth can increase the period of exposure to size-limited predators and decrease sizemediated competitive ability. With such complex possibilities, it is not surprising that frog reproduction follows a variety of phenological patterns. Blair (1961) summarized the reproductive phenology over a four-year period of seven species of anurans that bred at several sites near Austin, Texas. One species bred p r i m d y in winter, four in summer, and two in spring and autumn. Six species cded for more than 1month. The c a h g activity and initiation of first calling of all species was in response to heavy rains. After initial calling, some species continued to call any time that it rained or even with no rain, whiie others lirnited calling activity to periods of heavy rainfai. The sequence of c&ng and spawning by the species varied considerably among years. Temporal separation of caiiing and spawning activity of adult frogs inhabiting a temporary pond in Mexico was documented by J. R. Dixon and Heyer (1968); they suggested that this phenology enforces temporal habitat segregation and resource-use differences in the tadpole assemblage. Seasonal occurrence of tadpoles and salamander larvae was documented at a site in Maryland over a 2-year period by Heyer (1976a). He found that the phenology and density of some species was very similar between years, while considerable pattern changes of other species in the same pond produced large changes in the total species assemblage. The same pattern continued over a further two years of monitoring at the same site (Heyer 1979). Brook (1980) identified three seasonal patterns of c&ng and spawning by frogs in Victoria, Australia: frogs with 4-6 week reproductive seasons occurring at roughly the same time each year regardless of precipitation hstory, frogs with short reproductive seasons (2-3 days) apparently initiated by heavy rainfai, and frogs with extended reproductive seasons. The frogs breeding for 4-6 weeks began c a h g at different times depending on the genus; Litoria, Philoria, and Uperoleia initiated caiiing in early sumrner, while Pseudophryne started in late summer. Eggs or amplectant pairs occurred in more than 1month in 20 of 33 species. Diaz-Paniagua (1986, 1988) studied the reproductive

phenology of seven anuran species at Donaia, Spain, over four years. The heaviest rains usuaiy fell during the autumn and early winter months. Pelobates and Pelodpes had short re~roduc&eseasons associated with heaw kinter rains. Four other species bred over longer periods during early spring, and Ranaperezi had an extended reproductive season during summer. Temporal overlap in pond occupancy of larvae varied considerably among years; in 1978-1979, the heaviest rains arrived in mid-winter, and most species reproduced almost simultaneously. Temporal overlap varied less among the other three years of the study. Timed tape recorders were used (W. Martin 1991) to examine details of the caiing phenology of frogs in seasonaiy dry eucalypt woodland in the Australian tropics near Townsd e , Queensland (table 10.4; see Pearman 1995 and C . L. Rowe and Dunson 1995). He used daily c a h g data to construa simdarity matrices for c&ng patterns among species within sites. Classification analvses of these matrices revealed only two clearly delineated phenologies; Cyclorana brmpes and C. nuvaehollandiae cai early in the wet season in response to heavy rainfai, whde ai other species have extended c a h g seasons with peaks in activity often corresponding to heavy rains. Although the classification analysis did not reveal clear phenological groups within the second category, Mantel tests comparing the similarity matrices among sites showed high degrees of correlation, implying that interspeciic similarities in caiiing pattern remained constant between sites. This suggests that the large group of species with extended calling seasons contains a complex set of subgroups with slightly different phenologies. The longest series of data available on reproductive phenology for a single species (Terhivuo 1988) involves locations and dates of spawning by Rana temporaria in Finland between 1846 and 1986, with an extremely detailed supplemental data set for 1983. Terhivuo found that the date of the onset of spawning correlates well with temperature at any site. Populauons in southwestern Finland (60N) spawn earliest, and spawning proceeds north and east as temperatures increase. The northernrnost populations (70N) spawn an average of 35 days later than the southernmost ones. Unfortunately, data include only the onset of spawning, not its duration. Semlitsch et al. (1996), in their analysis of 16 years of detded data from a natural pond, showed that late spawners rnay sometimes gain a strong advantage: Scaphiopus holbrookii and Gas~r~bhrvne carolinensis were most successfd I when they spawned afier the pond had filied and dried earlier in the season, leading to low densities and diversities of competitors and predators. Larval anuran species assemblages are frequently complex but often less complex than adult species assemblages in the same geographical region. D u e h a n (1978) reports 10 species of tadpoles found in a single habitat simultaneously in an aseasonal tropical region with at least 53 species of aquatic-breedmg anurans, and Hero (1991) found a s d a r number in Brazil. In Madagascar, Blommers-Schlosser and Blommers (1984) found 15 species in one brook, and R. Altig (personal communication) collected 1 3 species within an area of about 2 m2 in a stream in Madagascar. Nearly as
i I

268

ROSS A. ALFORD

many species have been found simuitaneously in seasonal tropical and temperate habitats with much lower aduit anuran species richness. This apparent low limit on the nurnber of simuitaneously occurring species may be caused in part by competition. Temporary pools are relatively unpredictable in space and time and contain a relatively unpredictable set of competitors and predators. This may limit the degree of specialization in resource use attainable by tadpoles. Because most tadpoles are relatively unspeciahzed filter-feeders, most species in a pond may be potential competitors. Differentiation of site selection and reproductive phenology by aduit frogs may serve to decrease the levels of competition experienced by larvae to tolerable levels.

Density and Resouvce Use Relatively few studies describe or explain the composition of tadpole species assemblages in the field. Although R. M. Savage (1952) did not set out to produce a community study, the detailed studies of the autecology of two species in the same ponds can be interpreted as one. Heyer (1973b, 1974) examined species distributions among and within ponds at a seasonai tropical location in Thaiiand, examined gut contents and relative gut lengths (Heyer 1973b), and calcuiated niche breadths and overlaps (Heyer 1974). His quantitative microhabitat use and niche measurements are based entirely on species occurrence in sweep net samples taken as an indicator of rnicrohabitat usage. H e found a broad range of overlap vaiues, as might be expected from a community containing 8 species and thus 28 possible species pairs. He examined the food particle sizes used by species pairs showing the greatest spatiai overlap and found little indication that food might be partitioned based on particle size. He suggested that differences in feeding mode overlooked by his relatively coarse-grained sampling techruque, for example surface-iim versus midwater versus bottom feeding, might separate some of the more overlapping species (see Bowen 1980). Heyer (1973b) also reported a series of qualitative observations on microhabitats that include the observation that different cohorts of some species may occupy ddferent microhabitats. Heyer (1976a) studied habitat use and partitioning by tadpoles at sites in Maryland and on Barro Colorado Island, Panama. He defined habitat partitioning as including use of breeding ponds by adult frogs, microhabitat use by tadpoles within ponds, and temporal occurrence within ponds. As discussed earlier, I would suggest that habitat partitioning of the first type is better viewed as a consequence of aduit breeding biology rather than as a component of tadpole ecology. Heyer examined rnicrohabitat use at his Maryland site by dividing the available habitat into surface film, rnidwater, and bottom. Heyer sampled tadpoles of six species during 1974 and 1975. AU species for which sample sues were reasonably large showed evidence for microhabitat preference. The microhabitat of Pseudmk cmcifer larvae changed as they grew; larger larvae occurred relatively more ofien in deeper water. Indices of microhabitat overlap between species generally changed markedly between years. Oniy three species pairs maintained consistent overlaps in

both years. Heyer suggested that overlap values for most species pairs indicated that competition was unlikely. He also noted that most species experienced very different environments in the two years. The study on Barro Colorado Island (Heyer 1976a) cbncentrated o i examining species occurrences in a series of smaller streambed pools. Most pools contained only one species of tadpole. This might be caused by temporal or spatial partitioning of reproductive habitat by aduits or by a variety of other factors including interspecific egg predation or its avoidance. Heyer (1979) summarized the results of two additional years of monitoring tadvole habitats at the Marvland site. H e found considerable between-years variation in habitat use, popuiation sue, and recruitment. H e suggested that this variation resuited from complex interactions among causative factors including adult reproductive pattern, food resources, intra- and interspecific competition, physical-ciimatic factors, and predation. Turnipseed and Altig (1975) looked simultaneously at the density and age structure of Acrisgryllus, Hyla avivoca, and H. cinerea in a 4,000 mZfarm pond in Mississippi. They took weekly samples over 1 7 weeks starting on 14 A p d . Two samples were taken along each of four sections of the pond bank with a quadrat sampler placed at the waterline. Captured tadpoles were counted, staged, and weighed. Their collection methods allowed a more quantitative examination of spatial and temporal variation habitat use than previous studies. They found thatA.gryllus made up 95.7% of the tadpoles in the area sampled, and the spatial distribution of Acris tadpoles showed a consistent bias away from one area of the pond. The density varied through time, indicating some combination of changes in their propensity to remain near the shore and changes in total population size. The stage composition of the Acris population changed as cohorts developed and new cohorts were added by reproduction. The populations of both Hyiu species were apparently more evenly distributed spatially but showed temporal patterns similar to Acris, which suggest that they were responding to the same environmental stimuii. In 1980 and 1981. I used a circular auadrat samvler to examine the size structure and spatial distribution of populations of tadpoles in a pond in northern Florida. In March 1980, oniy larval Rana sphenucephala were present, but they were divided into three sue classes that were distributed differently in the pond (Alford and Crump 1982). An experiment in an aauarium showed that habitat use diferences I among the size classes persisted in a simplified environment. In 1981, I sampled the pond from just after h a t c h g of the first species of frogs that used it until it dried (Alford 1986a). Larvai Pseuducris m a t a and R.sphenucephaiu were present throughout the sampling period. B u . americanus and Pseuducris ocularis appeared after sampling began. AU species persisted until the pond dried, and B. americanus, PseudaEms ocularis. and l?&ata vroduced metamorvhs. I selected sampling lcations haph&ardly within severi habitat types and sampled most areas of the pond which was at most 60 cm deep (table 10.1; see Wiid 1996). I found R. sphenoeephaiu at densities of up to 3.l/liter early in the season and at densi-

ECOLOGY

269

ties as high as 8.0/liter as the pond began to shrink. Tadpoles of B. americanus, which often form dense aggregations, reached densities as high as 47.31liter on 17 April. I measured a number of physical and biotic environmentalparameters, includiig plant cover, substrate type, temperature and water depth. Tadpoles of B. americanus, Pseuducris ocularis, and R. sphenocephala showed differences in spatial use between individuais of Merent sizes within sampling dates, which suggest that habitat use changes ontogeneticaiiy and that cohorts of different ages may function as separate "ecological species" (Polis 1984; Wemer and Gilham 1984). Kehr (1997) also showed stage-speciic shifis in microhabitat use by l?ocularis. In rny study, aspeas of the spatial distribution of ali four species responded to one or more environmental parameters. Spatial distributions of some species pairs were correlated, which suggest that habitat use might, in part, be caused by behavioral responses of species or cohorts to one another. I evaluated this hypothesis with a laboratory experiment in an environment devoid of asymmetrical spatiai features. Rana sphenocephala tadpoles tended to be overdispersed with respect to other species. This may have been caused by other species avoiding Rana which were the largest tadpoles. Overaii, this study suggested that tadpole habitat use is species-spechc,changes ontogeneticaiiy within species, and may be plastic with tadpoles responding to short-term factors such as the presence of larger tadpoles. This complexity is not surprising given the great environmental variability encountered by temporary-pond tadpoles of each species during their evolutionary history. Diaz-Paniagua (1987) and Loschenkohi (1986) studied tadpole habitat use in Europe. Loschenkohi examined microhabitat, resource use, and seasonal diflerences among five species in the same pond. Four species (Hyh arborea, Pelobates jkcus, Rana dalmatina, and R. ridibunda) overlapped seasonaiiy. The f i f i , Bufo bufo, reproduced earlier in the year, and most Bufo tadpoles had metamorphosed before the other species bred. The four seasonaiiy overlapping species had very similar gut contents but Mered in their distributions among microhabitats. Waringer-Loschenkohi (1988) examined Merences in the use of three depth strata in experimental aquaria. Aquaria were established with diflerent population densities, vegetation, and species composition. The results showed that the species differed in microhabitat use within the aquaria. Tadpoles of B. bufo always preferred the bottom and those of H. arborea preferred the surface. Tadpoles of R. dalmatina preferred the bottorn when alone but altered this preference when other species were present, and Pfscus preferred the bottom when alone but was distributed more widely when Hyh were present. Tadpoles of R. rzdibunda preferred the bottom of a bare aquarium, but this preference disappeared when the aquarium contained rnacrophytes. Diaz-Paniagua (1987) defined five vegetation mnes w i h ponds: (1) grass, (2) short floating vegetation, (3) ernergent vegetation, (4) subrnergent vegetation, and (5) no vegetation. Samples frorn 16 ponds showed that aii tadpoles avoided mne 5. Hyh meridwnalis tadpoles were found pri-

marily in zone 4 and some were in zone 3. Tadpoles of Bufo bufo were found mostly in mnes 1and 2. Bufo calamita, Disco&sus~a&anoi, and Ranaperezi tadpoles were distributed widely in zones 1 through 4. Diaz-Paniagua suggested that diflerences in zone use were related to Merences in feeding method among species. Degani (1986) found tadpoles of Bufo viridis, Hyla arborea, Pelobates syriacus, and Rana ridibunda in a 40 X 60 m temporary pond in Israel and examined their microhabitat use in aquaria. He recorded the proportion of time spent in seven depth strata and in areas with and without macrophytes. ALI species except B. viridis spent 60%-70% of their time near macrophytes, while Bufo spent only 45% of their time there. Bufo and Rana tadpoles spent most of their time at or near the bottom, Hyh tadpoles were distributed nearly evenly among depth zones, and Pelobates tadpoles were bimodaiiy distributed between the surface or near the bottom. Degani used an experirnent to determine whether these microhabitat preferences afeaed tadpole growth rates. He foiiowed the growth of each species in three cages placed near the surface and three on the bottom. Results agreed with the habitat preference study; Bufo and Rana grew best in cages on the bottom, whde Hyh and Pelobates grew equaiiy well at either depth. Barreto and Moreira (1996) dernonstrated differences in microhabitat use among the larvae of six species of frogs breeding in a pond in central Brazil. A of the studies above were carried out in still water environments, many in highly productive temporary pools with relatively high nutrients. A few studies have examined rain forest stream environments, which are generaiiy thought to be relatively low in nutrients and rates of primary production (Bishop 1973). Inger (1969) comrnented on tadpole habitat use by frog communities along smaii rain forest streams. Odendaal et al. (1982) documented S e r ences in habitat use between Crinia riparia and C.s&nifera in streams of South Australia. In laboratory experiments they showed that the habitat preference of C. stgnijiera shified to reduce overlap when C. riparia was present. Inger et ai. (1986a) documented differences in microhabitat use among 29 species of tadpoles collected in streams in prirnary rain forest in Borneo. They identified four taxonomicaiiy heterogeneous groups: leaf-pack, pothole, riffle-shingle-openpool, and side-pool pothole species. Variation in species composition was greater between than within sites over al of their sampling. Species composition was least variable in the leaf-pack habitat. They suggested that little evidence for cornpetition existed and that most diflerences in microhabitat use probably were because of diflerences in adaptations for coping with the physical environment. Trenerry (1988) examined habitat use by four species of tadpoles living in a tropical rain forest stream in northern Queensland, Australia. He divided the habitat into pools, riffles, and runs. Litoria~enimaculataand Mixophyes schevilli primarily lived in pools, and L. nannotis and Nyctimystes dayi inhabited only riffles. Runs were not regularly censused but appeared to be inhabited at low densities by pool species. Both r f i e species were suctorial, while the pool species had

270

ROSS A. ALFORD

typicai lotic body forms. Population densities reached relatively high levels in summer months. In a pool measuring about 10 x 6 m with a mean depth of 40 cm, he estimated a population of 621 14 M. schevilli (l/m2; 0.0025Jliter) and 9162258 L.~enimaculata(15.3/m2; 0.04Jliter) in late summer. Hawkins et ai. (1988) presented s d a r information for tadpoles ofRccaphus.

Interspecific Competition and Predation

Field Studie~ Field studies cited above have shown that, regardless of habitat, species usudy overlap broadly in diet and microhabitat use and that tadpoles often reach high densities. Many tadpole habitats show a great deai of year-to-year and siteto-site variation in seasonai availability, duration, and size. Species composition of tadpole assemblages varies sidarly. This spatial and temporal variation probably reduces the degree of permanent niche differentiation among species. The combination of high population densities and relatively low niche dserentiation suggests that interspecific competition may be relatively common in tadpole assemblages. Interspeciic competition has been studied by the same general approaches used for intraspecific competition; populations have been monitored in the field, experiments have been performed in field enclosures and large-scale artificial communities, and experiments have been performed in laboratory systems. Few field experiments have examined interspechc com(1964) used a field experiment to expetition alone. amine possible competitive exclusion between Rana pipiens and R. pretwsa. The percentages and numbers surviving contradict the numbers that he introduced, so the foliowing is mv intervretation of his methods and results. H e remoied allJtadpoiesfrom 4 s m d natural ponds and then introduced 400 Ranapretiosa to 1 pond, 400 R. pipiens to another, and mixed groups with 200 individuais of each species to the other 2 ponds. Nine weeks later, 44% ofR. pretwsa and 39% of R. pipiens had survived in the single-species ponds. In the mixed-species ponds, survival of R. pipiens was 86% and 84%. while R. pretwsa survived at rates of 38% and 52%. resp&tively. H suggested that these results indicated tha; : R. pipiens was capable of expandrng its range and displacing R. pwtwsa. Wiibur (1977a) examined the relative strengths of intraspeciic and interspecific competition by raising Bufo americanus alone and with larvai Rana palustrw in 6 1 X 6 1 X 244 cm pens placed in a pond. Bufo alne were at densities of 67, 101, 134, 202, 336, and 605/m2 (0.44, 0.67, 0.88, 1.33, 2.22, and 4.00Jliter). Bufo raised with Rana were aiways present at a constant density (100 per pen; 67/m2; 0.44/liter). Rana densities were 0, 17, 34,67,134, and 202/m2 (0, O. 11, 0.22, 0.44, 0.88, and 1.33Jliter). Each treatnlent was replicated twice. A reanaiysis of his means for treatnlents with 100 Bufo larvae raised with 50,100, and 200 Bufo and 50, 100, and 200 Rana suggests that there was no effect of density (F2,6 = 0.08, P = 0.922), competitor species (F2,6 = 0.02, P = 0.900) or the interaction of density and

umas

competitor species (F2,6 = 0.02, P = 0.980) on mass at metamorphosis ofBufo. There were aiso no significant effects on arcsine-transformed proportion survivai. When these treatments are compared to the treatment with 100 Bufo raised aione, there is some evidence for reduaions in proportion surviving and mean mass at metamorphosis, but the reductions caused by Rana appear to be equivaient to reductions caused by Bufo. Wiitshire and Buli (1977) used a field enclosure exveriment to examine competition between Pseu4hlyne bibronii and l? semimamata, which have narrow zones of geographic overlap in southeastern Austraiia. They evaiuated the hypothesis that the observed geographic distribution might be maintained in part by competitive exclusion. They colleaed eggs of both species from allopatric populations, hatched them in the laboratory, and introduced hatchlings into experimental enclosures (30 X 30 X 130 cm) placed in a pond in the range of l? bibronii. Tadpoles were raised at 40 or 80 individuais per enclosure (116 or 232/m2; about 1.5 or 3.0Iliter). At densitv 40 there were 3 treatments: E bibronii Ao&, I semimakarata aione, and 20 of both species. ? At density 80, there were 3 similar treatments that were repeated with and without supplemental food. Survivai of l? semimamorata was lower than survivai of E bibronii in ali treatnlents. When both species were present without food semimamorata survived to metamorsupplementation, no l? phosis. With supplementai food, the presence of l? bibronii reduced the mass of E semimamorata metamorphs. Wiitshire and Buil suggested that under their experimental con? ditions E bibmnii outcomveted I semimamorata and that larvai competition might partidy explain the narrow geographic overlap between the species (aiso see Kupferberg 199%). Interaaions between another species pair with primarily dopatric distributions were examined by Odendaal and Buil (1983). Crinia signifera typicdy occupies creeks with slow flow rates, and C. riparia lives in faster moving water. They measured growth and survivai of both species in enclosures placed in a creek in the range of C. signifera, in the range of C. riparia, and in an overlap zone. Tadpoles were captured in the field and relocated to cages measuring 100 x 50 x 40 cm high. Densities placed in the experimental cages ranged from 124 to 222/m2 (0.62-l.ll/liter). Some cages were shielded from stream flow while others were exvosed. In general, both species grew and survived s i d a r l y i shielded ; cages; but in unshielded mixed cages, Crinia riparia grew and survived better than C. sgnifira. The results from the shielded and single-species caies-indicated that C. sgnijira could survive in areas normdy occupied by C. riparia, but the results from the mixed, unshielded cages show that tadpoles of C. signifera are at a competitive disadvantage when C. riparia occur with them in those areas. Odendaal and Bull suggested that competition was unlikely to be caused by exploitation but was more hkely a result of interspeciic avoidance behavior. Tadvoles of C. ribaria were more active in laboratory experiments and were more capable of maintaining position in flowing water. Their movements may disturb C. signifera when both species occur in high flow-rate environI

ECOLOGY

271

ments and reduce their abilitv to remain settled on favorable substrates. This could result in C. signifera being swept away by currents or could simply reduce their growth and survival rates by forcing them to expend greater proportions of their energy in returning to sheltered sites. An experiment by Woodward (1982) exarnined potential competitive interactions between temporary and permanent pond tadpoles from a desert anuran cornrnunity. He excavated an array of 124 (1 x 1 x 0.4 m) pools in a flat field; each pond contained 227 liters of water. He used these ponds to perform pair-wise conipetition experiments involving tadpoles of Bufo woodbousii and Rana pptens, Scapbiopus m b i i and Spea multiplicata, R.pipiens and Scapbwpus couchii, and large and smail R. catesbeiana and Scapbicrpus couchii. Total tadpole densities were between 50 and 200 individuals per pond (50-200/m2; 0.22-0.88fiter). Tadpoles of Spea multiplicata eliminated R. pipiens, apparently via predation. B. americanus and Scaphiapw couchii had negative effects on one another, with the effect of Scapbwpus on Bufo stronger than the reverse. Scaphwpus couchii also negatively affected R. p p n s while R. pipiens had oniy a slight affect on S. couchii. R. catesbeiana of both sizes had negative effects on S. couchii while being affected rather little themselves. Woodward suggested that most of the competitive effects he observed were because of ex~loitation coni~etition. His results also indicated that in most cases, pernianent pond species could be excluded from temporary ponds by competitively superior temporary pond species. The tadpoles of R. catesbeiana, the oniy permanent pond species able to hold its own in competition with Scaphwpus, rnight be excluded from temporary ponds by its normdy extended larval period. Morin and Johnson (1988) examined competition between Pseudacris m c i f e r and Rana sylaatica in G i c i a l pond communities. They raised tadpoles of Pseudacris alone at 83 and 166/m2 (0.15 and 0.30/iiter), Rana aione at 83 and 166/m2,and PseudaEris and Rana mixed 1:1at a total density of 166 and 332/m2 (0.30 and O.O/liter). These treatments ailowed an independent examination of the effects of intraspecific and inteispecific competition and compared the interspecific effects of both species with each other. Phytoplankton standuig crops were examined at intervals, and water from the ponds were assayed for inhibitory effects on the growth of Pseudacris tadpoles in the laboratory. The mean mass at metamorphosis of both species decreased when density changed from 83 to 166/m2. The effect on Rana was about twice as " areat as that on Pseudacris. The soecies d8ered greatly in their interspecific effeas; Pseuamis had very little effect on Rana at either density, while Rana strongly affeaed most responses of Pseudanw at both low and high densities. The interspecific effea of Rana on Pseuducris was greater than the intraspecific effect of a similar number of PseudaEris. Rana tadpoles had a greater effea on the standing crop of periphyton;han did Pse&acris tadpoles. The assays for growth inhibitors suggested that interference competition through inhibition did not occur. They suggested that the greater effeas of Rana were likely because of their greater body size and effect on the availability of resources.

Trenerry (1988) used a reciprocal-transplant enclosure experiment to evaluate the potential for competition between larval Litoria genimaculuta and Nyctimystes ahyi in tropical rain forest streams. Tadpoles of L.genimaculuta norm d y occur in slow-flowing pools, and N. dayi tadpoles are suctorial and primariiy occupy f i e s . Both species are found in runs with intermediate flow rates. He measured and staged tadpoles, placed them in 3.5-liter plastic jars with screen wire sides, and anchored the jars in the stream. Nine jars were placed in a pool and nine in a riffle: 3 containing 6 L.genimaculuta (390/m2; 1.7/l1ter), 3 containing 6 N. dayi, and 3 containing 3 tadpoles of each species. After 34 days, he measured the mean growth increments of tadpoles and the number surviving in each jar. Both species showed significant increases in body length oniy when they were alone in their normal microhabitat. The data also suggested that mortality of each species was lower in its normal rnicrohabitat. Results indicate that L. fienimd~ulataand N. dayi are suffciently specialized that they perforni poorly in each other's microhabitats. The fact that presence of the other species caused theni both to perforni poorly in their normal microhabitats indicates tha; thts s~ecializtion mav be remforced by competition when they co-occur. Penned populations in the field are cornmonly used to examine competitive relationships. For example, R. A. Griffiths (1991) found that R. temporaria increased the larval period and decreased the growth rate, mass at metamorphosis, and survival to nietamorphosis of B. calumita even when the species were not dowed physical contact (see Bardsley and Beebee 1998). Likewise, R. A. Grfiths et ai. (1991) found that the growth of larval B. calamita was inhbited by co-occurring R. temporaria, and by R. temporaria feces alone. The effeas of pond drying and competition between tadpoles of Bufo calumita and Hylu arborea and either Rana esculenta,R. bssonae, or R. a b u n d a were exarnined by Semlitsch and Reyer (1992b). Rana esculenta is a hybrid species that is a sexual parasite on the two nonhybrids. In outdoor artificial ponds containing 1,100 liters of water, they used a 3 (Rana species) x 2 (interspecific competitors presentlabsent) x 2 (ponds dry/do not dry) factorial design. Each tank contained 40 tadpoles of one Rana species. Haif of the tanks also contained 35 Hylu and 100 Bufo. Haif of the tanks in each conibination-of Rana specie~/competitiontreatment dried completely in 60 days; remaining tanks had a constant water level. The tadpoles of R. ridibunda survived at low rates in d treatments: There was a significant interaction between the effects of species and drying regime on the proportion of tadpoles that metarnorphosed in 60 days. Several other interactions involving species identity that appear in their figures 1-5 were not statisticdy significant. Competition and drying regime also had significant main effeas on percentage of individuais metamorphosing and on mass at metamorphosis. Days to metamorphosis was significantly affected oniy by competition. Tadpoles of R. esculenta performed better under poorer conditions than either of the nonhybrid species, whde the nonhybrid R. lessonae outperformed R. esculenta under favorable conditions. Sernlitsch and Reyer suggested that environmental heterogeneity may

272

ROSS A. ALFORD

be an important factor in maintaining popuiations of aii the species necessary for the stabiiity of this hybridogeneticcomplex (see Schrnidt 1993).

Laboratory Studies Compared to the large number of laboratory studies of intraspecific competition, relatively few have examined interspecific effects. Some experiments, such as those ofAiun (1966) and Licht (1967), have already been discussed with reference to growth inhibition. Smith-Gill and Gill (1978) examined interferente competition between Ranapipiens and R. ylvatzca and fed the tadpoles boiied lettuce ad libiturn. The effects of conspecific density on R. ylvatica were curvilinear and R. pipiens had iittie effect on R. ylvatica. Conversely, the effect of R. ylvatica density on R. pipkns was greater than the conspecific effect of R. pipiens. The interspeciic effect of R. 91vatica on R. DiDiens was also curvilinear (i.e.. increased more rapidly thanLiensity).They also found ekdnce for positive effects of increasing density at lower densities. They concluded that the intensitv and even the sign of intersoecific competition coefficients may depend on density and that extrapolation from single-densityexperiments to predict interactions over a wide ranp.e of densities is unlikelv to resuit in realistic predictions. It wouid be interesting to perform a similar experiment under conditions where exploitation competition was likely to be dominant as it usually is in nature (Petranka 1989). Travis (1980b) examined the responses of Hyhgratwsa larvae belonging to six sibships to competition with H.femoralis lamae from a single sibship. Six of 7 experimental treatments included 20 individuais from 1H.gratiusa sibship and 20 H.femralis sibships, and the final one included 4 H.gratiosa from each of 5 of the sibshios olus 20 H. fimmalzs. The initial densities in aii treatments were equal with 40 individuals in a plastic pan holding 6 liters of water (555/m2; 6.661 iiter). His analysis indicated that H.~ratwsa sibships differed in their compe&tiveeffects on ~.fet&raliswhose mass at day 47 of the larval period varied from means of 125-185 mg depending on the H.gratwsa sibship with whch they were competing. The H.gratwsa families also differed in their responses to the combination of inter- and intraspecific competition. The treatment that included H.g~atiosa from a mixture of famiiies showed responses intermediate between, but more variable than, the sGgle family groups. The laboratory experiment of Aiford (1989b) was designed to determine whether variation in reproductive phenlogy of aduits might interact with gene6c variation for factors deterrnining competitive abhty in larval H. chlysoscelis. I designed a 2 x 2 x 3 factor experiment in which the first 2 factors were male and female oarentage and the third was competitive regimen. Crossing two males with each of two females gave four sibships of tadpoles. I raised these in 3 competitive regimens: individuais alone in plastic boxes containing 250 ml of water (83/m2; 4/iiter), 1 H. chlysoscelis with 4 R. catesbeiana introduced simuitaneously (4 15/m2; 20/iiter), and 1 Hyh with 4 Rana introduced on dav 11 of the Hvh larval oeriod. These treatments simuiated variation in the relative tirning of reproduction of Hyh and
V

Rana similar to that encountered in monitoring of field popuiations. I suppiied food to ali containers at the same dady rate and evaluated the effects of competition on the H. chlysoscelis tadpoles by weighing them o i day 41 of their larval periods. The presence of R. catesbeiana significantiy reduced the mass of Hyla tadpoles. The four H. chlysoscelis sibshps responded differentiy to competition, and the offspring of different females differed signiicantly. Offspring from different male parents Mered significantly, but their raniung from large to s m d changed depending on whether competing Rana were introduced simuitaneously or later. These results suggest that seleaion caused by phenological variation may be a factor leading to the retention of genetic variation in the competitive abiiity of tadpoles because different suites of characters may be advantageous depending on when competitors reproduce. In anothercase (pers&xd observation), I examined competitive effects of tadpoles of Limnudynastes ornatus and Litoria inermis from Australia on lamae of the exotic Bufo marinus. I raised tadpoles in 10-cm diameter x 7-cm deep plastic containers holding 250 ml of water and suppiied food at 1 and 6X. As part of the experiient, I used 5 experimental treatments to examine interspechc competition: 1Bufo tadpole raised alone with 1X food, 1 Bufo raised alone with 6X food (density 127/m2;4/iiter), 1Bufo tadpole raised with 5 L. inemzis (762/m2;24/iiter), 1 Bufo with 5 L. ornatus, and 6 Bufo tadpoles. There were large differences in relative'effects of each species on Bufo growth rate. Tadpoles of L. inemzis had very iittie effect on growth rates of Bufo, which grew at nearlv the same rate in comoetition with five L. inermis as they 'did when raised alone. btraspecific competition from Bufo had an intermediate effect, while competition from L. ornatw had the greatest effect; Bufo raised with L. ornatus grew very slowly and ali died before metamorphosis. These results are consistent with what is known of the basic biology of tadpoles of each species. Limnudynactes are specialized for rapid growth in extremely ephemeral pools; they hatch from relatively large eggs with large yoik suppiies that d o w them to reach large sizes within 2 days posthatchmg. They are voracious feeders and behavioraiiv interfere with the access of other tadpoles to food sources by butting and can reach metamorphosis in as iittie as 13 days (personal observation). ALI of these characteristics suggest that L. ornatus tadpoles shouid be strong competitors. Tadpoles of L. inermis, which normaiiy occur in semipermanent or permanent dams and water holes, hatch from much smaiier eggs than L. ornatus and at a much smaiier sizes, grow more slowly posthatching, and appear to be less aggressive (personal observation).Bufo marinus are intermediate in aii characters except aggression while feeding; they appear to be as aggressive as L. matus, but because thev are usuaiiv smaiier after hatching and thek relative size decieases withke passage of time, L. ornatw iarvae can outcompete them for macroscopic food items. Semiitsch (1993) examined competition between tadpoles of species of the hybridogenetic Rana esculenta complex and showed that interspecific competition exists, with the hybrid R. escuienta a somewhat stronger competitor than R. iessonae.

ECOLOGY

2 73

Competitwn o Tadpooles with Other Organhms f Aithough it seems likely that tadpoles and many other groups of primarily herbivorous aquatic organisms overlap in their food requirements, very iittle inforrnation is available on competition outside the tadpole guild. In her laboratory study of interspecific growth inhibition, Akin (1966) found that water conditioned by other taxa did not inhibit the growth ofRanapipiens tadpoles. Morin et ai. (1988) examined the responses ofHyh andersonii tadpoles to competition from Bufo woodhousii folaeri and an assemblage of aquatic insects. Twelve artificial ponds containing H. andevsonii tadpoles at a density of 83/m2 (0.151iiter) were used. Six ponds also contained B. w.fmleri tadpoles (83/m2).They dowed aquatic insects to colonize three of the ponds with both tadpole species and three with only H. andersonii. Both Bufo and inse&s reduced the mean metamorphic mass of H y h tadpoles. The combined effeas ofBufo and inseas were similar to the effeas of either group aione. Competition had or no effea on larvai ~ e r i o d survivai rate of Hvh. Bv measuring the standing crop of periphyton on glass slides on a single date, they found that inseas reduced the standing crop and presumably the availability of periphyton for tadpole consumption. They suggested that in naturai temporary ponds lacking vertebrate predators, competition with inseas rnay be at least as important as competition within the tadpele guild. competi2on of tadpoleskith other taxa has aiso been shown by Alford (1989c, newt larvae), L. Blaustein and Margalit (1994, 1996; mosquito larvae), Bronmark et al. (1991, snails), and Kupferberg (1997b, insect lamae). These results strongly indicate that such effeas should be considered in hture synthetic studies of tadpole ecology. Interactwns between Competitwn and Predatwn Competition and predation can have strong effeas on the development, growth, and survival of tadpoles. In naturai systems, tadpoles are commonly exposed to both of these factors, which rnay often interact. Predation obviously affeas competitive interactions by aitering the density of competing species. DXerences in palatabhty or predator avoidance among species rnay interact with differences in competitive abhty; species with better antipredator mechanisms rnay suffer less from interspecific competition as predators redce the numbers of their com~etitors. The iniensitv of competition rnay affect predation rates and cumulative risks of predation by changing the time periods that prey spend at sizes that are vulnerable to particular predators (reviewed by Sih et al. 1985). The complexity of these possible effeas makes it necessary to experimentaily evaiuate their importance. DeBenedictis ( 1974) examined cornwtition between larvai Ranapipiens and R. rylaaticu in pens placed in a pond (101-806 individuals/m2;0.67-5.36/iiter) over a 2-year period. In the first year, density did not appear to lumt growth or sumivai of either species in any experimental treatment. He suggested that this indicated that the experimental densities were below the carrying capacity of the environment. . In the second year, there were strong effects of density and

evidence for interspecific competition with inter- and inuaspeciic competition having about equal effects. Predators (fish, leeches, and odonate naiads) reduced sumivai and increased metamorph size of both species. Many of the results of DeBenedictis are difficult to interpret because of highly variable repiicates. Artificial pond communities generally exhibit less variation among replicates than pond enclosures and therefore ailow complex experiments with interpretable results. Interactions between competition and predation were first examined in artificial pond communities by Morin (1981). He established 16 experimental ponds containing adult brokenstriped newts (Notophthalmusviridescens donalis) at densities of 0,2,4, or 8 per pond (0, 1.1,2.2, or 4.4/m2). Five days later he added hatchluigs of Pseudmi crttcifeer,Rana @emcephah, and Scaphiopus holbrookii at densities of 0.2,0.3, and O.l/liter (110, 165, and 55/m2), respectively. The community was completed 36 days later with the addition of Bufo tewestris hatchlings at a density of 0.3/liter (165/m2).Morin found that ScaphUIpushad very high sumival to metamorphosis in the absence of predators (3 = 93%). Bufo ais0 survived at a relatively high rate when predators were absent (3 = 40%). Morin attributed the low survival of Pseudacris when predators were absent (2 = 4%) to interspecific competition. The survival rates of Buj and Scaphiopus deched with increasing newt density and reached means of 8% and 18%, respectively, at the highest newt density. Survival of Pseud m i increased to 28% at a newt density of l.l/m2 and remained constant at higher newt densities. Mean body mass at metamorphosis of Pseudmi increased frorn 72 mg at a newt density of 1.l/m2 to 211mg when 4.4 newts/m2 were present. Data from his table 1suggest that total metamorph biomass produced per pond deciined from about 93 g when newts were absent to about 46 g when 4.4 newts/m2 were present; this suggests that some competition rnay have been caused by interferente. Aiternatively, this result could have been caused by increased S U N ~ Vand competition from lar~ vai Rana that did not metamorphose during the period covered by his anaiyses. Morin (1981) demonstrated that predation could aiter the competitive hierarchy in a temporary-pond system but suffered from some lack of realism by includiig only a single predator. In a second overlapping experiment, Morin (1983) manipulated the abundance of a second predator, larval tiger saiamanders (Ambystoma t@rinum). Newts were established at the same densities (0,2,4, and 8 per pond) used in the first experiment. Two additional treatments were added: one with four adult N. v. dorsalis per pond and four A. tigrinum per pond and one with only four A. tigrinum per pond. Hatchling anurans were added in the sequence and densities previously outluied in Morin (1981). Twenty-two days after the introduction of Bufo tewestris, Morin added Hyla chrysoscelis and H. ~ratiosaboth at densities of 83/rn2 (0.15/liter). Larval A. tt&rinurnrernoved nearly d tadpoles from d ponds they occurred in. The effeas of newts on the tadpole metamorph assernblage were similar to those reported in Morin (1981), although the obsemations were over a longer period. Indusion of data on H. chrysosdlts and

274

ROSS A. ALFORD

H.~ratiosashowed that total metamorph biomass did not decrease consistently with increasing newt density, which in&cates that competition was probably caused by exploitation of resources. Correlational analysis of interspecific competition suggested that it was strongly asymmetrical, with B. tewestrs, H. chlysoscelis, R. sphenocephah, and S. holbrookii negatively afFecting the growth and survival of H. ~ratwsa and l?mzrcqer, whiie not being affected by them. Morin suggested that H. jratwsa and l? crucqer may depend on reduction of the density of competing species by predators for their persistence in anuran communities. He also &scussed the possibility that some of the asymmetry in competitive effects may have been caused by the phenology of introduction of species. In both of his experiments, Morin (1981, 1983) could infer competition only through retrospective correlations of metamorph abundances with performance. Wilbur et ai. (1983) addressed this problem with an articiai pond experiment by manipulating the abundances of adult newts and tadpoles simultaneously. Newt densities were 0 , 4 , or 8 per pond (0,2.2, or 4.4/m2), and Bufo woodhousiifmleri tadpoles were introduced to ali ponds at a constant density of 1651 m2 (0.3Jliter).Three other species were introduced to low-, medium-, and high-densityponds as follows: Bufo tewestris (55,110, or 220/m2; 0.1,0.2, or 0.4/liter), Ranasphenocephala (28, 55, or 110/m2;0.05,0.1, or 0.2/liter), and Scaphwpus holbrookii (55, 110, or 220/m2; 0.1,0.2, or O Alliter). Total tadpole densities in the ponds were: low = 138/m2(0.251 liter), medium = 275/m2 (0.5/liter), and high = 550/m2 (l.O/liter). Bufo species were competitively dominant at medium and high densities when newts were absent. Scaphqus had low survivai rates, which was attributed to their inability to feed on the ilamentous aigae that dominated the ponds. Rana appeared to be strongly afFected by interspecific competition at the high density. When newts were present, they eliminated ali S. holbrookii. Newts at 2.2/m2 greatly reduced survivai of both Bufo species, and at 4.4/m2 they eliminated ali Bufo from 8 of 9 ponds. Newts also greatly reduced the survivorship of R. sphenocephah, but some Raha survived in most ponds with newts. Surviving Rana benefitted from reduced inter- and intraspecific competition and metamorphosed at mean body masses at least twice as large as Rana from ponds without predators. Wilbur et ai. (1983) suggested that R. sphenocephah was the oniy species capable of attaining a size refuge from newt predation and that predation rnight be the dominant organizing factor in many tadpole assemblages. Other studies focused primariiy on interactions between competition and predation as influences on tadpole ecology. Van Buslurk (1988) examined whether effects of multiple predators can be reiarded as additive. He studied predatLn by dragonfly naiads on a tadpole assemblage in artificial ponds by independentlyvarying the densities of Tramea carolina (O or 9.9/m2; O or 0.019Jliter) and A n a juniw (O or 2.2/m2; O or 0.004/liter). The tadpole assemblage consisted 38/m2; 0.26/liter), PseudanZs mzrcqer of Bufo americanus (1 (39/m2; 0.07/liter), l? triseriata (50/m2; 0.09/liter), and Rana sphenocephala (83/m2; O.l/liter). The composition of

the metamorph assemblage was afFected strongly by predators. Rana were almost eliminated from ponds with no predators, but about 10% survived in ponds with one species of predator. Rana survival decreased to 2% when both predators were present. Bufo and l?mzrczyer were afFected similarly by A. junius, which reduced their survival rates from 74% and 62%, respectively, to 30%. Tramea carolina afFected l? cmcqer and Bufo differentialiy with 0% and 35% surviving when Tmmea were the oniy predators. When both predators were present, their effects on Bufo appeared additive, with about 19% of Bufo reaching metamorphosis. The survivai of l?triseriata was afFected strongly by both dragonfly species by declming from 62% when predators were absent to about 10% when either predator was present and 0% in the presente of both. Predation benefitted surviving B. americanus and l? triseriata by increasing their growth rate and decreasing their larval periods; it did not afFect the performance of the other tadpole species. Van Buslurk concluded that the effects of predators on the tadpole assemblage were essentialiy additive and aliowed the effects of both species together to be predicted from the effects of each one alone. This offers some hope for the eventual construaion of models of species interactions in complex communities. Cortwright (1988) exarnined interaaions among predators and the effects of predators on their prey in 1 x 2.5-m pens placed in a permanent pond. His predators were adult Ambystoma *um and larval A. jefiersonianum, and A. macuhtum. The presence or absence ofA. opacum (O or 4.4/m2) was crossed with 3 initiai densities (17, 34, or 67/m2) of hatchling A. jeffeersonianurn. Other species were added at equal initial densities to ali pens: hatchling A. macuhtum (34/m2),hatchlmg Hyh chlysoscelis (92/m2),hatchling Rana sylvatica (200/m2),and year-old ovenvintering Rana clamitans (12/m2).Cortwright found that the yearling R. chmitans were not vulnerable to salamander predation presumably because of their large body size. Hyh chlysoscelis were virtualiy eliminated from ali pens with a maximum survival rate of 0.2%. The fate of R. sylvatica depended on the presente ofA. opacum; when these were present, R. syltatica were almost eliminated. WhenA. opacum were absent, about 10% of R. sylvatua survived to metamorphosis. There was some evidence that surviving R. sylvatica benefitted from reductions in conspecific density when the initial density of A. j@eersonianum was higher. Cortwright and Nelson (1990) report on additional results from this experirnent. Wilbur and Fauth (1990) examined competition and predation in a mo-predator, two-prey system in artificial ponds. Their experiment crossed the presence and absence ofAnax junius naiads ( m o per pond) and adult newts (Notupbthalmus virihscens; m o per pond) with the presence and absence of tadpoles of Bufo americanus and Rana palustrs, both at 276/m2 (0.5Jliter). When predators were absent, there was significant intra- and interspecic competition. The leve1 of competition was reduced when predators lowered survival rates. The survival of both species was reduced in ponds with one predator and one prey species but to a species-specific extent; Rana were more strongly afFected by Anax than were Bufo. When one or two predators and both prey were pres-

ECOLOGY

275

changes in priority of reproduction by Bufo americanus and Rana sphenocephala in artificial pond systems. We crossed three Rana treatments (absent, 100 [55/mZ,O l/liter] intro. duced on day 0, and 100 introduced on day 6) with three Bufo treatments (absent, 500 [275/m2, 0.5/liter] introduced on day 0, and 500 introduced on day 6). For each species, these treatments yielded ponds with no interspecific competition and ponds with interspeciic competitors introduced simultaneously, 6 days earlier, and 6 days later. The treatment with neither species introduced was used in a separate esperiment (Wilbur and Alford 1985; see below). Both species performed best in the absence of interspeciic competition. Both species aiso performed better when they were introduced on day 6 of the experiment than when they were introduced on day 0. The effect of Rana on Bufo appeared to increase when thev were introduced earlier than Bufo and to decrease when they were introduced later than Bufo. This result was consistent with the idea that the effects of interspecific competition are size-specific, with larger animais having greater competitive effects. Rana actuay performed better when Bufo preceded them into ponds and worse when Bufo foilowed them into ponds. This may have been caused b i a reduction in the number ofBufo competitor days esperienced by Rana when Bufo preceded them into the ponds. The performance of tadpoles can depend on the absolute fine timing of reproduction by adult frogs and on the relative timing of reproduction by competing species. L. Blaustein and Margaiit (1995) have demonstrated similar effects of sma changes in priority on the interactions between larval Bufi viridis-and mosquitoes (Culisetalongiareolata). The study of Alford (1989a) was designed to examine the effects of short-term variation in reproductive phenology in greater detail than had been revealed in the preceding esperiment (Aiford and Wilbur 1985). I used 9 experimentaltreatments to examine the responses of tadpoles of Bufo americanus and Rana palus~rz'r 7 relative timings of introduction to of the other spe& that spanned a range f 11days preceding to 11days foiiowing. A treatments included both speU cies at the same initial densities of 55 Rana/m2 (O.l/liter) and 274 Bufolm2 (0.51liter). Growth and survival of both Effeets o Environmental Uncertainty f species failed ;o resbond to &e timing of introduction ofthe Environmental uncertainty has been featured prominently other species. Correlations between growth rates and numthroughout this chapter. Reproductive site choice by frogs, ber of metamorphs indicated that both species responded and thus possible tadpole community composition, is d u - to their own densitv. but there was little vidence of interenced by a complex interaction between environmental specific competition. These species frequently co-occur in quaiix environmental uncertainty, life history, predation North Carolina ponds, and I suggested that the lack of interrisk, and resource availabiiity. Tadpoles are capable of re- specific competition might reflect co-adaptation to avoid sponding to alterations in the environment, both physical competition and that the effects of phenological variation on and biotic, by aiteritig their behavior and physiology and species interactions depend on the species being examined. Effects of short-term variation in reproductive phenology changing their relative rates of growth and development. As Morin and Johnson (1988) pointed out, competition be- were also explored by Gascon (1992a) and S. E Lawler and tween species of tadpoles is ofien asyrnmetrical and may de- Morin (1993). Gascon examitied competitive relations bepend on relative body sizes. Relatively recently, a number of tween the early reproducing Osteocephalus taurinus and the field experiments have attempted to evaluate how environ- later-reproducingEpipedobatesfemoralir and Phyllomedusa tomental uncertainty of several types effects ecological rela- mqptema i11 Amazonian Brazil. He conducted three experitions within and between species of tadpoles. ments in 45-cm diameter x 19-cm deep basins placed outAiford and Wilbur (1985) were interested in determining doors in the forest. Neither of the late-breeding species whether the outcome of competition was dected by sma dected the early species, but the early species deGed both ent, the prey survived at higher rates than when they were alone with the predator; they appeared to share the rise of predation so that each species benefitted from the presence of the other. When both predators were present with one prey species, the prey suffered about additive mortaiity from both predators-the survival rate of prey in two-predator, one-prey systems could be predicted from their survival in one-predator, one-prey systems. Wilbur and Fauth exaniined their results further with several models of multispecies interactions. Fauth and Resetarits (1991) reported on the results of an experiment in which newts (Notopbthalmusvirideseens) and sirens (Siren internedia) preyed on a five-species tadpole assemblage. They found that newts acted as keystone predators, altering the composition of the tadpole assemblage, while sirens were nonselective. The predators interacted, apparently through predation by sirens on newts. Peacor and Werner (1997) found complex interactions between competition, predation, and behavioral antipredator mechanisms in experimental systems containing two size classes of larvai Rana catesbeiana, larval Rana clamitans, and ~ W O odonate predators. These field experiments demonstrate that competition and predation interact in a complex mamer in determining the success of tadpoles. Some species that have effective antipredator mechanisms (e.g., Pseudacns cmcifer) may benefit unambiguously from predation on competing species. Surviving individuais of many species may benefit from increased growth rates and body sizes at metamorphosis caused by reductions i11 the levels of inter- and intraspeciic competition. The extent of these benefits may depend on interactions between freeing of resources by the thinning of competitors and the costs of antipredator mechanisms such as reductions in rates of movement (Van Buskirk and Yurewicz 1998). Whether the advantages outweigh the disadvantages of decreased survivorship for species with poor defenses probably varies from time to time and place to place depending on the densities of competitors and on the presente of other mortaiity sources such as pond drying. Some of these questions are addressed in the next section.
2 ,

276

ROSS A. ALFORD

of the late-breeding species. The relative tirning of introduction altered the effects of 0. taurinw on l?tomoptema; they were reduced when both s~ecies were introduced simultaniously. Relative tirning did not alter the competitive effect of 0. taurinus on E. fenzoralis. S. P. Lawler and Morin (1993) used artificial pond experiments to determine how Bufo woodhousii and Pseuducris crucqer larvae responded to O, 7, and 14 day lags in tirning of reproduction. They found that the species responded asymmetricdy; Pseudacvis were affectedmore strongly by Bufo when Bufo were introduced 7 days before them, while Bufo were not affected by Pseudacvis. They concluded that the usual pattem in nature, in which Pseudami's reproduce very early, probably maxirnizes the success of this competitively inferior species. Competition may be direct, through the use of the same resources or behavioral or chemical interferente, or indirect, through alterations to the physical or biological structure of the environment. Wilbur and Alford (1985) demonstrated that tadpoles may affect other species through persistent effects on the environment. We examined the effects of pond history and competition on the performance of ~ ~ l a h r soscelis larvae. Hyh were introduced at a density of 274/m2 (0.5lliter) to the artificial ponds used in a separate experiment (Alford and Wdbur 1985. discussed above) 80 davs after &e ponds were initiaiiy filld. They were als raised in a set of ponds that had been filled only 15 days previously and had never been occupied by other tadpoles. This led to H. chvsoscelis tadpoles experiencing ponds with 10 different hstories: 15- and 80-day-oldponds that had never been occupied by tadpoles, 80-day-old ponds that had been occupied by Rana sphenoc~.phala Bufo americanus for 66 or 72 or days, and 80-day-old ponds that had been occupied by both Rana and Bufo for various combinations of 66 and 72 days. Most Bufo had completed metamorphosis before any Hyla were introduced to the ponds; effects of Bufo on Hyla thus were caused by environmental alterations rather than direct competitive effects. Growth and survival ofH. chrysosceliswas greatest in 15-day-old ponds. Eighty-day-old ponds that had never been occupied by other tadpoles produced the next best Hyh performance. The presence of Bufo or Rana decreased the performance of Hyla metamorphs, which was also iduenced by the order in whch Bufo and Rana had been introduced. Indirect effects between tempordy separated species were examined further by Morin (1987), who introduced Pseudacvis crucifer tadpoles to experimental ponds at 55, 110, and 220/m2 (0.1, 0.2, and 0.41hter). These were crossed with densities of the predatory newt Notophthalmw viridescens at 0, 2, or 4 per pond (0, 1.1, or 2.2/m2). Forty-three days later, about 1week before Pseudacri's began to metamorphose, Morin added hatchling Hyh versicolor to d ponds at a density of 110/mZ(0.2lliter). Newts had little effect on the growth .and survival of Pseudacri's, but increasing newt density strongly decreased the survival to metamorphosis of Hyla. When newts were absent, the total froglet biomass exported from ponds was relatively constant (about 75 g). The proportion of that biomass attributed to Hyla depended strongly on the initial density of Pseudami's; although d

Pseudacvis completed metamorphosis early in the Hyh larvai period, there was a significant negative correlation between Pseudumis biomass and Hyla biomass exported from ponds. Hyla versicolor growth and survival rates also decreased with increasing densities of Pseudacris. Morin pointed out that this pattern is consistent with a hypothesis of preemptive exploitation competition for resources; a limited total quantity of nutrients is avadable when the pond fills and the export in metamorphs of early breeding species may reduce the avadability to later reproducers. He also noted that Pseudacri's appeared to be competitively superior to Hyh and that their temporal separation might not preclude competitive exclusion under some conditions. Warner et ai. (1991) examined interactions among the effects of density, long-term reproductive priority, and pH on the larvai biology of Hyhfemoralis and H.gratwsa in outdoor ta&. M remaining H.gratwsa were removed before the introduction of H. fenzoralis 108 days after H. gratiosa had been introduced. They found that H.gratwsa larvae had longer larval periods and lower metamorphic body sizes ywhen densities were higher. Lower environmental p H increased the effects of intraspecific density on H. gratiosa. Hyhfemoralis responded similarly, but the effects of pH did not interact with the effects of density. H. femmalis introduced to ponds that had previously contained H. gratwsa performed better than ones in ponds with no previous tadpole occupants. Warner et al. (1991) suggested that this facilitation might have been caused by effects of H.gratiosa on resources; their grazing may have changed the abundantes of aigae, favoring types with better nutritive quality, or may have enhanced nutrient cycling or reduced the prevalence of some toxic substance. The consequences of variation in reproductive tirning within a species were examined by Morin et al. (1990). They introduced a second cohort of Hyh andersonii tadpoles at a density of 55/m2 (O.l/liter) to 12 artificial ponds used in an experiment discussed earlier under competition with other taxa (Morin et al. 1988).The second cohort was introduced 34 days after the first. Their resuits indicated that the second cohort experienced longer larval periods and reduced growth rates and survival to metamorphosis when compared to the first cohort. Mean masses at metamorphosis were similar between the cohorts except when interspecific competitors and predators were excluded; in that treatment, the earlier cohort reached metamorphc masses nearly twice as great as the later cohort. Hearnden (1992) demonstrated similar effects in Australian popuiations of Bufo marinus, with the additional factor that when a second cohort was introduced as eggs, the earlier cohort preyed upon them, adding predatory reductions in survival to competitive effects. The precediig experiments showed that the reproductive phenology of frogs can affect ecological interactions among tadpoles. Variation in reproductive phenology has the potential to alter the outcome of competition between some species. Variability in the persistence of ephemeral ponds is probably another aspect of environmentai uncertainty that is important in tadpole ecology. Wdbur (1987) examined interactions between competition, pond drying rates and pre-

ECOLOGY

277

dation in experimental pond communities. He hypothesized that predation may benefit some tadpoles in temporary habirats by reducing interspecific and intraspecific competition by aiiowing sumivors to grow more rapidly and leave ponds before they dry. He tested this hypothesis with an artificiai pond experiment including three rates of pond drymg: connant water depth of 50 cm, drained after 50 days, or drained afier 100 days. H e aiso included 2 predation regimes of adult Notophthalmus viridescens dorsalis (absent or present at a density of 2.2/mZ [0.004/liter] and low and high tadpole densities (totals of 242 and 970/m2: 0.44 and 1.76Ilite1-). , , The tadpoles included in each pond were Bufo americanus (110 or 441/m2; 0.2 or 0.8/liter), Hyia chrysoscelis (69 or 276/m2; 0.125 or 0.5/liter), Ranasphemcephau (28 or 1101 m2; 0.05 or 0.2/liter) and Scaphiopus holbrookii (36 or 1431 mZ;0.065 or 0.26lliter). The composition of the metamorph assemblage was d u e n c e d significantly by predation, drying time, density, and their two-way interactions. Scaphiopus appeared to be the dominant competitor. Rana negatively atfected the growth of ali species except Scaphiopus. Bufo and H appeared to have negative effects only on themselves. + ,t low initial densitv. newts elimnated nearlv aii tad~oles. i -it high density, si&& total numbers of tadp61es metamorphosed regardless of newt presence, but the composition of the metamorph assemblage changed; newts reduced the dominance of Scaphzupw and aiiowed other species to metamorphose in greater numbers. This effect was more pronounced in ponds that dried. With predators absent, only Scaphiopzw eherged from ponds thatdried in 50 days, b i t mith predators present, both Bufo and Scaphzupus metamorphosed. Sirnilarly, only Bufo and Scaphiopw emerged in any numbers from ponds that dried in 100 days when predators n-ere absent, b;t the presence of predatois allowed significant numbers of Hyla to metamorphose as weli. There was aidence that tadpoles responded directly to pond drying by initiatinp: metamomhosis earlier. Wibur inter~retedhis results as indicating that competition and desiccation probably dominate the dynamics of populations exploiting very ephemeral habitats, with predation possibly acting as a moderating influente. The importance of predation should increase as habitat duration increases. Harris et ai. (1988) documented the temporal patterns of abundance of adult and lamal newts (N. v. dorsalis) in a pond in North Carolina over a 3-vear period. We showed that the riming of newt arrivai and depkure and the duration of newt presence varied among years. This variation was not nreli correlated with variatin in the tad~oleassemblage (personal observation). Ali of the experimental studies cited thus far which examined predation did not explicitly d o w for the fact that predators typicaiiy foliow their own phenologicai patterns. I conducted an artificial pond experiment (Alford 1989c) to examine the effects of predator phenology and phenologicai variation. My experiment was combined with the one reported bv Wilbur (1987). The same four anuran species were introduced at the same two densities. I established 4 predation regimes that included 2 extreme treatrnents in which adult Notophthalmw at a density of 2.21 m2 (0.004/liter) were absent or continuously present. Two
0 ,

additional regimes mimicked the extremes in newt departure time obsemed by Harris et ai. (1988)-newts initiaiiy present but removed from ponds on 1 Aprii and newts initially present but removed on 13 and 14 May. A correlation analysis suggested that interspecific competition foliowed a hierarchy with S. holbrookii being the dominant competitor foiiowed in order by R. sphenoczphaia, B. americanus, and H. chrysoscelis. The presence of predators for any length of time aitered the composition of the metamorph assemblage. The anurans that reproduced earliest (Rana and Scaphzupus) were affected most negatively. When newts persisted for longer periods, they aiso affected later reproducers. The abundance and contribution to metamorph biomass of Buf declined when newts persisted until 13-14 May, while Hyia increased. The relatively abundant Hyia in this predation regime appeared to have competitive effects on the other species, which suggests that interspecific competition may not be completely described by the simple hierarchy postulated earlier. Forcing newts to remain in the ponds through the summer reduced the abundance and biomass of Hyla by increasing the relative contributions of the other species to the metamorph assemblage. Bufo, Rana, and Scaphzupustadpoles appeared to reach size refugia after which newt removai did not aiter their sumivai patterns. There was strong correlationai evidence for negative effects of increased tadpole density on the performance of newt lamae. Because the primary resource of tadpoles is plant materiai and newt larvae feed primarily on zooplankton, this negative effect probably was caused by diffuse competition through two intermediate levels of the trophic web; tadpoles aitered the aigai community, which altered the zooplankton community, which affected the performance of newt larvae. This experiment confirms that incorporating predation into models of temporary pond and other communities requires some knowledge of predator phenology in addition to predator abundance.

Summary
The avaiiable information on the ecology of tadpoles has increased greatly since Wilbur's (1980) review. Tadpoles have proved to be a valuable model system for investigation of many aspeas of population and cornmunity ecology. In the course of these investigations, it has become clear that a iii understandmg of the ecology of any tadpole assemblage requires data of severai types. First, there must be information on reproductive habitat choice and phenology of the frog species mhabiting that region. Choices of reproductive habitats by frogs set the range of possible tadpole assemblages that may develop. Reproductive phenology can predetermine the outcomes of interactions between tadpoles and their competitors and predators. Second, mformation on the reproductive habitat use and phenology of potential predators is necessary for the same reasons. I suggest that information on potential competitors belonging to other phyla will aiso be needed. Third, it is necessary to have mformation on tadpole, predator, and probably interphyletic competitor densities. Finaiiy, information on resource availabhty, habi-

278

ROSS A . ALFORD

tat stnicture, behavioral responses of tadpoles to competitors and predators, and the previous history of the habitat may be needed. A lack of any of this information may lead to incorrect predictions about the course of interactions in any tadpole assemblage. Extrapolation from ecological interactions experienced by tadpoles to their effects on the population biology of frogs requires much more information on the biology of metamorph and juvenile fkogs than is currently avadable for nearly d species. This complex picture is far from the optimisticdy sirnple views of community ecology that prevailed in the early 1970s (see May 1974). It is also far from the nondeterminism espoused by some authors (see Strong 1983). Rather, it appears that tadpole assemblages, and probably complex assemblages of other species, are highly deterministic but that a great deal of information is needed to unravel the sources of that determinism. I identified several areas that are in particular need of h r ther investigation. To fully understand the autecology of tadpoles, detaded studies on their foods, feeding rates, and assimilation efficiencies must be initiated. Such studies also would aid in understandmg the role of tadpoles in communities. Although the evidence is sparse, tadpoles may provide a major source of ine particulate organic matter for detritivores. Studies of interphyletic interactions between tadpoles and other aquatic herbivores would also aid in examining this connection. Despite the comparatively great amount of work that has examined the relative imponance of interference and exploitation as competitive mechanisms in tadpoles, some good areas for investigation in these fields remain. Combining the quantitative genetic approach with ecological experiments

designed to separate interference and exploitation may lead to some insight into the evolution of these competitive mechanisms. The extreme competitive mechanism of carnivory or cannibalism, which often appears to be facultative in tadpoles, also should be looked at more extensively with quantitative genetic analysis. Interactions between tadpole behavior, inter- and intraspechc competition, and predation risk also have been explored only preliminariiy. I suspect the reader wiii have arrived independently at a iist of research needs that differs from mine. T h s should emphasize the fact that although much more is known about tadpole ecology in 1998 than was known in 1980, much more remains to be discovered.

ACKNOWLEDGMENTS
I would Me to thank S. J. Richards for h s considerable assistance in reading and commenting on drafts of the manuscript and for access to h s comprehensive reprint collection. R. Altig and R. W. McDiarmid provided imrnense assistance in keeping up with developments in the field during the prolonged gestation of the book and made many invaluable editorial comments and corrections. I would also like to thank L. J. Alford and M. N. Hearnden for comments on the manuscript and my students K. S. Bradfield, M. R. Crossland, W. Martin, and M. I? Trenerry for discussions and access to their data. Support for my research was provided by several Australian Research Council grants, the Council of Nature Conservation Ministers, and the Cooperative Research Centre for Tropical Rainforest Ecology and Management.

280

REID N. HARRIS

the evolutionary hypothesis of interest is that the larvai stage . .is where adaptations to maximize growth rate are expected. Evidence consistent with this hypothesis would include the presence of nonadditive genetic variation for growth rate &d a response to naturaiselection on genotype< that M e r in growth rate. Trade-offs between growth rate and other traits, such as competitive ability or antipredator mechanisms, might aiso be expected (Aiford 1986a, 1989a, b). If the gowth rate maximikon hypothesis is correct, selection has favored growth over other traits to the degree that any confia among fnctions exist. Rapid g r o d rate d o w s a tadpole to reach a sue refuge from many gape-limited predators and achieve a large body size at metamorphosis. Aiso, larvae quickly attain the minimum size needed for metamor~hosisand have the o ~ t i o n to metamorphose i the pond kgins to dry or if pre&on f pressures increase. Body size at metamorphosis can be negatively correlated with age at first reproduction and positively correlated with sue at first reproduction (Berven and Giii 1983; D.C. Smith 1987). Thus, a high larvai growth rate is also of potential value in terrestrial survival, fecundity, mating success, and age at first reproduction. Predator avoidance and resistance to parasitism are other important traits that can be disassociated from growth rate and may be involved in phenotypic or geno<pic trade-offs with-growth rate. Morphologicdy, tadpoles appear primarily adapted for feeding and therefore achieving maximal growth. A large proportion of a tadpole's biomass (50%) is accounted for by its digestive tract (Wassersug 1974). Many of the morphological variations of buccopharyngeal (buccal filters) and oral (papiiiae, labial teeth) features are correlated with the mode of feeding (e.g., carnivorous, suspension feeding, scraping; Altig and Johnston 1989; Orton 1953; Wassersug and Hever 1988). In " general. carnivores have reduced branchiai filters, buccal papdiae, and labial teeth, whereas suspension feeders have the most elaborate buccopharyngeai structures. Suaoriai and surface feeding forms have major morphologicai features associated with feedmg (Altig and Johnston 1989; Orton 1953).
1

growth is restrained and the degree of restraint can be modeled by the coupled equations

and

Here, y is the size-specificgrowth rate, and a is the exponentiai decay of the size-specificgrowth rate. Physiologicdy, y can represent the rate of ceii division; (Y can represent ceii death, slowing of division rates, and the truncation of growth as tissue differentiation occurs. The solution of the coupled equations is

where W, is initiai size and y, is initial size-specificgrowth rate. The specific growth rate y exponentidy decays as described by (5) This model ~rovides im~ortant an framework for a discussion of evolutionary and ecological questions (Travis 1980a; Wilbur and Collins 1973). First, it deimeates three wavs in which final size can be afTected: f. 1)variation in initia size (W,), (2) variation in size-specific growth rate (y), and (3) variation in the exponentiai decay of the size-specific rate (a). AI parameters can be afeaed intrinsicdy by genotype and extrinsicdy by environmental faaors. In addition, variation in initial sue ( W,) could result from variation in egg sizes, including yolk content, or from geneticaliy or environmentdy caused variation in embryogenesis. The parameter (Y can be thought of as the rate of change of the sizespecific growth rate and may increase or decrease throughout the iarval ~ e r i o d .Thus. differences in size-s~ecific growth rate, y, can be based on Merences in y,, a , or both (Travis 1980a). Based on developmentai events, the previous equations predict that growth wiii be size specific; this leads to interactions among the parameters. Dserences in initial size among individuais or sibships wdi be translated into differences-h dWldt. An increase in final size at metamor~hosis can be achieved by beginning at a larger size that ;an be maintained by a constant size-specificgrowth rate and a constant damping parameter (a). large size at metamorphosis A can aiso be achieved by beginning growth earlier and at a higher size-specificrate (y,). Early growth may be a conservative trait maxirnized by selection, but the damping pararneter, a, may be plastic and underlain by a large amount of genetic variation (Travis 1980a). Rapid early growth aows the tadpole to escape gape-limited predators and to reach the size suitable for metamorphosis if the pond begins to dry or predation pressure increases. Tadpoles are under selection for rapid growth; thus, a variable a would seem maladaptive.
1

y = yo Exp (-at).

Growth and Metamorphosis

Grmvth Models Growth rate is a criticai adaptation in anuran tadpoles, and I wiii review the basic growth models that workers have used to study tadpole growth. Growth is a physiological process that can be mathematicdy modeled (M. J. Katz 1980; Laird et ai. 1965); the presentation below foiiows Laird et al. (1965). Umestrained exponential growth can be represented by

where W is size, t i s time, and k is a constant. This equation models the case of unlimited ceii division. Exponential

T H E ANURAN TADPOLE

281

Maintenance of genetic variation for the damping parame- ability to dissociate growth rate and developmental rate can ter, a,may be explained by relatively weak selection on late occur throughout the larval period afier a minimum metagrowth rate as compared to early growth rate in the tadpole morphic si; is attained. stage (Travis 1980a). Once a tadpole achieves the minimal Travis (1984) presented a fixed-rate model that predicts size necessary to metamorphose, the selective pressure on that growth rate and developmental rate are dissociated oniy growth rate may relax because of the option of metamorpho- during the early parts of the larval period. Developmental sis. If resources decline or predation pressure remains high, rate is set toward the beginning of the larval period, and subthen selection would favor the metamorphic decision (Wil- sequent changes in growth rate change oniy size at metamorbur and Cohns 1973). A negative genetic correlation be- phosis. Hensley (1993) modified the Wilbur-Collinsmodel tween growth rate and length of larval period (Travis et ai. ;o take into account a fixed develo~mentdrate late in the 1987; discussed below) is another explanation for the persis- larval period. Changes in growth rate that occur late in detente of genetic variation for later growth rate. velopment affect size at metamorphosis and not developA t e m for developmentai rate (Mord and Harris 1988; mental rate. The proportion of the larval period that is sensiHensley 1993; Leips and Travis 1994; Smith-Giil and Ber- tive to changes in growth rate is the fundamental difference ven 1979; Wilbur and Coilins 1973) is lacking in the above between the Wilbur-Coiiins and fixed-rate models. This senequations. Wdbur and Coliins (1973) predicted that growth sitive period may be variable. Indeed, K. A. Berven (perrate and developmental rate should be dissociable (i.e., de- sonal communication) has evidence that populations of velopmental rate should be responsive to changes in growth Rana sylvatica differ in the fraction of the larval period that rate). In an experimental study, Bufo woodbousiifowle~tad- is responsive to changes in growth rate. This difference sugpoles did slow their developmental rate in response to an gests that variation among populations and species could be enhanced growth rate (Mord and Harris 1988). In Scaphw- correlated with selective environments or revresent different pus couchii, the degree of correlation of developmental and "solutions" to maximizing size at metamorphosis. growth rate varied among sibships. Sibships that had a rapid Mord and Harris (1988) argued for an experimental developmentai rate could not alter it in response to an en- manipulation to evaluate adequately the Wilbur-Collins or hanced opportunity for growth (e.g., a long-duration pond). fixed-rate models because Merent correlative patterns are This implies a "tight link between growth and development" consistent with either model. For example, Coiiins (1979) (Newman 1988a). Other sibships characterized by a weaker stated that a positive relationship between length of larvai correlation between growth and developmental rates had period and mass at metamorphosis of field-coiiected metaa slower overail developmental rate and were able to take morphs would support the Wdbur-Cohns model. As the advantage of enhanced growth oppomnities. Temperature first larvae metamorphose, more resource per capita is left. affects growth and developmental rate differentiaily in tad- Larvae delay metamorphosis to take advantage of the inpoles of Rana ylvatua (Smith-Giiiand Berven 1979). Exper- creased " nrowth rate. This assumes that resource level is conimental evidence indicates that growth and developmental stant or increasing in the pond. Resource level can decline rates are separate biological entities and are dissociable in during the season if primary productivity decreases or if prianuran larvae. A quantitative model that incorporates the mary productivity is constant, and increasingly larger tadgrowth parameters discussed above with developmental pa- poles consume more. In this case, a negative correlation rameters is needed. would be generated even if the larvae are conforming to the Wilbur-Collins model. Models ofMetamorphu TzmingThe fixed-rate model also can generate either a positive The ability to time the change between iife-history stages in or negative correlation of length of larval period and size at a way that wiil enhance fitness is a key adaptation of any metamorphosis, assurning some variation in developmental species that has more than one iife-history stage. Wilbur and rate is present. Larvae with a fixed rapid developmentai rate Coilins (1973) proposed a model that predicts the timing of metamorphose early, freeing resources for larvae that have a amphibian metamorphosis. The cornerstone of their hy- slow developmentalrate. This situation would lead to a posipothesis is that growth rate (GR) and developmental rate tive correlation. If resources are deciining over time, a nega(DR) are dissociated (see chaps. 9 and 10). The degree to tive correlation would be produced. Variation in developwhich dissociation is possible represents an adaptation; the mental rate can be caused by underlying genetic variation or degree to which dissociation is not possible represents con- by a response to competition early in the larval period. In straint. Growth rate is the change in body size over time. summary, since either model can generate a positive or negaDevelovmental rate is the avvea&ce of new features (bio- tive correlation of length of larval period and size at metachernical, morphological, behavioral) in the organism over morphosis, an experimental analysis is required to ascertain time (Gosner 1960; A. C. Taylor and Koiiros 1946). The how individuais are making metamorphic decisions. Wdbur-Coiiins model ~redicts if the nrowth rate of the Three ex~erimentsthat tested larval resDonses to inthat I larva increases, developmental rate is retarded to capitalize creases and decreases in food level address predictions of on the improved growth conditions. If growth rate de- the Wilbur-Coiiins and fixed-rate models. Mord and Harris creases, developmental rate is predicted to increase to escape (1988) exposed tadpoles of Bufo wooa!housiifowle~ eight to to a deterioratini environment.~ccordin~ the model, &e experimental treatments. Four treatments initiaily had low
L L

T H E ANURAN T A D P O L E

283

of Werner nor L. Rowe and Ludwig explicitiy considers changes in size-specific growth rates during a season, aithough it should be possible to incorporate such changes into their model. Changes in recent growth history are the cornerstone of the Wilbur-Colhs model. Other models of metamorphic timing are based on staging series. Pandian and Marian (1985a) and Srnith-Gill and ~ & e n(1979) proposed a reliance on developmentai rate to predict tining of metamorphosis and found that differentiai rate is a good predictor of metamorphic tirning. Staging series used in both studies are intended ex~licitlvto reflect changes in forrn over time leading to rnetamorphic climax. Thus, extrapolation to the end point of a developmentai stage versus time relationship should predict when metamorphosis occurs, if environmentai conditions withn a population remain constant. Indeed, both studies found that different population densities and temperatures could aiter the relati6nShip of developmentai rate and growth rate, a prediction made by Wilbur and Coilins. The claim that growth rate is nctionay dependent on deveio~rnentairate (Smith-Gdi and Berven 1979) is not supported in the sense that variation in developmentai rate has not been shown to affect changes in growth rate. The hypothesis that an underlying developmentai process exists that effects changes both in size and stage, however, has not been disproved. As the biology of growth and development becornes better known, t h s hypothesis can be evaluated. White and Nicoli (1981) discussed evidence that ~rolactin and thyroid horrnones are antagonistic. A switch in concentration from growth hormones (i.e., prolaain) to a rising concentration of thvroid horrnones (TH) with the dominance of TH could generate an interaction between growth and development. Wassersug (1986) noted that prolaain increases mucus production in tadpoles, and mucus contains prostaglandin, which slows development of an adult stomach. The hypothesis that developmentai rate is adjusted for changes in growth rate is predicted by both the WilburCollins and Werner models. The adaptation proposed by Wdbur-Collins and Werner is subjea to some constraints. Developmentai rate is not responsive to changes in growth rate throughout the entire larvai period. This was suggested by Wdbur-Coliins and by the different results obtained by Aiford and Harris (1988), Audo et ai. (1995), Hensley (1993), Leips and Travis (1994), and Travis (1984). Seemingly, changes in growth rate have no effect on developmental rate beyond a certain developmentai stage. An extreme case ma; be when metamorphosis is initiated bv a surge i11thyroxine. Any change in growth rate at that time, no matter how dramatic, is unlikely to affect that process. Hensley (1993) suggested that at least by Gosner 38, Pseudan2s nztcqer initiates metamorphosis and that developmental rate is fixed at that point. The verbal model of Leips and Travis (1994). which tries to unite the models of Wilbur-Coliins and Smith-Gill-Berven, assumes that resources are docated between growth and development differently in early and late larvai stages. At early ages, resources are mainly aliocated to development: "development has the priI
i
\ ,

macy." At later stages, the aocation pattern changes such that fewer resources are docated to development and that "the proportionate allocation of resources to growth . . . is total after the developmentai trajectory is fixed. Presumably they mean that after the developmentai trajectory is fixed, any additionai resources are aocated to growth. They derive predictions from this model which they state "accommodates d of the extant data" relevant to amphibian responses to changes in food level. The main prediaion of the model is an asymmetry in response to increasing and decreasing food levels at different larvai ages. Earlier increases in food level affect body size less than do later increases because of an increasing proportionai aocation of resources to growth. Earlier decreases in food level affect body size more than do later decreases because of the compensatory effect of a greater proportional aocation of resources to growth late in the larvai period. I propose an alternative graphicai model (fig. 11.1)that posits that energy allocated to development increases late in the larvai period as metamorphosis is initiated. This rnodel also distinguishes the proportion of energy aocated to growth and developrnent from the absolute amount of energy allocated to growth and development. At low food levels, individuais devote proportiondy more resources to development and less to growth in order to rnetamorphose at the minimum size in the shortest time. At high food levels,

a"
Time

a>

Time

Time

Time

Time
Fig. 11.1. Graphicai model of energy aiiocation during development and growth of larvai amphibians. (A) The proportion of energy aocated to development by larvai anurans. (B) ~ h proportionai amount e of energy aiiocated to growth. (C) The absolute amount of energy aocated to development. (D) The absolute amount of energy aiiocated to growth. (E) The absolute amount of energy allocated to growth if food Irvel changes after metamorphosis has been initiated. The rectangle at the right side of each graph denotes that period of time from the initiation of metamorphosis until metamorphic climax. Abbreviations: D = development, G = growth, H F = hgh food condition, and LF = low food condition.

r ,

284

REID N. HARRIS

individuals devote proportionally fewer resources to development and more to growth in order to take advantage of favorable conditions for growth (figs. 11.lA, B). Individuals are able to devote absolutely more energy to both development and growth at high food conditions (figs. 11.lC, D). It foliows that individuals at low food conditions metamorphose at a smaiier size after a long larval period, and individuais at high food conditions metamorphose at a larger size in a short time. Upon the initiation of metamorphosis, suggested to be between Gosner 35-38 (Hensley 1993), develpmental rate is ixed and begins to consumi a greater proportion of energy because of morphologic and other changes associated with metamorphosis induced by increasing titers of thyroxine. Proportional and absolute amounts of energy devoted to growth decline. The observation that individual growth curves tend to level off or drop toward metamorphic climax supports the idea that less energy is devoted to growth, presumably in favor of an increased allocation to development. Changes in the food level prior to fixation of developmental rate will have prediaable results; increases in food level wiii lower the proportion of energy aiiocated to development but allow a greater absolute amount of energy to be allocated to development. A greater proportion and absolute amount of energy is aiiocated to growth; length of larvai period will decrease and size at metamorphosis will increase. of Decreases in food level will increase the ~ r o ~ o r t i o n energy aiiocated to development but a small absolute amount of energy will be available for development. Proportionaiiy and absolutely, little energy is allocated to growth. Length of larvai period increases and size at metamorphosis decreases relative to individuais that experiente constant high food. Changes late in the larvai period occur after the developmental trajeaory is fixed; the absolute amount of energy devoted to development also is ixed. A change to high food conditions means that a smaii but increasing amount of energy is committed to finishing metamorphosis. A very large vro~ortionof the increased resources can be devoted to growth, which results in a large increase in size at metamorphosis (fig. 11.1E).A change to low food implies that a large and increasing amount of energy is required to complete the fixed developmental trajeaory. Essentially, all energy is devoted to development and none to growth, with perhaps some negative growth required to complete metamorphosis on the rapid developmentai trajectory (fig. 11.1E). A large decrease in body size is possible. N o asymmetries in response to increasing and decreasing food levels at different aaes are ~redicted. " The Aidence from three studies that have investigated larvai responses to increases and decreases in food level, in addition to supporting the modified Wdbur-Collins model as presented by Hensley (1993), also supports the graphical model of resource allocation discussed above. Recall that Leips and Travis (1994) prediaed asymmetrical effeas from chGging food levels at di-fferenttimes, whereas the graphicai model does not. Aiford and Harris (1988) found no asymmetry from changing food levels at different times. All groups that received increased food at different times were
I

not significantly different from each other in mass at metamorphosis. Two groups that received decreased food, including the 1 s t group to be switched, metamorphosed at the same mass. Hensley (1993) also found no asymmetry in responses. Early increases and decreases in food levels had their greatest effea on size at metamorphosis. Leips and Travis (1994) presented evidence that responses were asymmetric for Hyla cinerea but not for H. gratwsa. Body size (SVL) at metamorphosis did not differ significantly for all groups that received increased food levels at different times in both species. Two groups of H.gratwsa that received decreased food, including the last group to be switched, metamorphosed at the same s&. Earlier decreases in food level had more of an effect than late decreases in H. cinerea. N o asymmetry of responses was found in recent studies by Beck (1997) on H. squirella and Tejedo and Reques (1994b) on Bufo calamita. Metamorphosis does not seem to be possible until a minimum size is attained. A panadaptationist might suggest that metamorphosis at a very smaii size is better than death in a drying pond. Metamorphosis at a smaii size can decrease terrestriai performance (Emerson 1986; Pough and Kamel 1984; Taigen and Pough 1985).A developmental constraint may exist because adult organs may not form below a cem& size. The maximum size that a species attains at metamorphosis is very likely an adaptation to average opportunities for growth and mortality rates experienced on land. However, mapping metamorphic trait values on anuran phylogenies would be vaiuable as a means of assessing the relative role of seleaion and evolutionary history.

Ecological Genetics of Larva1 Traits

Several investigators (e.g., Newman 1988a, b; Travis et ai. 1987) provided a coherent picture of various aspeas of the genetics of growth rate and possible trade-offs with other fitness traits in larvai anurans. They used the breeding designs of quantitative genetics (Faiconer 1989) to assay for degree and type of genetic variation present and for genetic correlations of fitness traits. For a fuii-sib design, eggs are coiieaed from single, isolated matings; egg masses coiiected in the field will not guarantee that all larvae are fuii sibs. In a mating aggregation of Bufo americanus or Rana sylvatica, for example, it is possible that one female's eggs may be fertilized by two or more males (personai observation). A fuiisib design can provide considerable information relativelv easily, but several sources of variation in the response are confounded into the effect of family: additive genetic variation (necessary for a response to selection), dominante variation, and maternal effeas. The inability to isolate matemai effeas is a major problem because egg size variation could be caused by nongenetic factors (e.g., feeding history of the mother) and lead to a spurious estimate of the amount of genetic variance present in a population. Frogs could be raised through one or more acchation generations to elimitiate nongenetic maternal d e a s , but this usuaiiy is impractical given the long generation times. In general, a fuil-sib design would be appropriate as a pilot study to estimate vari-

T H E ANURAN TADPOLE

285

ances for use in a statistical power calculation to determine the number of replicates to use in futther studies. Half-sib designs and partial didel designs d o w partitioning of the effect of family into additive genetic, nonadditive genetic, and environmental components of variation. In the half-sib design, females are mated to a number of males. The effect of male parent on offspring fitness is free of any maternal effects and is used as a measure of additive genetic variation. Unless an acclimation generation is used, the estimate of nonadditive genetic -variation is confounded with nongenetic maternal effects. In a partial dialiel design, ali males are each mated to ali females and vice versa. The external fertilization of anurans facilitates a partial dialiel design. For practical reasons, it may be posGble to use relatively few parents. For example, an 8 females x 8 males design with 5 replicates would require 320 experimental units (e.g., pools or cages). The amount of maternal effect can be estimated by comparing the magnitude (sums of squares) of the male effect with that of the female effect. The male effect is an estimate of additive genetic variation; the female effect is an estimate of additive genetic variation plus maternal effects. Thus, maternal effects can be estimated by subtraction. Partial dialiel designs also d o w for estimation of nonadditive genetic effects (male x female interaction). In order to interpret the nature of genetic variation for a particular trait, the ecological context of the population must be considered (Travis et ai. 1985a). Any given value of a genetic parameter can be caused by several different evolutionary scenarios. For example, an estimate of zero genetic variante can be explained by strong directional selection or by a iow mutation rate cou~ied with weak directiond selection. Conversely, an ecological study alone cannot determine whether a trait is adaptive without a knowledge of genetic underpinnings. For example, the large body sizes of montane species is largely caused by the low temperature environment of the larvae (Berven et al. 1979). Studies of the ecological genetics of frogs have progressed to the point where both kinds of information are Dresent and credible scenarios of past selection can be presented.

Genetics of Growth Rate Srnith-Gdi and Berven (1979), Travis et al. (1987), and Wilbur and Cohns (1973) argued that growth rate is an important measure of larval fitness under almost any imaginable environmental regime and suggested that growth rate is a vahd measure of competitive abiiity. Individuals that grow rapidiy are the winners of competition; individuals that grow slowly are the losers. Slow growth places larvae at an increased risk of predation for longer periods of time and at an increased risk of dying in a drying pond (Newman 1989; Wilbur 1987). In an early experiment in the ecological genetics of anuran larvae, Travis (1980b) raised six full-sib famiiies ofHyla patwsa in competition with H.femoralis under laboratory conditions. Significant variation among the families of H. patwsa with regard to early growth rate (7)and length of larval period suggested possible genetic variation for those traits. Maternal effects were also suggested as the potential

cause of variation among famiiies. Individual tadpoles were considered as observations (degrees of freedom) in the statistical analysis, which led to a nonconservative analysis. The suggestion of genetic variation for growth rate was initialiy troubling because if the trait were so important to an individual's fitness, then growth rate should be maximized by selection, and genetic variation for growth rate should be exhausted. A resolution was suggested by the finduig that early growth rate and the length of larval period were negatively correlated, as were early growth and size at metamorphosis (Travis 1980b). Individuals that initidy grew rapidiy metamorphosed early at a smali size. Individuais that grew slowly early in the larval period metamorphosed later at a large size. This led to a positive correlation between size at metamorphosis and length of larval period. A trade-off between minirnizing time spent in the pond (reducing risk of predation and desiccation) and maxirnizing size at metamorphosis (important for terrestrial survival and later reproduction) could maintain genetic variation for growth rate. The combination of correlated traits evolves to a joint optimum value in the population (Via and Lande 1985), which is not the individual optimum of either trait. The presence of variation in growth rates between populations of H. dratwsa was confirmed by the results of four breeding episodes spread over three years (Travis 1983a). Fuii-sib families differed in growth rate in three out of four bouts of breeding, and the difference among f d e s tended to be consistent over two larval densities. No consistent relationship between average egg sizes and average growth rates was found. Thus, the effects of nongenetic maternal effects on significant among-family variation in growth rate were discounted. Within f d e s of interacting individuals, size at metamorphosis and length of larval period were positively correlated. Among famiiies, size at metamorphosis and length of larval period were negatively correlated. This means that rapidiy growing farnilies tend to develop quickly and metamorphose rapidly. Individuals of more slowly growing families on average metamorphose at a smalier size, but given proper environmental conditions, such as a constant or increasing food supply, they could attain large sizes at metamorphosis. In summary of fd-sib studies, variation in growth rate and usualiy in length of larval period exists among sibs in natural populations (Travis 1980a, b, 1983a). However, variation in size at metamorphosis is not always found. The differencesamong sib groups could have been caused by maternal effects, additive genetic variation, or nonadditive genetic variation. Experiments including half-sib or partial dialie1 designs could distinguish among these effects. In a half-sib design with PseudaEns criseriata, differences in growth rate, length of larval period, and size at metamorphosis were found among families in the laboratory (Travis 1981b). Differences in growth rate among families appeared due largely to nongenetic maternal effects. In another study involving a half-sib design using H.gratiosa, Travis ( 1980a) found genetic variation in growth rate and length of larval period, but not in size at metamorphosis. By the mid 1980s, genetic and nongenetic bases had been demonstrated for

286

REID N. HARRIS

growth rate, and species-specific differences had been found. Progress would be made when more elaborate breeding designs and ecological information were combined to infer the historv of natural seleaion. At 'least three extensive breeding designs (partial didels or chain block designs) have been published. Travis et al. ( 1987, Pseudmi crctcifer) and Newman (1988b, Scaphwpus couchii) found no or l i d e additive genetic variation but significant nonadditive genetic variation for growth rate. By pooling maternal and additive genetic variation, Travis et al. (1987) showed that their results were similar to those reported earlier; the implication was that nonadditive genetic effects for growth rate were important in earlier studies (Travis 1980a, b, 1983a, b). Blouin (1992b) found a positive heritability for larval growth rate in H y h cinerea but did not explicitly test for additive versus nonadditive genetic variation. In a partial &de1 experiment, evidence of significant additive genetic variation for growth rate was found in H. chrysoscelis (Alford 1986a, 1989a). Species-specific dBerences in the distribution of genetic variation for growth rate remain to be fully discovered and explained.

size at metamorphosis. Not enough information on the actuai genetic architecture of these traits exists, but the partial didel designs of Alford (1986a, 1989a), Blouin (1992b), Newman (1988a, b), and Travis et ai. (1987) show irnportant progress in this area. The pattern of existing underlying variation (additive, nonadditive, maternal) is expected to differ depending on past selective environment of the population (see discussion below).

As phenotypic characters, growth rate and length of larvai period invariably are inversely correlated. The correlations of growth rate with size at metamorphosis and length of larvai period with size at metamorphosis can be negative, positive, or zero and depend on envirormental conditions. Travis et ai. (1987) reported a positive additive genetic correlation between growth rate and length of larval period, but Newman (1988b) found a negative correlation between size at day 5 (early growth) and length of the larval period for Scaphiupus cowhii. This result should be viewed cautiously because Newman detected no additive genetic variation for size at day 5. Blouin (1992b) found no additive genetic corGenetics of L e n . of Lama1 Period relation between growth rate and length of the larval period In full-sib expenments, Travis (1980a) found significant in H y h cinerea. Travis et al. ( 1987) reported a positive addidifferences among sibships in length of larval period. Plytycz tive genetic correlation between growth rate and size at et al. (1984) reported differences among maternal half-sibs metamorphosis and between length of larval period and size in length of larval period. In more complex designs, l i d e at metamorphosis; they found a negative nonadditive correadditive genetic variation for length of larvai period was lation between growth rate and length of larvai period and found, but significant nonadditive variation was present in between size at metamorphosis and length of larval period. the populations of Pseudacris mcijer (Travis et al. 1987). A positive, nonadditive correlation existed between growth Working with Scaphwpus couchii, Newman (1988b) found a rate and size at metamorphosis. Newman (1988b) detected predominance of additive genetic variation for length of lar- no other significant additive genetic correlations and did not vai period, or the equivalent trait, developmental rate. Ber- estimate nonadditive correlations because none of his charven (1987) reported high levels of additive genetic variation acters showed nonadditive variation. Blouin (1992b) found for development time in lowland and montane populations a positive genetic correlation between length of larval peof &na sylvatica. Blouin (1992b) found a positive heritabil- riod and size at metamorphosis but no correlation between ity for length of larval period in H. cinerea. growth rate and either larial period or juvenile growth rate. Berven (1987) described a negative additive genetic correlaGenetics of Size at Metarnorphosis tion between developmental time and size at metamorphosis In full-sib designs, variation among sibs in size at metamor- in montane populations of &na sylvaticu but no correlation phosis -independent of variation caused by growth rate - in lowland populations. was found by Travis (1983a) in H.~ratwsa. This contrasted Antagonistic pleiotropy (M. J. Rose 1982) between with the results from a more limited study (Travis 1980a) in growth rate and the reciprocai of length of larval period was which no variation in size at metamorphosis was found. Two a key frnding of Berven (1987) and Travis et ai. (1987). Separtiai didel studies gave mixed results: Travis et al. (1987) lection wdi a a to maximize growth rate and minirnize length found significant additive genetic variation for size at meta- of larval period under many ecological conditions. Genes morphosis, but Newman (1988b) did not. Blouin (1992b) that lead to this desirable combination of traits wdi be fixed found a positive heritability for size a metamorphosis but quickiy and will not contribute to genetic variation for either did not test for nonadditive variation. Woodward (1986) trait. Genes that "code" for a slow growth rate and a rapid found no genetic variation among sib groups that differed development, or vice versa, will not be fixed but wiii come by male parent in size at metamorphosis but did find that to joint equilibrium in the population. This effect can exoffspring from larger sires were larger than offspring from plain the existente of genetic variation for either trait (table s m d e r dames 30 days after metamorphosis. Berven (1987) 11.1A, Falconer 1989). As Travis et ai. (1987) noted, neifound no additive genetic variation for size at metamorpho- ther growth rate nor length of larval period in t h s popusis in lowland populations of R. sylvatica but high levels of lation had much additive genetic variation that needed to genetic variation in montane popdations. be explained. Berven (1987) also uivoked the explanation of In summary, variation among fll sibs exists routinely in antagonistic pleiotropy to explain the maintenance of high the fitness traits of growth rate, length of larval period, and levels of additive genetic variation for development time and

T H E ANURAN TADPOLE
Table 11.1 Heuristic rnodels that explore the outcorne of selection on larvai and adult traits. (A) Depiction of a rnodel by Faiconer (1989) and (B-D) extensions of that model. Fitness vaiues for genes affecting larvai (horizontai axis) and adult (verticai axis) traits are shown in left pane1 of each section. Genetic combinations that are h e d , selectively eliminated, or rernain neutrai are shown in right panel. A. Correlated selection on larvai and adult stages: no selection for coupling or uncoupling of stages, and the life cycle is cornplex. Genes affecting the same trait in the larva and adult in favorable and unfavorable ways are given fitness values of + 1and - 1, respectively. Fitness Values Fixed Genetic Cornbinations that Are Eliminated Neutrai

B. Selection on the adult stage, uncouplmg of stages, and no evolution in the larvai stage; the life cycle is complex. Genes that affect a trait in the adult in favorable and unfavorable ways are given fitnesses of 1 and - 1, respectively. Genes that affect the trait in the larva in any way are selected against (fitness of - 1). There is selection for uncoupling (+ 1) and against coupling (- 1) of the trait in larva and adult. Fitness Values Fixed Genetic Combinations that Are Eliminated Neutrai

C. Selection on the adult stage in a species with direct developrnent. Genes that affect the trait in the adult in favorable (+) and unfavorable (-) ways are given fitnesses of 1and - 1, respectively. Genes that affect the trait in the ernbryo in any way are selectively neutrai. There is no selection for uncoupling or against coupling of the trait in larva and adult. Fitness Values Fixed Genetic Combinations that Are Eliminated Neutrai

D. Selection on the adult stage for uncoupling of stages and against any cturent or past correlated evolution of the ernbryo; the life cyde is now cornplex after a period of direct developrnent. Genes that affect the adult in favorable (+) and unfavorable (-) ways are given fitness vaiues of 1and - 1, respectively. Genes that affect the larva in any way are selected against (fitness of - 1). There is selection for uncoupliig and against coupling of the trait in larva and adult as in B. Fitness Vaiues Fixed Genetic Cornbinations that Are Eliminated Neutrai

size at metamorphosis in montane populations of Rana 9 1

vatzca.

The work ofTravis et ai. (1987) included a principal component analysis (PCA) the phenotypic, additive, nonaddiof tive, and environmentalcorrelation matrices for growth rate, length of larval period, and size at metamorphosis. For the phenotypic variation, the first principal component that

maximized the amount of variation explained aiso refleaed a gradient between large size at metamorphosis and rapid growth to s m d size and slow growth. Length of larval period did not come into play in the PCA but was inversely related to growth rate and positively correlated with size at metamorphosis. Tadpoles that grew rapidly achieved a maximum size in a short time and metamorphosed.

288

REID N. HARRIS

Anurans possess morphologicay and ecologicaily divergent larval and postiarval stages, each with an array of specializations. Larval specializations include keratinized teeth, a Summaly buccal pump mkhanisrn, a long coiled intestine, and caudal Estimates of genetic variation can be coupled with ecological locomotion. Many of these specializations are related to information to construct probable scenarios of past selec- feeding. Adult specializations include lack of a tail, fsed tion, adaptation, and constraint. For example, Travis et al. postsacral vertebrae, an elongated iliurn, a radius h e d to (1987) reported nonadditive genetic variation for length of the ulna, and a tibia fused to the fibula. These adaptations larval period, whereas Newman (1988b) did not. Although are related to saltation. At metarnorphosis, the larval ta deBerven (1987) and Blouin (1992b) did not estimate nonad- generates; and the locomotory, digstive, and nervous sysditive variation, they found additive genetic variation for a tems are largely or completely replaced (Aiberch 1987; Fox correlate of larval period length. These differences are related 1981; Kohos 1981). In contrast, the shortened presacral to ecological factors. The desert ponds containing Scaphiq~us vertebral colurnn remains constant across stages. How can such divergent iife-history stages evolve? The that Newman studied were arguably more variable in duration than the ponds studied by Travis, Berven, and Blouin. origin of a free-living larval stage can be explained with the Newman (1988b) docurnented a negative correlation be- phylogenetic nuii hypothesis, i.e., anurans evolved from antween developmental rate and plasticity in growth rate. cestors that had a complex iife cycle. The ancestral larval to Rapid developers did not delay metamorphosis if growth postlarval size versus shape trajectory is thought to have inrate increased and could not take advantage of the increased cluded a minor metamorphosis (Aiberch 1987; Bolt-1977; growth rate. The cornbination of slower development and Eaton 1959; Noble 1925a; Orton 1953; Wassersug and plasticity in growth rate may be favored in pools of longer Hoff 1982; fig. 11.2A). I define a minor metamorphosis as duration. Spatial and temporal variations in seleaion means minimal variaiion between manv traits in the larva &d oostthat no particular developmental rate is favored, and no larva. Compared with frogs, salamanders have a minor metadorninance variation, and therefore no dorninance covari- morphosis (Harris 1989; Reay and Lauder 1990). A nunor met&norphosis seems to have.occurred in the dissorophids ance structure, is expected in Scaphzopus. The presence of additive variation for the inverse of the (Bolt 1977, 1979), the presumed ancestors to modern amlength of the larval period in Rana sylvatica was explained phibians, but the question of anuran origin is still unrein two ways (Berven 1987). In lowland populations, size at solved. Three scenarios for the evolution of divergent phemetamorphosis is canaiized. Large size at metamorphosis is notypes in larval and postlarval anurans are (1) the larval correlated with earlier age at first reproduction. For short- stage evolved specializations before the postiarval stage, (2) iived colonizing species such as lowland wood frogs, age at the postlarval stages evolved speciaiizations before the larval first reproduction should be minimized. Larvae are selected stage, or (3) specializations occurred in both stages simultato attain a large size at metamorphosis regardless of develop- neously. ment time, which would be under weak selection. In mon~elction specializations in the adult stage wdi cause for tane populations, the inverse of larval period length shows a ailometric and other changes in the larval stage unless a state negative genetic correlation with size at metamorphosis. of independente or lack of correlation exists between stages. Early metamorphosis that d o w s terrestrial growth or late Change in a size-shape trajectory in the adult stage, in eimetamorphosis that allows a large body size at metamorpho- ther elevation or slope, will affect the larval trajectory (fig. sis wiii lead to a large juvenile body size. This may enhance 11.2B).Adaptive evolution in the adult stage would produce ovenvintering survival in the rigorous montane environ- correlated responses in the larval stage that may not be adap-

Environmental correlations were also analyzed with PCA. The first principal component described an axis with slow growth, long larval periods, and large sizes at metamorphosis at one extreme, and rapid growth, short larvai period, and sma sizes at metamorphosis at the other. These environmental effects can be produced by temperature variation and some types of competitive environments. The PCA on the nonadditive correlations were of considerable interest because theoretical population genetics predias that the outcome of past selection is i~kely be reflected in these to correlations (Travis et ai. 1987). Favorable genetic combinations should be protected from recombination by inversion that would cause linkage or gametic disequilibrium and be refleaed by an estimate of nonadditive genetic variation. The first principal component described an axis with rapid growth, short larval period, and large size at metamorphosis at one extreme. This was contrasted with slow growth, a long larval period, and sma size at metamorphosis. In many ecological situations, this mirrors a selection gradient from traits that are adaptive to traits that are not.

ment. In both environments, large size at metamorphosis is favored by selection, but the different genetic architectures that have evolved suggest that multiple adaptive solutions are possible (Harris et ai. 1990). Quantitative genetics, when combined with ecological information, has yielded important insights into adaptation and constraint of larval anuran fitness traits. Aithough not enough studies have been conducted to ailow any broad generalizations, it seems that (1) half-sib and diaiiel breeding designs are necessary because nonadditive genetic variation is an important quantity to measure, (2) genetic correlations must be estimated in order to evaluate hypotheses about adaptation and constraint, and (3) ecological information must be combined with genetic data to generate a credible analysis of past selection pressures.

Origin of a Complex Life Cycle

T H E ANURAN TADPOLE

289

A 1

Ancestral trajectory

Ancestral trajectory

Ancestral traiectorv

D Specialized
b ~ e c

Ancestral trajectory i a l i z adult e

,
SIZE

--.

Fig. 11.2. Graphicai model to iiiusuate the evolution of divergent Me history stages in amphibians. (A) Ancesuai size-shape uajectory that leads to little phenotypic vanation between stages. Metamorphosis occurs at the break in the trajectory. (B) Selection for a specialized adult reduces the slope of the larvai trajectory if the developrnentai programs are correlated. Metamorphosis occurs at the breaks in the trajectories. The dashed lines in the late larvai and adult trajectones indicate new shapes that are not characteristic of previous ontogenetic stages. (C) Selection for a s p e c i k d adult when the developmental programs are uncoupled. The larva of the specialized adult foiiows the ancestral larvai uajectory. Metamorphosis occurs at the breaks in the trajectories. The dashed line in the adult trajectory indicates a new shape that is not characteristic of previous ontogenetic stages. (D) Once uncoupling occurs, the larva aiso can diverge from the ancestral trajectory. Metarnorphosis occurs at the breaks in the trajectories. The dashed lines the adult and larvai trajectories uldicate new shapes that are not characteristic of previous ontogenetic stages.

tive. Similarly, adaptive evolution in the larval stage would produce correlated responses in the adult stage that may not be adaptive. To the degree that maladaptive morphologies are produced, selection should occur simultaneously for an uncoupling of stages (R& 1987). A partial uncoupiing could speed the rate of speciahzation in both stages. As each stage comes to sit on different "adaptive peaks," ieleaion for specialization and uncoupiing would stop. Developmentdy, uncoupling of stages may occur by compartmentalizationof embryonic cels. One group of cels Merentiates into the larvalalstructure. whileano;her u grouv of cels remains undderentiated. At metamorphosis, the previously undifferentiated ceii iine produces the adult structure (Aiberch 1987). 1s it vossible to selea simultaneouslv for svecialization in one stage and for uncoupling of stages? Simultaneous selection for two traits eventuaily leads to a negative correlation between the traits ( ~ a k o n i r1989). In Fdconer7s model (table 11.lA), the two traits would be the same trait in both the larval and postlarval stages. A negative correlation arises because pleiotropic genes that &ect the trait in the Iamal and postlarvai stages as desired (defined here as wiii come to iixation. Aiso, genes that &ea both traits in the undesired direction (-/-) wiii be selectively eliminated, which means that much of the remaining covariation is
L

caused by genes that &ect one trait favorably (+/-) and the other trait unfavorably (-/+) and are selectively neutral. Thus, the heritability of the traits in the larval and postlarval stages simultaneously becomes zero and wiii not respond to joint selection (Falconer 1989; M. J. Rose 1982). The two traits under selection in the evolution of the protoanuran Me history are a hypothetical trait (e.g., characteristics of the digestive tract) in one Me-history stage and a uncoupling of that trait in the other stage. Decoupiing is considered a trait in t h s model, and uncoupled traits are defined as traits that are genetically and developmentally undcorrelated. Evolution of the trait in one stage wiii not &ea evolution of that trait in another stage. I extend the heuristic model by Fakoner (1989) to include the case of selection for specialization of an adult trait and for uncoupling of the trait in the larva and in the adult. I define total fitness of the genotype as the genotypic effect on the trait in the adult plus its effect on the same trait in the larva plus its effect on the correlation between the trait in the larva and in the adult. It is assumed that any correlated evolution (allometry) in the larval stage is maladaptive (fig. 11.2B). This means that an early larva that begins to develop specializations of an adult fiog instead of maintaining the ancestral larval form is selected against. In this example, the frog larva would begin to develop a carnivorous digestive tract before metamorphosis. Thus, we can assign genes that &ea the adult trait in the desired direaion (+) a fitness of 1, genes that &ect the adult trait in the undesired direction (-) a fitness of - 1, and genes that cause evolution of the larval trait in any direction and therefore away from its adaptive peak (+ or -) a fitness of - 1. A correlation between traits is assigned a fitness of - 1, and no correlation between traits is assigned a fitness of 1 (table 11.1B). For example, the O / + genes have a fitness of 2 (+ 1 for &ecting the adult trait in the desired direction and + 1 for causing no correlation between the trait in the larva and adult) and wiii move toward iixation. The +/+ and -/+ genes have a fitness of - 1 and the +/- and -/- genes have a fitness of -3. Genes with negative fitness values wiii tend toward selective elimination. Combinations of -/O, +/O, and O/are neutral. This heuristic model has two outcomes: (1) selection wil uncouple larval and adult traits, and (2) neither stage wiii evolve because + and - genes that are iixed or neutral balance each other. The o/- combination is neutral and wiii not be seleaed against; it lirnits evolution of the adult in the desired direaion. Simultaneous selection for adult specialization and uncoupiing leads only to uncoupling. Once uncoupling has evolved, presumably via compartmentaiization of embryonic cels, mutations can arise that would d o w the evolution of adult and larval speciaiizations. The simple model does suggest that simultaneous seleaion for evolution in one stage and uncoupling is not likely to be effective. The same result occurs when simultaneous selection occurs for larvai speciaiization and uncoupling; uncoupling evolves but neither stage changes. Alternatively, direa development is a developmentai mode that potentidy can facilitate postlarvai speciahzations

290

REID N. HARRIS

and uncoupiing (Jagersten 1972). Direa development removes a larva from manv of the faaors of external selection present in the environment of a free-iiving larva (Lutz 1948). Seleaion for adult specialization would produce correlated larval morphologies that may be maladaptive, but such morphologies would be exposed to weak selection. The genes that coded for adult speciahzations and larval evolution, +/+ , -/+, and o/+, would tend toward ixation (table 11.1C). The neutral combinations of -/O, +/O, and 010 would covary. Adult evolution occurs, but the embryo does not evolve (fig. 11.1C). Should seleaion again favor an aquatic larva, the embryo (now tadpole) would be exposed to seleaion. With seleaion for uncoupling of larval and postlarval stages, the +/+ and -/+ genes would be eiiminated rapidly (table 11.1D). The loss of + genes that d e c t adult specialization in the desired direction would lead to regression of the adult specialization. After selection, only the +/O and -/O genes would contribute to the covariance because the O/- genes had been selectively eliminated. What are the differences between simultaneous selection for a change in adult specialization and uncoupling in species with a complex Me cycle (table 11.1B) as opposed to a species that secondariiy evolved a complex iife cycle from direa development (tables 11.lC, D)? The simple models predia that in both cases the O/+ genes wiil tend toward ixation and that the evolution of uncoupling of traits has occurred. One difference is that the O/- genes are neutral in the simultaneous selection case but eliminated selectively in the direa development scenario. Selection for uncoupli& after direa development allows a greater initial speciaiization in the adult stage. This may make it more probable that a complex Me cycle with a major metamorphosis will evolve from an ancestor with direa development. This sirnple heuristic model needs to be developed quantitatively in order to explore further the evolution of divergent life history stages. The model suggests an alternative pattern to the one proposed for plethodontid salamanders. Based on work with Eulycea, Aiberch (1987) proposed that compartmentahzation of larval and adult development programs may have d o w e d a suppression of the larval structures by direa developers. If true, compartmentahation preceded direct development. The model presented here for protoanurans suggesis that direct development preceded selection for compartmentahzation. 1s there any evidence that direct development did or did not facilitate the evolution of divergent Me stages in anurans? Can direa development be considered primitive in frogs? I argue that if direa development is primitive, then an analysis of Leiupelmu, which retains many primitive traits, or a fossii ancestor should reveal a larval size-shape trajeaory that retains prirnitive, perhaps salamander-like traits. If dir e a development is derived, then Lewpelma and other species with direa develo~ment should have a iarval size-sha~e trajectory that retains some trace of a tadpolelike trajeaory. Most workers have considered a complex life cycle to the primitive in anurans ( D u e h a n 1989; Noble 1925a; Orton 1953; Szarski 1957; but see Bogart 1981). The argument
I

for primitiveness is that most anurans have complex iife cvcles and that most fish and salamanders also have com~lex life cycles. However, commonness of a trait cannot be used to establish a charaaer polarity. For example, most mamrnals are placentals, but it is generaliy concluded that the placenta is a derived charaaer. The life cvcles of the coelacanth or lungfishes are not helpful in det&mining what life cycle is primitive in frogs. The crossopterygian Latimeria has direa development and the lungfishes have complex Me cycles. Fossii evidence does not seem conclusive in determining whether a complex life cycle is primitive in frogs. Bolt (1977) suggested that the Dissorophoidea is ancestral to the three modern groups of Lissamphibia. The evidence includes simiiarities in the pedicellate teeth of both groups. Adult and juvenile remains of the dissorophids have been found. ~ o lproposed a paedomorphic origk of the modern t amphibians based on their sirnilarity to juveniie dissorophids. It cannot be concluded that dissorophids had a freeliving larval stage because larval remains have not been found. Protoanurans are known from the fossil record. but larval stages of these transitional forms are not known. The Recent frog genus that retains the most primitive traits may be Leiopelma, and it has direa development (exovivipary of Aitig and Johnston 1989) to various degrees. Can it be determined if direa development is primitive in Leiopelma? If direa development is derived in Lehpelma, then the prehatching stages should have some trace of a tadpole trajeaory, although extensive ontogenetic repatterning associated with direa development may have occurred. It is ~ossibie the tad~ole that traiictorv ma;, be truncated relative , to a free-living tadpole. If direa developrnent is primitive, then the trajeaory of the prehatching stages of Leiopelma should begin as a primitive (perhaps Urodela-like) trajeaory, then diverge toward the adult trajeaory (fig 11.2B). The literature has contradiaory reports on the trajeaory of Lewpelmu. Eaton (1959) and N. G. Stephenson (1951a, 1955) Stated that Leiopelma does not resemble a tadpole at any stage during its development, and N. G. Stephenson commented that prehatching Leiupelma resemble caudate larvae. Larval Lewpelma have a medially expanded ceratohyal, which would suggest a derived direct development (Wassersug 1975). Further analysis is needed in the form of a series of complete size-shape trajeaories for a variety of traits in Leiopelma. Even if ~eio~elma out to have a derived direa deidor>turns ment, the h o t h e s i s that direct development can facilitate' a phenotypic divergente between iife stages remains testable should suitable fossil material be found. The argument that a complex life cycle cannot be derived from direa development seems to be false as a general d e . Some members of the hemiphractine genus G ~ o t h e c a apparently evolved a complex life cycle from ancestors with direa development ( D u e h a n 1989; Wassersug and Duellman 1984). It is unclear ifa primitive complex life cycle reappeared or if the complex life cycle is in some way derived. I' f ~ u e h a n ' phylogenetic hypithesis is supported, it suggests s that several pathways are possible in the evolution of anuran developmental modes. I stress that the suggestion of direa development being
I

T H E ANURAN TADPOLE

29 1

ponds ddfer in mean phenologies of primary productivity cannot be adequately addressed with current information. Predation pressuie on tadpoles rnay be less in recently filled ponds than in permanent ponds (Wilbur 1980,1987). Upon filling, a temporary pond rnay be a relatively predator-free environment. Fish wouid be absent, but saiamanders and other ~redatorsmight be important. In western " Virginia, newts (Notophthalmusvi&scens) migrate to srnd montane ponds in time to eat hatchlings of the wood frog (GLU1978, Rana sylvatica; personai observation). IndividuMaintenance of the Complex Life Cycle ais of a species of ~ i r e n (caudata: Sirenidae) emerged from food-rich aquatic habitat, a hgher probabiiity that free- a dried pond bed in the southeastern United States within living larvae wdl escape predation or parasitism better than two days of refiiiing (personaiobservation) and began eating embbos of direct deGelbping species, -and the increased op- tad~oles. Predaceous insects mav be scarce in a recentlv N e d portunity for kin selection in free-living larvae compared to temporary pond, but predators of s m d tadpoles, such as embryos in an egg mass can be associated with high larvai diving spiders, rnay colonize quickiy. Unless saiamanders growth and survivai and therefore might promote the reten- and insects are much less efficient Dredators than fish. it is tion of the larvai stage. It has been suggested that the mod- not clear that recently filled temporary ponds are characterem tadpole rnay be an adaptation to exploit abundant but ized by lower predation pressure than permanent ponds. ephemeral resources found in temporary pools (Wassersug More study is needed to compare predation on tadpoles and 1975; Wibur 1980; Wilbur and Colhs 1973). Indeed. the embryos in temporary ponds, permanent ponds, and terspecies diversity of anurans is higher in temporary ponds restriai habitats. If predation pressures vary systematically than in permanent ponds in the southeastern United States among temporary ponds, perrnanent ponds, and terrestriai (Wilbur 1987; personal observation). Free-living larvae rnay settings, then predation can be invoked as an evolutionary have enhanced survival in the presence of predators because factor that selects for larva1 habitat. Pond size is an important they can attain a size refge f;om predat&-sand aiter their variable; extremely srnd temporary pools (e.g., the size of activity level to avoid detection by predators (Skeiiy 1990, cow hoof print) wi not contain fish or salamanders but wiii 1991). Aithough the increased opportunity for kin selection retain waier long enough for some species of tadpoles to caused by behavior of tadpoles has not been suggested as a metamorphose (e.g., Gastrophiyne carolinensis and Physaluereason for the maintenance of a complex life cycle, behav- muspustulosus). ioral ada~tations wouid have the effect of increasing: the fitDirect development rnay be favored when predation or V ness of individuais or family groups that retain the complex parasitism is higher on free-livingtadpoles than on direct delife cycle. Kin selection aiso acts on antipredator responses velopers (Dueliman and Tmeb 1986; Lutz 1948; Magnusson and Hero 1991). Parentai care can interact with direct (chap. 9). Aithough some have asserted that temporary ponds pro- development to deter predation and lower the incidente of vide a flush of primary productivity as they fill and recently parasitism (McDiarmid 1978). Criticai data on these points faiien leaves and organic matter become saturated and re- are lacking and a need exists for frther investigation. Direct lesse nutrients (Wassersug 1975; Wilbur and Collins 1973), development is more prevaient in the tropics and might be little limnologicai evidence supports their claim. Nutrients related to the intensity of aquatic predation. Average presurely are released upon pond fiiiing (Osborne and McLach- dation rates in temporary ponds in tropicai and temperate lan 1985); but the rate of that release, and the relationshp zones rnay be so different that a comparison wouid be between the release rate and ~rimarv ~roductivitv avdable profitable.- Vermeij (1976) compared predation on gastroto tadpoles, is debatable. In faa, larvai resources might be at pods in the Pachc and Atlantic Oceans, and both prey and their highest level as the pond dries. At that time nutrients predators in the Pacific have evolved stronger defenses and become concentrated, and a flush of primary produaivity predatory abiiities, respectively. An increased opportunity for kin selection rnay favor the rnight occur. Before the hypothesis thai a cokpiex life cycle is an adaptation to exploit a flush in primary produaivity can retention of a complex life cycle. An individuai's inclusive be addressed criticdy, studies on the phenology of primary fitness is composed of its individual fitness and its effect on productivity and its avaiiabiiity to tadpoles in temporary the fitness of other related individuais (Hamilton 1964a). ponds are needed. Kin selection among embryos in a terrestriai egg rnass might Conversely, Wassersug (1975) argued that the larvai stage lead to the evolution of a communaiiy produced toxic subrnight be abandoned when ponds do not have a predictable stance. However, I argue that the opportunity for kin sein productivity. ~ r o ~ i ponds rnay lack a predictable leaion is greater among free-living tadpoles because the be&l pulse in productivity, and that rnay have been one selective haviorai repertoire is greater. Options such as schooling pressure for direct development in the tropics. Data on the behavior and release of aiarrn chemicais are not avaiiable to phenology of productivity of tropicai ponds aiso are l a c h g , ernbryos restricted to an egg mass. Enhanced behaviorai posso a meaningfui comparison with temperate ponds is impos- sibilities can increase an individuai's inclusive fitness by aisible at this time. The hypothesis that tropicai and temperate lowing individuais to aid related individuais. Kin selection on a sirnple genetic model (sensu Faiconer 1989) of the evohition of morphologicaiiy divergent life-history stages. A broad anaiysis of morphologicai transitions among a variety o metarnorphic and direct developing anurans is needed. f Combining this anaiysis with a weii-supported phylogenetic scheme will be essentiai in resolving whether direct development is a primitive or derived trait in the Anura.
i l

a primitive feature of "protoanurans" is a hypothesis based

292

REID N. HARRIS

has played an important role in the larvai biology of many pended aigae. How resource concentration would interact species (chap. 9), and this rnay lead to an overali increase in with relative size of the filtering surfaces to affect size-specific fitness to the individual with a complex life cycle in environ- growth rate needs investigation. ments that aliow it. A comparison of size-specific growth rates of ranid tadComplex life cycles have been stable over long periods poles tends to support the hypothesis that a large body size of evolutionary time (Bolt 1977; Wassersug 1975). A theo- decreases the efficiency of suspension feeding. Werner reticai model by Istock (1967) suggested that such stability (1986) reviewed data for severai species of&na and found is unexpected, thus creating Istock's paradox. A critical as- that size-specific growth rates increase and then decrease sumption of Istock's model was that no genetic correlation over the larvai period. The decrease in growth rate as body exists between traits in both stages thereby allowing inde- size increases late in the larvai period is not associated with pendent evolution. This leads to the suggestion that stages the typicai depression in growth rate at metamorphosis. d differ consistently in rates of adaptation. Genotypes as- These data are consistent with a mechanistic explanation of sociated with a reduction or elimination of the less adapted decreasing efficiency of tadpole feeding as body size instage wiil be favored. The assurnption of genetic correlition creases and with the hypothesis that larvae (see above) d o has been evaiuated in only two studies with anurans (Blouin cate greater proportions of energy to development late in the 1992b; Emerson et ai. 1988) and is discussed below. Istock's larvai period. However, some species can grow exponenparadox can be resolved with the foliowing three explana- tially throughout their larvai period if resources are adequate tions: functionai constraints, fluctuating selection, and ge- (e.g., ALford and Harris 1988, Bufo woodhousiifwleri). Exponentiai growth implies a constant size-specific growth rate netic correlations. Functionai constraints of tadpole morphology and physi- throughout the larvai period (see eq. [I]). ology rnay lead inevitably to metamorphosis. Anuran tadUnder some ecologicai conditions, aquatic growth as a poles become less efficient feeders as they grow. Wassersug large larva, with its relatively smalier filtering surfaces, rnay (1975) argued that the surface area-to-volume ratio of the stili carry a higher fitness than metamorphosis to a terrestrial filtering surfaces of a tadpole decreases as biomass of the tad- stage at a smder size. This would select for paedomorphosis pole increases, and this creates less efficient feeding at in- or the relative lengthening of the larvai stage. An unfavorcreasing size. This ratio creates an upper limit to the size of able terrestrial environment would create such a selective rethe branchiai basket in tad~oles ali s~ecies. order to gime. More snidies that integrate the relationship of larvai of In overcome this constraint, evolution in the tadpole would size to filtering efficiency, assimilation percentage (Aitig and have to proceed via a peak shift (an ontogenetic change in McDearman 1975), metabolic costs of feeding, and specific trophic morphology and behavior tantamount to a meta- growth rates are needed. For example, tadpoles of species that metamorphose at a small body size (e.g., Bufo) apparmorphosis; S. Wright 1931). Morphological anaiyses provide an explanation for the ently can maintain a constant size-specific growth rate while decreasing trophic efficiency hypothesis. Oral trophic stmc- those of species that metamorphose at a large body size (e.g., tures show isometric growth or negative dometric growth Rana) cannot. Functionai constraints in the form of decreasing trophic with body size (Wassersug 1976a). In the latter case, feeding structures become relatively smder as the tadpole increases efficiency with increasing larvai body size can provjde a parin size. A positive allometsr of filtering surfaceand body size tiai resolution of Istock's paradox. The tadpole stage is would lead to a tadpole with an increasingly crowded body thought to be an adaptation to exploit ephemerai but favorcavity. Wassersug (1975) suggested that there rnay be an up- able growing conditions, aithough the isometry of filtering Der limit to the size of the branchiai basket: the branchiai surfaces to body size rnay lead to a slowing of further growth I baskets of large Pseudis paradoxa tadpoles are no larger in (Wassersug 1976a). Wassersug's argurnent leads to the sugabsolute terms than those of &na catesbeiana larvae. gestion that the adult stage wiil displace the larvai stage over Wassersug (1975) tested the hypothesis that large body evolutionary time because the rate of adaptation slows in the size decreases the relative size of the filtering surfaces and larval stage (i.e., larvae are as "adapted" as they can get bethus decreases feeding efficiency. He found that 12 smali cause of allometric constraints) but not in the adult stage. Ranapipiens tadpoles cleared their guts more rapidly than This resolution rests on the assumption that large size in a three large tadpoles of the same combined weight. Wasser- tadpole is always adaptive. The existente of species that typisug (1975) did not measure the metabolic costs of feeding caliy metamorphose at a smali size, such as Buf, suggests in small and large individuais. Large tadpoles rnay be less that metamorphosis at a smali body size can be an adaptaefficient feeders, but they rnay spend less energy feeding. If tion. Furthermore, the rate of larvai adaptation can proceed true, the net energy return per unit time rnay have been aiong numerous other axes, including the evolution of greater in the large tadpoles. Further experiments with a va- mechanisms of predator avoidance, increased competitive riety of species with the same and different feeding modes abdity, and digestive efficiency. need to be conducted to test W y the hypothesis. For exWassersug (1975) considered life-cycle evolution in diample, R. pipiens has keratinized mouthparts, which suggests chotomous terms: either the population retains or eliminates that the species rnay not function primarily as a fiiter feeder a stage. It seems more likely that selective pressures wiil op(R. Aitig, personal communication). Sede and Beckvar erate to increase or decrease the relative durations of a stage. (1980) and Sede and Wassersug (1979) showed that tad- If selection favored the larvai stage for a long period, two poles can adjust filtering rates to the concentration of sus- outcomes are conceivable. One would be the elimination of

THE ANURAN TADPOLE

293

the adult stage (paedomorphosis). As Wassersug noted, this relation between lengths of the two stages is of interest. If would require a reworking of the tadpole body plan to make longevity is constant among individuais, then the length of space for eniarged gonads, to aiter amplexus or other mating larvai and adult stages necessarily are negatively correlated, behaviors, and to overcome a lack of emetic behavior that is but longevity in the larvai and adult stages might be posirequired for oviposition and sperm extmion (Naitoh et ai. tively geneticaiiy correlated. Individuals with longer larval 1989). Paedomorphic tadpoles may not be an evolutionary periods might aiso live longer after metamorphosis. If there possibility. This is a consequence of the high degree of diver- is no genetic correlation between length of larval stage and gente between life stages. The same phenomenon occurs in length of adult stage then persistent or fluctuating variation insects; the oniy truiy larvai reproduction in insects is asso- in length of larval period wdi have iittle effect on evolution ciated with par;henogenesis (H. F. Nijhout, personal com- in the adult stage and therefore persistence of a complex iife munication). Wassersug's question, "Why there are no pae- cycle. This has not been estimated in anurans, and its estimadomorphic tadpoles?" could be foliowed by the question, tion in a variety of species would be of interest. "Why are there no parthenogenetic tadpoles?" The occurrence of genetic correlations for Me-history The relative lengthening of the larvai stage, resulting in traits across stages provides a t h r d resolution of Istock's para single breeding period shortly after metamorphosis (e.g., adox. The possibility of antagonistic pleiotropy (M. J. Rose ephemeropteran insects; Istock 1984) is another possible 1982) could explain the maintenance of a complex Me cycle. evolutionary outcome of a highly adapted larvai stage. This If the genotype that codes for a complex of adaptive adult hypothesis could be tested by geographical comparisons. For traits does not code for the most favorable complex of adapexample, one might expect that populations that occupy fa- tive larvai traits, then genetic variation for those traits and vorable adult environments but consistentlv use I Door larval the concornitant complex life cycle could be maintained. A environments should have a relativelv short larvai period. polymorphism would be achieved as two solutions come to Populations from the center of a geographical range may oc- an equilibrium: less adapted larvae and more adapted adults cupy equally favorable larval and adult habitats. Populations versus more adapted larvae and less adapted adults. If geat the edge of the range iikely exist in a poor larvai or poor netic correlations across life-history stages exist, it would be adult habitat. The goal would be to identi@ areas where the unlikely or impossible for one stage to evolve at a faster rate. environment was iimiting only on the larva or only on the This hypothesis could be tested with a conventionai quantiadult. A geographical comparison of the relative durations of tative genetic anaiysis, such as a ha-sib design that foliows larvai and adult periods would be of great interest. Indeed, tadpoles into adulthood and measures trait values for both Berven et ai. (1979) found that montane populations of stages (Faiconer 1989). Additive genetic correlations beRana clamitans have evidence of past selection for a shorter tween larvai and adult Me-history traits could be estimated larval period relative to lowland populations. The montane (e.g., Emerson et ai. 1988). Selection experiments on larvai ponds arguably are poorer larvai environments than lowland and adult traits with an assay for correlated selection reponds. The techniques of common garden (e.g., Berven sponses of larvai and adult traits aiso could be used. The long et ai. 1979) and reciprocal transplant experiments could be generation time of many anuran species would make selecemployed to test whether any observed differences have a tion experiments problematic. Species with generation times genetic basis. Because of the long generation times of an- of one year would be good candidates for a test of the antagurans, these experiments would be far more practicai than a onistic pleiotropy hypothesis. As argued above, anuran evoselection exper&ent. lution probably has been characterized by a genetic or develIstock (1984) postulated another resolution of Istock's opmental independence between stages. Such independence paradox: persistent genetic variation in length of larvai pe- would allow divergence of stages and would allow each stage riod could preserve a complex Me cycle. When ecological to adapt to different environments. If true, the antagonisfactors periodically favor lengthening the larvai period via tic pleiotropy hypothesis is unlikely to provide a resolution fluctuating selection, genetic variation for the length of the to Istock's paradox. Viability selection acts on the larval larvai period is maintained. If length of the adult stage is stage before viability selection or fecundity selection acts inversely related to length of the lakal stage, then such selec- on the adult. Whether this could create a differential rate of tion would concomitantly decrease the length of the adult adaptation in favor of the larval stage is unclear because stage. Fluctuating selection might aiter the relative lengths such difficult models have not been constructed (Hartl and of the two stanes. but neither stane would be eliminated if Clark 1989). the larval period were periodicaiiy favored. Fluctuating selection is one factor that leads to the maintenance of genetic Experimental Approaches to the Evolution variation (Hartl and Clark 1989), in this case for length of of the Complex Life Cycle larval period. ~ h fluctuating-selection explanation could be addressed Quantitative Genetics e first by surveying populations for genetic variation for A central characteristic of the anuran complex Me cycle is a length of larval period. Such genetic variation has been phenotypic divergence between larval and postlarval stages. found in some populations (Berven 1987; Blouin 1992b; I have argued that uncorrelated developmental programs are Newman 1988b; P h c z et ai. 1984; Travis et ai. 1987). Sec- necessary for such divergence to have evolved. Support for ond. the assum~tion the leneths of the larval and adult this hypothesis would be a quantitative genetic analysis that that I periods are inversely related could be examined. Genetic cor- estimates no genetic correlations between a trait in the larval
i
V , V

294

REID N. HARRIS

The complex iife cycle of anurans is characterized by a phenotypic divergence between larval and postlarval stages. Selection for specialization of one stage in protoanurans may have been accompanied by seleaion for uncoupling of traits in the other stage. Genetic and developmentalindependente is expected between stages that d o w independent evolution.-~irect developmen; may have facilitated the origin of a divergent complex life cycle by dowing greater initial addt specialization. The divergent larval stage (tadpole) seems to be specialized for growth. Adaptations that promote rapid growth and provide an optirnal timing of metamorphosis are expected and have been examined by many workers. The evolutionary stability of a complex life cycle has been questioned (Istock 1967), and several solutions have been discussed. Once highly divergent life cycles evolve, elimination of the addt stage may be impossible without parthenogenesis. A parthengene;ic solu6on has been exiloited by insect groups (H. F. Nijhout, personal cornrnunication) but not by anurans. However, the relative lengths of the anuran larva17andaddt stages can be adjusted. In Gurans, the length of larval period is under genetic control, but genetic control of addt longevity is unknown. By i d e n e i n g larval and addt environments that d a e r in auaiitv. it shodd be DOSsible to test the hypothesis that le&h &f larval has been adjusted accordingly. For example, a consistentiy lush larval habitat surrounded by a harsh terrestrial environment wodd select for ~aedomorbhosisin saiamanders but codd select for a long?arval period and short addt iife span in frogs. Further work is needed in several areas. Mor~hometric and phylogenetic anaiyses are needed in a variety of species An aiternative to the quantitative genetic methodology is to assess phenotypic correlations of traits across metamorrooted in descriptive histology and developmental biology. phosis. The case of Leiopelma is of particular interest in that Using this approach, Aiberch (1987) identiied two devel- it is unclear if the earlysize-shape kajectory of the embryo opmental pathways that led to the formation of the larval resembles a tadpole (fig. 11.2D), an ancestral salamanderand postlarval epibranchial in the saiamander Eulycea. The like larva (fig. 11.2C), or neither (fig. 11.2B). Quantitative unique formation of the epibranchial cartilage during the genetic analises coupled with ecoGgical information are embryonic and larval periods is a primitive pattern. During invaiuable for testing hypotheses about current seleaion metamorphosis, larval chondrocytes can undergo ceii death on anurans. Such analyses aiso can test the hypothesis that or remain static. The compartmentalization of the precur- tad~oie addt traits are eeneticdv uncorrelated. Develand sors to larval chondrocytes (neural crest cells) into prospec- opmental biology wili become increasingly important in tive larval and addt ceiis is the derived pattern. Whde some addressing the general issue of uncoupling of iife-history neural crest ceiis differentiate into the larval epibranchial, stages (Alberch 1987). The combination of morphological, others remain quiescent. At metamorphosis, undifferenti- ge&ic, ecological, developmental, and phylo&netic apated cells develop into the postlarval epibranchial and the proaches surely wdi clarie our understanding of iife-history larval chondrocytes die. The derived compartmentalized de- evolution in anurans. velopment d o w s for independent evolution of larval and postlarval stages. Anurans also are expected to have compartmentalized de- ACKNOWLEDGMENTS velopmental programs (Aiberch 1987). The greater &ver- I thank R. Aiford, J. Hanken, S. Marks, and G. Wyngaard gence between stages in anurans perhaps is explained by a for helpful comments on the manuscript. James Madison greater number of compartmentalized developmemai path- University provided me with a research leave during which ways. A research program involving the marking of larval I developed the ideas presented in this chapter.
V

stage and the same trait in the addt stage. An analysis of this kind has not been performed, although genetic correlations between larval iife-history traits and postlarval morphological traits have been estimated. Emerson et al. (1988) showed that time to metamorphosis and growth rate of larvae were not correlated with hind iimb shape of metamorphs, and Blouin (1992b) demonstrated that growth rate of larvae was not correlated with head width. These resdts suggest that larval life-history traits and postlarval hind limb traits are able to evolve independently. The generality of these resdts require further study. Estimations of the genetic correlation between values of the same trait in larval and postlarval stages are avenues for future study. One such trait is body shape in the tadpole and in the metamorph. For example, if selection is imposed for a narrower tadpole, will there also be a response to that selection in the postiarval stage? Lack of an additive genetic correlation wodd suggest that independent evolution is possible. Estimation of genetic correlations between traits in larval and postlarval salamanders wodd be an interesting companion study. I hypothesize that correlationsin salamanders wili be large, whde correlations in frogs will not be significant. If true, this wodd help to explain the far greater phenotypic divergence between stages in anurans. Lack of genetic correlations between stages wodd also explain why paedomorphosis is cornmon in caudates and not in anurans. The abiiity of a species to reproduce in a larval morphology (paedomorphosis) is inversely related to the degree of phenotypic divergence between stages. A lack of developmental flexibility (e.g., loss of possibility of paedomorphosis) has adverse macroevolutionary consequences (higher extinction rates) in benthic marine invertebrates (Valentine and Jablonski 1983). Such a pattern is unsnidied in anurans.

ceiis in a variety of species and assaying their fates across metamorphosis wodd be of great interest.

Summary

DIVERSITY
Farnilial and Generic Characterizations
Ronald Altig and Roy W. McDiarmid

Introduction
Orton (1953) clearly recognized the impedirnents that commoniy are encounteied w k n attemptGg to i d e n t q anuran tadpoles. If it were oniy the inherent morphologicai traits characteristic of the larvae of different species that were holding back our understanding of tadpole diversity, we would have made considerable progress over the past 45 years. Unfortunately, the tadpoles of many species remain to be identified, and we have oniy meager understanding of the nature of ontogenetic, geographic, &d metamorphc~variation(e.g., Cei and Crespo 1982; Goiiman and Goiiman 1991, 1995, 1996; Inger 1966; Korky and Webb 1996; L a d a 1984b; J. M. Savage 1960). Our abiiity to recognize and identq new forms progressively improves with each description, but many factors, including a chronic lack of detail in many descnptions, inadequate collections, and too few workers, harnper progress. Intangible insights come from looking at manY tad~oles many s~ecies, through such exPosure of and one begins to develop an understanding of their morphologicai diversity. Having a readily available surnmary of the known morphological diversity of tadpoles aiso would be beneficiai. At times, interspecific variation seems to be slight or absent; whether such Gariation simply is not recogni&d (e.g., Bufo)or poorly understood (many taxa) remains to be shown. Phenotypic plasticity, especiay of pigmentation and fin shape, is confusing. Conspecific tadpoles collected from turbid versus clear water vary tremendously in color, and cultured tadpoles often have aberrant mouthparts compared to wiid-caught individuais of the same species. Tadpoles of the same siecies may dXer in shape in Still versus flowing water (R. D. Jennings and Scott 1993; Van Dijk 1966), and ontogenetic changes in color and oral morphology are occasionay profound. Improper fixation and storage (chap. 2) tend to complicate a other issues. Examination of the sma, complex oral apparatus of a tadpole provides the major characteristics for species identifications but often is fmtrating. The discovery of a simple method for properly preserving a tadpole with its oral disc wide open and without distortion would facilitate consistent observations and be a major advance. In s ~ i t e manv tzains. we sti find ourselves in the of realm of k e "prelimk& studyn mentioned by Orton (1953). Because research on tadpoles is increasing rapidly, the need for properly identified tadpoles is becoming more common. Researchers are reaiizing that sampling tadpoles usuay is an efficient, viable means of assessing local biodiversi+. Tadpoles are present in aquatic hab:itats for longer periods than breeding adults and are often more easily collected. H . B. ShafFer et ai. (1994) presented useful ideas in

Tadpoles are ". . . highly specialized, and seerningly unlimited variation shows iittle organization on preiiminary study."
Orton 1953:68

296

R O N A L D ALTIG A N D ROY W. M c D I A R M I D

this regard, but Wassersug (1997b) raised some interesting counterpoints. The importance of understanding ontogenetic variation and patterns of development has been aiscussed previously (chaps. 2 and 3), and-the need for colleaions of ontogenetic series is even more apparent. Developmental data provided an understanding of the differences between the duai lateral spiracles of pipids and rhinophrynids and those of Lepidobatvachus (Leptodactylidae; Ruibai and Thomas 1988; see Nieuwkoop and Sutasurya 1976). Even so, morphologicai data of tadpoles have seldom been gathered or examined to answer question about character homology. The use of larvai data in systematic studies has been debated, and tadpole features have been used to various degrees (e.g., Diaz and Valencia 1985; Donneliy et ai. 1990a; Dueiiman and Trueb 1983; I. GriKths 1963; I. GriKths and De Carvalho 1965; Inger 1967; Kluge 1988; Kluge and Farris 1969; Kluge and Strauss 1985; Noble 1926b; 0 . M. Sokol1975; Starrett 1973; Yang 1991). Disagreements relative to the use of tadpole characters in systematic analyzes seemingly resuit from incomplete understanding of charaaer states and their distributions arnong taxa (e.g., Wassersug 1989b). When appropriate traits have been investigated, issues relative to character homology, functional capacities, and both geographic and ontogenetic variations can ofien be confronted. We reviewed the morphologicai diversity of tadpoles in chapter 3, and Harris discussed the evolution of the tadpole as ;life-history stage in chapter 11. In this chapter we &marize the known larvai morphology of pertinent taxa, and discuss the limitations of current knowledge. Two data sets are needed before meaninfl advances in understandmg the interactions of tadpole ecomorphology and phylogeny can be made. We must have a more detailed phylogeny of the farnilies and genera of anuran amphibians and a better understanding 6f the origin and m~;~hology the primitive of tadpole (see chaps. 4 and 11). Once this has been achieved, we shouid be able to examine the distribution of larvai traits amonp: the severai clades of modern anurans and s~eculate " on their evolution. The moduiators of growth and metamorphosis and their combined effects on tadpole ecology (e.g., Wiibur and Collins 1973) need to be considered, as do minute differences in developmentai patterns, if we are to expand the current notions of character homology and better understand the evolution of larva1 traits. We must be able reiiablv to distineuish between those features that are more " labiie in response to ecologicai pressures (i.e., result in ecomorphologicai convergences among lineages) and those that hold reliable phylogenetic information. For example, morphology of the chondrocranium (chap. 4 and citations therein), position of the front iimbs relative to the buccopharyngeai structures (Starrett 1973), and patterns of formation and position of the spiracle(s) might be expeaed to carry strong phylogenetic signai, whiie certain features of the oral apparatus more likely reflect ecologically pertinent seleaion. In the same light, the distribution of some oral features among lineages-seems to indicate lineage-specific constraints. From our perspective the majority of tadpoles fall within

boundaries that typicdy coincide with the phylogenetic patterns derived from aduit frogs, but the extremes in ecologicai and morphologicai diversity among anuran larvae (figs. 12.1-12.2) seem truiy astounding. Some of this diversity is captured in the composite of microhabitats and morphoTable 12.1 Guide to select literature on tadpole faunas and regional treatments of groups. Referentes with keys are indicated with asterisks. Europe General: Boulenger 1892*, E. N. Arnold and Burton 1985*, Nollert and Nollert 1992*, Berninghausen 1997" Austria: Grillitsch et al. 1983*, Lower Austria Belgium: De Witte 1948* France: Ame1 1946* Germany: gerninghausen 1994*, Brauer 1961*, Engermann et al. 1993" Italy: Lanza 1983* Portugal: Serra and Aibuguerque 1963* Russia: Terent'ev and Chernov 1965" Spain: Salvador 1985+ Turkey: Basoglu and Ozeti 1973 Africa East f i c a : Schiotz 1975, hyperoliids South Africa: Van Dijk 1966*, Wager 1986 West Africa: Rode1 1996, Schiotz 1967, hyperoliids Cameroon: Mertens 1938 Madagascar: Blommers-Sdilosserand Blanc 1991, Glaw and Vences 1992* Natal: Lambiris 1988aX,b* Zimbabwe: Lambiris 1989* Asia Southeast Asia; Bourret 1942* Borneo: Inger 1966*, 1985* China: C.-C. Liu 1940, 1950 Japan: Okada 1931, Maeda and Matsui 1989 Java: Schijfsma 1932 Malaysia: Berry 1972 Pakistan: Khan 1982aX Philippines: Inger 1954 Russia: Terent'ev and Chemov 1965* Taiwan: Chou and L i 1997" Thailand: M. A. Smith 1916a, 1917 Australia/New Guinea Australia: Watson and Martin 1973, Tyler 1989*; Queensland: Retallick and Hero 1988*, rainforest streams in one area North America Canada: Orton 1952*, genera, Aitig 1970*; British Columbia: Corkran and Thoms 1996"; Aiberta: A. P. Russell and Bauer 1993" United States: Orton 1952*, genera, Aitig 1970", Wright 1929*; Eastern United States: A. H. Wright and Wright 1949*; Florida: Fanning 1966*, Stevenson 1976*; Louisiana: Siekmann 1949*; North Carolina: Travis 1981ar; Oregon: Corkran and Thoms 1996"; Washington: Corkran and Thoms 1996* Middle America General: Dueilman 1970, hylids Costa Rica: J. M. Savage 1968*, dendrobatids; 1980ar, Lips and Savage 1996" Honduras: L. D. Wiison and McCranie 1993* Mexico: Aitig and Brandon 1971", genera, Aitig 1987" Panama: J. M. Savage 1968*, dendrobatids South America Argentina: Cei 1980, 1987; Kehr and Wfiarns 1990 Brazil: Hero 1990*, Manaus; Heyer et al. 1990, Boracia Chile: Cei 1962 Ecuador: Dueilman 1978*, Santa Cecilia/pper Arnazon Basin

DIVERSITY

297

types depicted in figures 9.2-9.3. The charaaerization of t-adpolesb y taxa pre&entedin ths chapter pinpoints some of the a c u l t i e s associated with species identifications, iilustrates gaps in the data, provides a point of departure, and lists ~ertinent references to im~rove likelihood of I I the DroDer identiication. In our experience, larval morphology is often concordant with accepted systematic treatments (e.g., Bufo vevaguensis, Hylu leucophyllata, and H. pam'ceps groups) and in other instances likely signals the need for adjustment (e.g., species in the Scinm vostrata group and Scarchyla). The literature that has proven usei for identiication of tadpoles generaliy is scattered. In this chapter we review the sources of major keys and other large works that include considerable tadpole dormation (table 12.1). As we have noted previously, makmg correa identifications of tadpoles often is a c u l t . In table 12.2, we list citations that correctiy ident* previously misidentiied tadpoles and in table 12.3 we provide identifications and references to tadpoles that were described orinindv without a name. Mouth~arts of tadpoles are admit;dly Ldd l o o h g structures without an obvious orientation, and they are sometimes printed upside down; known examples include Pseudis pavahxa (Dixon et al. 1995);Hylazeteki (Duehan 1970);Buf bufo, Kaloula m&2eva, and Rana japonica (C.-C. Liu 1940); and Mimobatrmhella sp. (Wager 1986). The tadpole described as Centrolem bucklevi bv Rada de Martinez (1990) either is misidentified or deviates strongly (e.g., complete marginal papiilae and many more than 213 tooth rows) from the conservative centrolenid morphotype; her description of Scinm rostrata also seems in error. The primitive tadpole presurnably was some sort of benthic inhabitant of a pond-type environment, perhaps ephemeral in nature. It likely fed by rasping food materials from submerged surfaces with keratinized mouthparts. Bogart (1981)and others have presented alternative ideas (see chap. 11). As previously noted, suggestions regarding the evolution of specific characteristics among anuran larvae are rare and usudy weakly defined. Given the state of knowledge, the concepts of ecomorphological g d d s (Altig and Johnston 1989) or similar designations of ecologicdy hnctional larval groups (Van Dijk 1972) at the moment appear to be the best summaries of morphological diversity among lineages.
2 ,

Table 12.2 Identiications of previously misidentified tadpoles. Names are arranged alphabeticdy by initial genus and species. Original Identification Citation Corrected Identification Citation

Bufo campons Korky and Webb 1973 Chinjcalw vittatus Pope 1931 Chimmantisnrfeswns Guib and Lamotte 1958 Colostethushaydeeae Rivero 1976 Colostethw subpunctatus Cei and Roig 1961 Gephyromuntis methueni Razariheiisoa 1973b HetenXdus amoulti Razarihelisoa 1979 H-vhloyuaxa Duellman 1970 Hyh siopeh Caldweii 1974 Hyh sp. Stuart 1948 Hyperoliuspicturatus Ai111931 Kivsina sengaiensis Lamotte and Zuber-Vogeli 1956 Leptobrachium hmelti Wassersug 1980 Leptobrachiumgracili~raciliP Inger 1966 Mantuictylus alutus Arnoult and Razarihelisoa 1967 Manttdactylus brevipalmatus Razarihelisoa 1973a Mantidaqlus ulce-rosus Razariheiisoa 1969 Nyctimystes montana H. W. Parker 1936 ?-Tvmmmystes duyi Czechura et al. 1987 Microhyh bmeensis H. W Parker 1934 . Otophlyne robusta Wassersug and Pyburn 1987 Paratelmatobiw lutzi Heyer 1976b

Bufo mucromistatus Mendelson 1997 ChinXal1*rdonae Heyer 1971 Phlyctr'mantU ieonardi Schietz 1975 HyhpWdaqh La Marca 1985 Colostethussemiguttata J. Faivovich, personal communication Manttdactylus bicaisuratus Blommers-Schiosser and Blanc 1991 HeteruCdus betsileo Blommers-Schlosser 1982 Unidentified Rana R. Altig, personal observation Hyh whta Toal and Mendelson 1995 Piemhvh teuchestes ~ u e h a and Campbeii n 1992 Rana albolab* Schietz 1967 Kivsina macuhta Schietz 1967 Leptobrachium hendhcbsonii Inger 1983 Leptobrachella mjobergi Inger 1983 Heterklw betsibo Blommers-Schiosser 1979a Mantdmtylw aenrmnalir Blommers-Schiosser and Blanc 1991 Manttdactylw curtus Blommers-Schiosser 1979a Nym'mystes cheesmanae Davies and Richards 1990 Litoria nyakaiensis S. J. Richard~ 1992 Kalophlynus sp. Inger 1966 0tophynepyburn.i Campbeii and Clarke 1998 Unknown W R. Heyer, personal . communication Nci&cupgaiimes ytbabsyrclPau Piai 1978 Probably Ranapipiens R Altig, personal observation Rana@eri R. Altig, personai obse~ation Ptychadena anchietae Van Dijk 1966 Unknown Inger 1966 Unknown Inger 1966
(continued)

Familial and Generic Characterizations

Understanding the diversity of tadpole morphology (figs. Phiuutus vanadilU 12.1-12.2; also see chaps. 2 and 3) is a requisite of successfl Annandale 1919 identiiications. Appreciating how those morphologies are Pseudami brachyphona Green 1938 distributed across taxa and which tad~oles have been described in each group also is invaluable. Because of gaps in Ptmhylafodiens D u e h a n 1970 our knowledge and the general conservativeness of tadpoles, Ptychadena oxyrhyachw are some tad~oles more difficult to characterize. and therePower 1927b fore idem+, than others. These obstacles may be at the ge- Rana hosii Van Kampen 1923 neric (e.g., ali centrolenids, most dendrobatids, Hyla vs. Osteocephalus, Scaphwpus vs. Spea) or specific levels of organi- Rana [= Limnonectes] macrodon zation (e.g., species of Bufo and the Ranapipiens complex, Flower 1899 HL i 1982; Scott and Jennings 1985; Wassersug 1976~). l Ls

298 Table 12.2 wntinued

RONALD ALTIG A N D ROY W. M c D I A R M I D

Rana [= Limnanectes]w o d a n Inger 1956 Rana microdiscapalavanensis Inger 1956 Rana nicobariensis M. A. Smith 1930 Ranapleske6 Annandale 1917 Rena vertebral& Hewitt 1927 Rhacophorzls leucomystm Inger 1956 Staurois natatm Mocquard 1890 Staurois latopalmatus Inger 1966 Rhqhoncs bimacuiuta Inger 1985

Limnanectes microdisca Inger 1966 Limrumectesparamurodon Inger 1966 Unknown Inger 1966 Perhaps Scutiger naumnauta Boulenger 1919 Rana umbraculata Van Dijk 1966 Polypedutes m m ~ Inger 1966 Unknown Inger 1966 R h q h w w bimaculatus Inger 1985 Rhqhwusgauni Inger and Tan 1990

(e.g., Lavilla 1985, 1988) or anatomical (e.g., Viertel 1982) data are uncommon, and ecomorphological gurlds (Altig and Johnston 1989) do not reflect phylogenetic relationIn ths chapter we provide the first summary of familiai and generic characterizations of tadpoles based on a common terminology. The taxonomic arrangement of D. R. Frost (1985), updated through 1996, was foiiowed and includes most of the ranoid arrangement proposed by Dubois (1987,1992; see comments by Inger 1996). Representative, but usuaiiy not exhaustive, citatiois concerning each genus are provided. Subfamiiies are included primariiy to separate groups of genera. The Gosner system (1960) of staging was used throughout. Afier each f d y , the number of genera and species, geographic range, major ecomorphologicaidivisions of Aitig and Johnston (1989), and other characteristics are given; for mono~eneric farnilies or subfamiiies. the com~arable o r d " mation is presented only under the higher category. Character states common to aii genera within a famiiy or subfamiiy are not repeated at the generic level. The state of a character for whichdata are lack&p:is indicated with a dash. Mor~ho" logical traits are given only for exotrophic (i.e., feeding; see chap. 7 for endotrophic forms) forms; genera, aii of whose species are entirely endotrophic, are included oniy to provide a complete listing (see aiso chap. 7). Character abbreviations used in the compiiation are'CO, COmposition (famiiy/subfdy: number of genera/number of species; genus: number of species); GR, Geographic Range; EMG, EcoMorphological Guild; LTRP, Labial Tooth Row Formula (see chap. 3; ontogenetic and other variations not included); OA, Oral Apparatus (typicai oral disc with labial teeth and jaw sheaths regardless of conigurations: position descriptor); MP, Marginal Papiilae (distribution: arrangement); SUP, SUbmarginal Papiilae; DEM, Disc Emargination; NA, Nares; VT, Vent Tube; EP, Eye Position; SP, Spiracle; UJ, Upper Jaw sheath (width notation, descriptor of shape of serrated edge); LJ, Lower Jaw sheath (width notation, descriptor of shape of serrated edge); DP, Dorsai Fin (general height descriptor, anterior site of origin, tip descriptor); BS, Body Shape (dorsai view/ lateral view); CP, Color and Pattern; TL, Total Length/ Stage; NO, Notes; and CI, Citation(s). Generalized drawingi of representative morphotypes appear immediately before the pertinent taxon, and data on the specimens used in the preparation of these drawings foiiow.

aSeeJ. C. Lee 1996 for correct description. bToothiess. 'See C.-C. Liu 1950. Table 12.3 Identiication of tadpoles originaliy described without a name. Names and citation are presented in alphabetical order by initial genus and species. Original Identification Citation
Amo@ sp. A Inger 1966 Amo@ sp. B Inger 1966 Rhqhorus sp. B Inger 1966 Unknown Pyburn 1980 Unknown hylid Stuart 1948 Unknown E. H. Taylor 1942 Unknown Inger and Wassersug 1990

Corrected Identification Citation

Merktgenysphaeomms Inger and Gritis 1983 Me&o~enyspoecilus Inger and Gritis 1983 Rana luctuosa Inger 1985 Otophynepybumi Wassersug and Pyburn 1987 Rana maculata Hiis and De S 1988 Rhinophrynw h a l & Orton 1943 Staurolc natatm Inger and Tan 1990

The apparent lack of daerentiating characteristics among species in some genera or inadequate study of certain groups sometimes forces workers to key tadpoles oniy to the species rank. In ths regard considerable work is needed on enigmatic taxa such as Buf, aii centrolenids and dendrobatids, Leptod&us, some ranids, and Telmatobius. Attempts to diagnose larvae above the species level with morphologicai

Fig. 12.1. Representative body shapes and fin configurations among anuran tadpoles. (A) Ranupalmipes (Ranidae, dasping). (B) Mgophys m t a n u (Megophryidae, neustonic). ( C )S c i m mbulosa (Hylidae, nektonic). ( D )Hyh vimduris (Hyiidae, suctorial). (E) Atclopw &nesm (Bufonidae, gaswmyzophorous). (F) Coiustcth nubicoh (Dendrobatidae, neustonic). ( G )Thmopapmopolitana (Leptodactylidae,semitemesmal). (H) LeptopclU byluih (Hyperoiiidae, benthic). (I) Hyh b?vmeliaiu (Hyiidae, arboreal Type 1. (J) Gashvphynecuwli& (Microhyiidae, suspension feeder). (K) Hyh nrinocephala (Hyiidae, macrophagous). ( L )Oudmy~aLISrna (Ranidae, macrophagous). (M) Hyh l i & (Hylidae, suctorial). (N) DucUinamclyh ummchvoa (Hylidae, adherent). (0) CochvamUap$thsi (Centrolenidae, fossoriai) ( P ) Anothea pinosa (Hylidae, arboreai Type 2). (Q) Cmaqhrys m u t a (Leptodactyiidae, camivorous Type 1). (R) Sttphop&amclr (Bufonidae, arboreai Type 3). From Altig and Johnston (1989), after Dueiiman and Tmeb (1986); original drawings by and L i d a Tmeb. Reprinted by permission of HevpetoIo~gicalMm~raphs Tmeb.

Fig. 12.2. Representative configurations of the oral apparatus among anuran tadpoles. (A) Ranapulr* n (Ranidae, dasping), (B)Megopby muntuna (Megophryidae, neustonic), (C) Hl riarlarLr (Hylya ya idae, suctorial), (D)Atehpa & n e s m (Bufonidae, gastromyzophorous), (E) Hl micmcephaia (Hylidae, macrophagous), (F) Cem&phtys m u t a (Leptodactylidae, camivorous Type l), (G) Hyla l i h (Hylidae, suctorial), and (H) DueUmumbh uranucbroa (Hylidae, adherent). From Altig and Johnston (1989) after D u e b a n and Trueb (1986); original drawings by L i d a Trueb. Reprinted by permission of H"pctob@ulMmpphs and Trueb.

DIVERSITY

301

ALLOPHRYNIDAE: CO: 111. GR: Guyana-Brazil. Tadpole unknown. Allophvyne: CO: 1.GR: Guyana-Brazil. Tadpole unknown. ARTHROLEPTIDAE: CO: 6/77. GR: sub-Saharan Africa. EMG: exotroph: lotic: adherent, neutonic, suaorial; lentic: benthic; endotroph. LTRF: several variations between 010 and 9/11. OA: typical: anteroventral, terminal, umbeiiiform. MP: absent, medium to wide dorsal gap, complete: uni- and muitiserial. SUP: absent, single row distal on lower labium, oblong papdae arranged radidy to mouth. DEM: absent, lateral. NA: nearer snout than eye. VT: medial. EP: dorsal, lateral. SP: sinistral. UJ: wide, smooth arc to V-shaped, sometimes serrations coarse and medial ones may be enlarged. LJ: similar to upper sheath. DF: low, originates on taii muscle posterior to the dorsai tail-bodyjunction, tip round to pointed. BS: oval to elongate/depressed. CP: uniformly dark to mottied. TL: 29-80136. Arthroleptinae: CO: 2/52. GR: sub-Saharan Africa. Arthroleptis: CO: 35. GR: sub-Saharan Africa. EMG: endotroph. Cardwglossa: CO: 17. GR: westem Africa. EMG: exotroph: lentic: benthic. LTRF: 010. OA: typicai: anteroventral. MP: wide dorsal gap: uniserial, long and widely spaced ventrdy. SUP: single row distal on lower labium. DEM: absent. NA: s m d , nearer snout than eye. VT: mediai. EP: dorsal. SP: sinistral. UJ: wide, smooth arc, coarsely serrate. LJ: wide, coarsely serrate. DF: low, tip rounded. BS: oblong/depressed. CP: unifonnly dark. TL: 29/36. NO: ventral in originates on tad muscle. CI: Lamotte (1961). Astylosterninae: CO: 4/25. GR: western Africa. Astylostemus: CO: 12. GR: Sierra Leone-Zaire. EMG: exotroph: lotic: adherent. LTRF: 011, 3(2-3)/3(1-2). OA: typical: aimost terminal, anteroventral. MP: medium to wide dorsal gap: uniserial. SUP: row distai to distal most lower row. DEM: lateral, absent. NA: nearer snout than eye, s m d medial tubercle at margin. VT: dextral. EP: lateral. SP: sinistral. UJ: narrow to wide, smooth arc with single large, medial serration. LJ: exceptiondy wide, V-shaped, coarsely serrate. DF: low, tip pointed. BS: oblonddepressed. CP: uniformiy dark. TL: 80137. NO: ventral in originates posterior to origin of dorsal fin; body with lateral sacs of variou degrees of development, function unknown (Arniet 1970, 1971). CI: Amiet (1971); Lamotte and Zuber-Vogeli (1954b). Leptoahctylodon: CO: 11. GR: Cameroon-Nigeria. EMG: exotroph: lotic: neustonic. LTRF: 010, but with ridges that suggest a 011-2 arrangement. OA: typical: umbeiiiform. MP: absent, margin crudely crenuiate. SUP: oblong papiiiae arranged radidy to mouth and scattered throughout disc. DEM: absent. NA: smd, about equidistant between snout and eye. VT: dextral. EP: lateral. SP: sinistrai. UJ: massive, d serrations large, especidy laterdy. LJ: massive, similar to upper sheath. DF: tip rounded. BS: ovai/depressed. CP: uniformly dark, mottied on tail. TL: 50136. NO: a tooth ridge-like ridge occws immediately distai to the lower jaw sheath; body with lateral sacs of various degrees of development, function unknown (Amiet 1970, 1971). CI: Amiet (1970). Scotobefs: CO: 1. GR: Nigeria-Gabon. Tadpole unknown; see Lawson (1993). Trichobatrachus: CO: 1. GR: Nigeria-Equatorial Guinea. EMG: exotroph: lotic: suctorial. LTRF: 9(7-9)/11(1-2). OA: typical: ventral. MP: complete: rnultiserial. SUP: absent. DEM: lateral. NA: equidistant between eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, strongly V-shaped. LJ: medium, V-shaped. DF: tip rounded. BS: oblongldepressed. CP: uniformly dark. TL: 91/36. NO: large subbranchial muscle along anterior two-thirds of body gives ventral sdhouette a distinctive shape; body with lateral sacs of variou degrees of development, h c t i o n unknown (Amiet 1970, 1971). CI: Mertens (1938). ASCAPHIDAE: CO: 111. GR: northwestern United States-adjacent Canada. EMG: exotroph: lotic: suaorial. LTRF: 2-319-ll(1). OA: typical: suctorial. MP: wide dorsal gap, but anterior lip fleshy and crenuiate: uniserial. SUP: muitiserial distal to distaimost lower row. DEM: absent but folds present. NA: nearer eye than snout, prominent tubuiar extensions. VT: medial. EP: dorsal. SP: midventral on chest, no dorsal w d . UJ: large, platelike lying aimost pardel with substrate, sometimes with medial break. LJ: minute, U-shaped. DF: low, originates at dorsal taii-body junction, tip broadly rounded. BS: ovalJdepressed. CP: rockiike mottling, tad tip typicdy white bordered by black. TL: 60136. NO: prominent s m d skm glands present ventrdy; LTRF A-1 and most distal posterior row rnultiserial.

Trichobatrachus

Ascaphus

RONALD ALTIG AND ROY W. McDIARMID

RFcaphus: CO: 1. GR: northwestern United States-adjacent Canada. CI: Altig (1970); H. A. Brown (1990a); Metter (1964, 1967); Stebbins (1985). BOMBINATORIDAE: CO: 218. GR: Eurasia-Philippines-Borneo Barboumh: CO: 2. GR: Philippines-Borneo. EMG: Tadpole unknown, perhaps endotrophic. Bombina: CO: 6. GR: Eurasia. EMG: exotroph: lentic: benthic. LTRP: 2/3[1], biserial. OA: typical: anteroventral. MP: complete: uniserial. SUP: row ventrolateraiiy. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: low on left side. UJ: medium, straight area mediaiiy, coarsely serrate. LJ: narrow, V-shaped, coarsely serrate. DE: low, originates near dorsal taii-body junction, tip rounded. BS: oval/depressed. CP: melanophores form network pattern in skin. TL: 30-38136 CI: Boulenger (1892); Lanza (1983). BRACHYCEPHALIDAE: CO: 214. GR: Brazil. EMG: endotroph. Br1u:h~ephalus: CO: 2. GR: Brazil. EMG: endotroph. Psyllophlyne: CO: 2. GR: Brazil. EMG: endotroph. BUFONIDAE: CO: 331381. GR: widespread except Australia. EMG: exotroph: lentic: benthic; lotic: benthic, suaorial, gastromyzophorous; endotroph. LTRF: 2/3,2(2)/3, 2(2)/ 3(1), 2(2)/2. OA: typical: anteroventral, rarely ventral and suaorial. MP: wide dorsal and ventral gaps, complete: uni- and biserial, wide dorsal gap only. DEM: lateral, absent. NA: often large, often not round and with mediai papilla, sometimes smaii and rounded, doser to eye than snout or reverse. VT: mediai, dextral. EP: dorsal. SP: sinistral. UJ: usuaiiy narrow and broadly arched. LJ: usuaiiy narrow and broadly V-shaped. DE: low, originates near or posterior to dorsal td-body junction, tip round to slightly pointed. BS: oval to roundldepressed. CP: typicaiiy uniformiy black; t muscle sometimes bicolored or banded. TL: d 1540136. Ahiphlynoides: CO: 1. GR: Ethiopia. EMG: endotroph. Andinophlyne: CO: 3. GR: Ecuador-Colombia. EMG: endotroph. Ansonia: CO: 19. GR: India-Borneo-Philippines. EMG: exotroph: lotic: clasping, suctorial, gastromyzophorous. LTRP: 213. OA: typical: ventral. MP: smaii, with medium to wide dorsal gap, complete: uniserial. SUP: row(s) distal to P-3, few lateraiiy. DEM: lateral, absent. NA: smaii, nearer eyes than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: narrow to medium, divided, absent. LJ:narrow, very open V-shaped. DE: low, originates on t d muscle or near dorsal taii-body junction, tip bluntiy pointed. BS: ovalldepressed. CP: uniformiy dark, with rniddorsal stripe. TL: 18-35/36. NO: large variations in oral morphology within genus; anterolaterai terrninus of abdominal sucker of A. "sucker" of Inger (1992a) has fingerlike projection. CI: Inger (1966, 1985, 1992a). Atelophlyniscus: CO: 1. GR: Honduras. EMG: exotroph: lotic: gastromyzophorous. LTRP: 213. OA: typid: ventral. MP: wide dorsal and smaiier ventral gaps: uniserial ventraiiy, biserial dorsally. SUP: few anterolateraiiy. DEM: absent. NA: s m d , nearer eye than snout. VT: medial, not attached to ventral fin. EP: dorsai. SP: sinistral. UJ: narrow, without serrations, V-shaped. LJ: narrow, V-shaped. DF: low, originates slightly posterior to dorsal taii-body junction, tip rounded. BS: oval to round/depressed. CP: black t d muscle boldly banded; other white marks on body. TL: 26/37. NO: ventral fin originates on t d muscle; resembles tadpoles ofAtelupus. CI: McCranie et al. (1989). Atelopus: CO: 50. GR: Costa Rica-Bolivia. EMG: exotroph: lotic: gastromyzophorous. LTRF: 213. OA: typid: ventral. MP: wide ventral gap: uniserial. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: narrow, smooth arc. LJ: narrow, open U-shaped. DE: low, originates near or slightly distal to dorsal td-body junction, tip rounded. BS: oval to round/depressed. CP: uniformiy black, may have white blotches or bands on t d . TL: 15-20136. CI: Duellman and Lynch (1969); Gascon (1989a); Lidquist and Hetherington (1988); J. D. Lynch (1986); Starrett (1967). Bufi: CO: 217. GR: widespread. EMG: exotroph: lentic: benthic; lotic: benthic, suaorial, gastromyzophorous; endotroph. LTRF: 2(2)/3(1), 2(2)/3,2/3,2(2)/2(1). Ok. typical: anteroventral. MP: wide dorsal and ventral gaps, complete: uniserial. SUP: few lateraiiy, absent. DEM: lateral, absent. NA: large and usuaiiy not round, often with me-

Bombina

Ansonia

DIVERSITY

diai papdla. VT: mediai, dextrai. EP: dorsal. SP: sinistral. UJ: narrow, smooth arc, angled lateraiiy. LJ: narrow, open V-shaped. DE: low, originates at dorsai td-body junction, tip rounded. BS: ovai-round/depressed. CP: typicaiiy uniformly black, tail muscle sometimes bicolored or banded. TL: 15-35/36. NO: smaii size is cornrnon; morphologicaliy conservative and species characteristics poorly docurnented; no consistently recognized morphologicai dserences among numerous species groups except in debilis group; Crump (1989b) reported facultative endotrophy in B.pm$enes. CI: Altig (1970); Berry (1972); McDiarmid and Altig (1990); Zweifel (1970). Bufoides: CO: 1. GR: hdia. Tadpole unknown; see Piai and Yazdani (1973). Capensibufo: CO: 2. GR: southwestern South Africa. EMG: exotroph: lentic: benthic. LTRP: 2(2)/3. OA: typicai: anteroventral. MP: wide dorsai and ventral gaps: uniserial. SUP: absent. DEM: lateral. NA: large, nearer eye than snout. VT: mediai. EP: dorsai. SP: sinistrai. UJ: medium, broad mediai convexity. LJ: medium, open U-shaped. DE: low, originates at dorsai taii-body junction, tip round. BS: ovai/depressed. CP: uniformly dark. TL: 35/36. CI: Power and Rose (1929); Wager (1986). Crepzdophryne: CO: 1. GR: Costa Rica. Unknown developmentai trajectory. Dendruphryniscus: CO: 3. GR: Ecuador-Pem-Guianas-Brazil. EMG: exotroph: lentic: benthic, arboreal. LTRF: 213, 2(2)/3. OA: typicai: anteroventral. MP: wide dorsai and ventral gaps: uniserial. SUP: absent. DEM: laterai. NA: large, nearer eye than snout. VT: medial. EP: dorsai. SP: sinistrai. UJ: medium, angled at dorsolaterai points. LJ: medium, open V-shaped. DF: low, originates at dorsai td-body junction, tip rounded. BS: ovai/depressed. CP: untformly dark. TL: 10-13136. CI: Izecksohn and Da Cmz (1972). Didynamipus: CO: 1. GR: Cameroon. EMG: endotroph. Frostt'us: CO: 1. GR: Brazil. EMG: endotroph. Laurentuphyne: CO: 1. GR: Zaire. EMG: endotroph. Leptupbryne: CO: 2. GR: Maiaysia. EMG: exotroph: lentic: benthic. LTRP: 2(1-2)/3. OA: typical: anteroventrai. MP: wide dorsai gap: uniserial. SUP: absent. DEM: lateral. NA: large like Bufo and equidistant between eye and snout. VT: mediai. EP: dorsal. SP: sinistrai. UJ: narrow, smooth arc. LJ: narrow, open U-shaped. DF: low, originates at dorsai tail-body junction, tip rounded. BS: oval/depressed. CP: uniformly black. TL: 151 34. NO: lack of gap in marginai papiae of lower labium is unusual. CI: Berry (1972); Inger (1985). Melanupbyniscus: CO: 11. GR: Brazil-Paraguay-Argentina. EMG: exotroph: lentic: benthic. LTRF: 2/3[1]. OA: typicai: anteroventral. MP: wide dorsai and ventrai gaps: uniserial. SUP: absent, few lateraiiy. DEM: laterai, absent. NA: nearer eye than snout. VT: mediai. EP: dorsal. SP: sinistrai. UJ: mediurn, smooth arc, coarsely serrate. LJ: medium, open U-shaped. DF: low, originates near dorsai td-body junction, tip round. BS: ovai/depressed. CP: umformly dark, tad muscle bicolored, fins speckled. TL: 19/36. NO: figure in Prigioni and Arrieta (1992) shows laterai gaps in marginal papiae. CI: Prigioni and Langone (1990); Starrett (1967). Mertensuphryne: CO: 2. GR: Zaire-Tanzania. EMG: exotroph: arboreal. LTRF: 112. OA: typicai: anteroventrai. MP: wide dorsai gap: uniserial. SUP: absent. DEM: laterai. NA: nearer snout than eye, withm the fleshy, hoiiow "crown" that encircles the eyes and nares. VT: mediai. EP: dorsai. SP: sinistral. UJ: wide, almost straight edge, coarsely serrate. LJ: similar to upper sheath. DE: low, originates near dorsal td-body junaion, tip broadly rounded. BS: oblong/depressed. CP: unrformly and lightly pigmented. TL: 13/30. NO: see Stephupaedes. CCI: Grandison (1980). Metaphryniscus: CO: 1. GR: Venezuela. EMG: endotroph. Nectophryne: CO: 2. GR: Cameroon-Zaire. EMG: endotroph. Nectuphrynotdes: CO: 5. GR: Tanzania. EMG: endotroph. Nimbaphymides: CO: 2 . GR: Liberia-Ivory Coast. EMG: endotroph. Oreuphrynella: CO: 6. GR: Venezuela-Guyana-Brazil. EMG: endotroph. CI: McDiarmid and Gorzula (1989). Osmnophryne: CO: 5. GR: Colombia-Ecuador. EMG: endotroph. Pedostipes: CO: 6. GR: hdia-Borneo. EMG: exotroph: lentic: benthic. LTRF: 2(2)/ 3(1). OA: typicai: ventrai. MP: wide dorsai and ventrai gaps: blunt and uniserial. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: mediai. EP: dorsal. SP: sinistral.

pjOPhtyne

RONALD ALTIG A N D ROY W. McDIARMID

Schismudenna

Stephopaees

wmz ena

UJ: medium, smooth arc, coarsely serrate. LJ: medium, open V-shaped. DF: low, originates near dorsai td-body junction, tip round. BS: ovai/depressed. CP: uniformly dark. TL: 16/36. CI: Inger (1966, 1985). Pelophvyne: CO: 7. GR: Philippines-China. EMG: endotroph. PeltcIphvyne: CO: 9. GR: Greater Anties. EMG: exotroph: lentic: benthic. LTRF: 2(2)/ 3, 212. OA: typicai: anteroventral. MP: wide dorsai and ventral gaps: uniserial. SUP: row lateral to lower tooth rows. DEM: lateral. NA: nearer eye than snout. VT: mediai. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc with slight mediai indentation. LJ: medium, open U-shaped. DE: low, originates near dorsal td-body junction, tip rounded. BS: ovai/depressed. CP: uniformly dark. TL: 2 1/36. CI: Ruiz Garcia (1980). Pseuubufo: CO: 1. GR: Borneo. Tadpole unknown. Rhamphqhvyne: CO: 9. GR: Panama-Peru. EMG: endotroph. Schismaema: CO: 1. GR: Tanzania-Zaire-South Africa. EMG: exotroph: lentic: nektonic. LTRP: 213. OA: typicai: anteroventral. MP: wide dorsai and ventral gaps: s m d and uniseriai. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: medial. EP: dorsai. SP: sinistrai. UJ: medium, long low convexity in margin; Wager (1986): upper sheath lacks serrations. LJ: narrow, open U-shaped. DE: low, originates near dorsai t d body junction, tip rounded. BS: ovai/depressed. CP: uniformly dark. TL: 35/36. NO: unique semicircular flap across body b e h d eyes; commonly schools in midwater/surface at least during the day. CI: Wager (1986). Spinophvynoihs: CO: 1. GR: Ethiopia. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3. OA: typical: anteroventral. MP: wide dorsai and ventral gaps: uniseriai. SUP: -. DEM: -. NA: -. VT: -. EP: -. SP: -. UJ: -. LJ: -. DE: -. BS: -. CP: -. TL: -. CI: Grandison (1978); M. H. Wake (1980). Stephqaehs: CO: 2. GR: Tanzania-Mozambique. EMG: exotroph: arboreal. LTRF: 2(2)/2. OA: typicai: anteroventral. MP: wide dorsai gap: uniserial, some appear branched and s m d gaps between papdiae. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: mediai. EP: dorsai. SP: sinistrai. UJ: wide, edge nearly straight. LJ: absent or not keratinized. DE: low, originates near dorsai td-body junction, siight arc about midiength, tip broadiy rounded. BS: ovai/depressed. CP: uniformly paie. TL: 201 37. NO: fleshy "crown" encircles nares and eyes; see MertenscIphvym. CI: Channing (1978, 1993). Tmebella: CO: 2. GR: Peru. EMG: endotroph. CI: Graybeal and Cannatella (1995). Werneria: CO: 4. GR: Togo-Cameroon. EMG: exotroph: lotic: suaorial. LTRP: 213. OA: typicai: ventral, suaoriai. MP: wide dorsai gap: biseriai. SUP: absent. DEM: lateral. NA: s m d , nearer snout than eyes. VT: medial. EP: dorsai. SP: sinistrai. UJ: narrow. LJ: narrow, open V-shaped. DE: low, originates at dorsai td-body junction. BS: oval/ depressed. CP: uniformly dark. TL: 27/36. CI: Mertens (1938); Perret (1966). Wolterstorffina:CO: 2. GR: Cameroon-Nigeria. EMG: endotroph. CENTROLENIDAE: CO: 31103. GR: Mexico-Brazil-Boiivia. NO: generic and specific drfferences poorly known. Centrolene: CO: 33. GR: Honduras-Peni-Venezuela. EMG: exotroph: lotic: burrower. LTRF: 5 213. OA: typicai: anteroventral. MP: wide dorsai gap: uniierial. SUP: absent. DEM: lateral. NA: s m d , about equidistant between eye and snout. VT: medial, dextrai. EP: dorsal: tiny and near sagittal line. SP: sinistrai. UJ: narrow to medium, smooth arc, sometimes coarsely serrate. LJ: narrow to medium, U- or V-shaped. DE: low, originates on t d muscle to near td-body junction, tip rounded. BS: oval/strongly depressed. CP: reddish in iife because of skin vascularization, uniformly tan in preservative. TL: 40-501 36. CI: J. D. Lynch et ai. (1983); Mijares-Urrutia (1990). Cochranella: CO: 44. GR: Nicaragua-Bolivia-Brazil. EMG: exotroph: lotic: burrower. LTRF: 5 213. OA: typicai: anteroventral. MP: wide dorsai gap: uniserial. SUP: absent. DEM: lateral. NA: s m d , about equidistant between eye and snout. VT: medial, dextrai. EP: dorsai: tiny and near sagittai luie. SP: sinistral. UJ: narrow to medium, smooth arc, sometimes coarsely serrate. LJ: narrow to medium, U- or V-shaped. DE: low, originates on t d muscle to near td-body junction, tip rounded. BS: ovai/strongly depressed. CP: reddish in iife because of skin vascularization, uniformly tan in preservative. TL: 40-501 36. CI: Duellman (1978).

Cochranella

DIVERSITY

Hyalimbatrachzztm: CO: 26. GR: Mexico-Bolivia-Tobago-Guyana-Brazil. EMG: exotroph: lotic: burrower. LTRF: 5 213. OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: absent. DEM: lateral. NA: small, about equidistant between eye and snout. VT: medial, dextral. EP: dorsal: tiny and near sagittal line. SP: sinistral. UJ: narrow to medium, smooth arc, sometimes coarsely serrate. LJ: narrow to medium, U- or V-shaped. DF: low, originates on taii muscle to near taii-body junction, tip rounded. BS: oval/strongly depressed. CP: reddish in Me because of skin vascularization,uniformiy tan in preservative. TL: 40-50136. CI: Duellman (1978); Heyer (1985).
DENDROBATIDAE: CO: 81164. GR: Nicaragua-Bolivia-Brazil. EMG: exotroph: lentic: benthic, arboreal; lotic: benthic, neustonic; endotroph. LTRF: number of variations from O/ O to 213. OA: typical: anteroventral, terminal, umbelliform. MP: medium to wide dorsal gap: uni- and biserial. SUP: few lateral, absent. DEM: lateral, absent. NA: nearer eye than snout or reverse. EP: dorsal. SP: sinistral. UJ: narrow to wide, smooth arc. LJ: narrow to wide, Uor V-shaped. DF: low, originates near dorsal td-body junction, tip rounded to slightly pointed. BS: oval to rounded/depressed. CP: uniform or muted variations. TL: 12-68/36. NO: generic and specific differences poorly understood. Aromobates: CO: 1. GR: Venezuela. EMG: exotroph: lotic: benthic. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial on upper labium, biserial on lower labium. SUP: few ventrolaterally. DEM: lateral. NA: nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: massive, finely serrated edge a uniform arc. LJ: large, moderately V-shaped. DF: low, originates near dorsal taii-body junction, tip rounded. BS: globular/depressed. CP: uniformiy dark. TL: 68/37. CI: Myers et al. (1991). Colostethus:CO: 85. GR: Costa Rica-Peru-Brazil. EMG: exotroph: lentic: benthic, arboreal, neustonic; lotic: neustonic; endotroph. LTRF: 2(2)/3[1], 212, 010. OA: typical: anteroventral,umbehform, largely absent in endotrophs. MP: medium dorsal gap: uniserial. SUP: row throughout extent of marginal papiiiae, absent, few laterally and distal to P-3. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium to wide, smooth arc. LJ: medium to wide, open V-shaped. DF: low, originates near dorsal td-body junction, tip rounded. BS: oval/depressed. CP: muted tones, dark line from nostril to snout common. TL: 15-30136. NO: C. nubicola group has umbelliform oral disc and jaw sheaths but no labial teeth. CI: Kaiser and Altig (1994); Lescure (1984); J. M. Savage (1968). Dendrobates: CO: 26. GR: Nicaragua-Peru-Brazil. EMG: exotroph: lentic: benthic, arboreal. LTRF: 1/1,2(2)/3[1]. OA: typical: anteroventral. MP: wide dorsal gap: uniserial, sometimes narrow ventral gap. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: mediun~to very wide, smooth arc. LJ: medium to wide, open U-shaped. DF: low, originates near dorsal taii-body junction, tip rounded to slightly pointed. BS: oval/depressed. CP: uniform and muted variations. TL: 15-25/36. CI: J. M. Savage (1968); Silverstone (1975). Epipedubates: CO: 24. GR: northern South America. EMG: exotroph: lentic: benthic, arboreal. LTRF: 2(2)/3[1-21. OA: typical: anteroventral.MP: wide dorsal gap: uniserial. SUP: few laterally, absent. DEM: lateral. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: narrow to medium, smooth arc. LJ: narrow to medium, open U-shaped. D E low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: u d o r m and muted variations. TL: 15-20136. CI: J. M. Savage (1968); Silverstone (1975). Mannophryne: CO: 9. GR: Venezuela-Tobago. EMG: exotroph: lentic: benthic, arboreal. EMG: exotroph: lentic: benthic, arboreal, neustonic; lotic: neustonic; endotroph. LTRF: 2(2)/3[1], 212, 010. OA: typical: anteroventral, umbelhform, largely absent in endotrophs. MP: medium dorsal gap: uniserial. SUP: row throughout extent of marginal papillae, absent, few laterally and distal to P-3. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium to wide, smooth arc. LJ: medium to wide, open V-shaped. D E low, originates near dorsal td-body junction, tip rounded. BS: oval/depressed. CP: muted tones, dark h e from nostril to snout common TL: 15-30136. NO: see Colostethus. CI: J. R. Dixon and Rivero-Blanco (1985). Minyobates: CO: 9. GR: Venezuela-Ecuador. EMG: exotroph: lentic: benthic. LTRF:

tnnitatU

Colostethw; C.fitatw

Dendrobates

RONALD ALTIG AND ROY W. McDIARMID

2(2)/3. OA: typical: anteroventral. MP: wide dorsai gap: uniserial. SUP: absent. DEM: absent, lateral. NA: nearer eye than snout. VT: rnedial. EP: dorsal. SP: sinistral. UJ: wide, srnooth arc. LJ: wide, open U-shaped. DE: low, originates near dorsai td-body junction, tip rounded. BS: ovai/depressed. CP: uniformly dark. TL: 12-18/36. CI: J. M. Savage (1968); Silverstone (1975). Nephalobates: CO: 5. G R Venezuela. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3. OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: absent. DEM: absent, lateral. NA: nearer eye than snout. VT: rnediai. EP: dorsai. SP: sinistrai. UJ: wide, srnooth arc. LJ: wide, open U-shaped. DE: low, originates near dorsal td-body junction, tip rounded. BS: ovai/depressed. CP: uniformly dark. TL: 12-18/36. NO: forrner Colostethus albogztlaris group. CI: La Marca and Mijares-U (1988). Phyllobates: CO: 5. G R Costa Rica-Colornbia. EMG: exotroph: lotic: benthic. LTRF: 2(2)/3. OA: typical: anteroventrai. MP: wide dorsal gap: uni- and biserial. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: rnediai. EP: dorsal. SP: sinistral. UJ: wide, srnooth arc. LJ: wide, open V-shaped. DE: low, originates near dorsai td-body junction, tip rounded. BS: oval/depressed. CP: uniforrnly dark. TL: 15-25/36. CI: Donnelly et ai. (1990b); Lescure (1976); J. M. Savage (1968). DISCOGLOSSIDAE: CO: 219. G R Europe. EMG: exotroph: lentic: benthic. LTRF: 2/3(1), biserial. OA: typical: anteroventrai. MP: complete: uniseriai. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: rnediai. EP: dorsai. SP: low on lefi side. UJ: narrow, srnooth arc. LJ: narrow, open U-shaped. DE: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: chainlike melanophore pattern in skin. TL: 3555/36. Alytes: CO: 4. G R Europe. EMG: exotroph: lentic: benthic. LTRF: 2/3(1), biserial. OA: typical: anteroventral. MP: complete: uniserial. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: rnedial. EP: dorsai. SP: low on lefi side. UJ: narrow, smooth arc. LJ: narrow, open U-shaped. DE: low, originates near dorsal td-body junction, tip rounded. BS: ovai/depressed. CP: chainlike rnelanophore pattem in skin. TL: 35-55/36. CI: Bouienger (1892); Lanza (1983). Dism~lossus: CO: 5. G R Europe-Syria. EMG: exotroph: lentic: benthic. LTRF: 2/3(1), biserial. OA: typical: anteroventral. MP: crnplete: uniserial. SUP: absent. DEM: lateral. NA. nearer snout than eye. VT: rnedial. EP: dorsai. SP: low on lefi side. UJ: narrow, srnooth arc. LJ: narrow, open U-shaped. DE: low, originates near dorsal tail-body junction, tip rounded: BS: oval/depressed. CP: chadke rnelanophore pattern in skin. TL: 35-55/36. CI: Bouienger (1892); Lanza (1983). HELEOPHRYNIDAE: CO: 115. G R South Afi-ica. EMG: exotroph: lotic: suctorial. LTRF: 4114-15(1). OA: typicai: ventral, upper jaw sheath absent, sometimes lower also. MP: complete: biserial. SUP: rows below distai rnost lower row. DEM: absent. NA: equidistant between eye and snout, slightly tubuiar. VT: rnedial. EP: dorsai. SP: sinistrai. UJ: absent. LJ: when present: rnedium, open V-shaped. DE: low, originates on taii rnuscle, tip rounded. BS: oblong/depressed. CP: rock/lichenlike rnottiing, yeiiow/gold ofien in background color in life. TL: 50-85/36. NO: fewer than 14 tooth rows on lower labium common; hypertrophed, isolated jaw serrations in the position of jaw sheaths of srnd tadpoles are soon lost (Visser 1985); large subbranchial rnuscle present ventrolaterdy along anterior two-thirds of body gives ventrai silhouette a distinaive shape. Heleophyne: CO: 5. CI: Channing (1984); Channing et ai. (1988); Van Dijk (1966). HEMISOTIDAE: CO: 118. G R sub-Saharan Africa. EMG: exotroph: lentic: nektonic. LTRP: 5(3-5)/4(1), 6(3-6)/3. OA: typicai: anteroventrai. MP: wide dorsai gap: uni- and biseriai, some ventrai papiae exceedingly long. SUP: absent, few laterally. DEM: absent. NA: srnd, doser to snout than eye. VT: dextrai. EP: lateral. SP: sinistral. UJ: rnedium to wide, strong rnedial convexity. LJ: rnedium, open U-shaped. DE: low, originates near dorsal taii-body junaion, tip pointed. BS: ovai/cornpressed. CP: stripes on taii, distal thrd dark, basai conneaive tissue sheath visible. TL: 40-60136. Hemks: CO: 8. CI: Lambiris (1988a, b); Van Dijk (1966); Wager (1986). HYLIDAE: CO: 411743. GR: Holaraic-Neotropic-Australopapua. EMG: exotroph: lentic: severai guilds; lotic: several guilds; endotroph. LTRF: rnany variations between 010 and

DIVERSITY

17/21. OA: typical: anteroventral, ventral, terminal. MP: dorsal gap, complete, dorsal and ventral gaps: variable, ofien bi- or multiserial. SUP: highly variable. DEM: absent, lateral. NA: round, variable in size, closer to snout than eye or reverse, ofien rimrned and countersunk. VT: dextrai, mediai. EP: lateral, dorsai. SP: sinistral, low on lefi side, ventral. LJ: highiy variable. DE: highiy variable. BS: oval to oblong/depressed, equidimensional and compressed. CP: highiy variable. TL: 20-60136. NO: marginal papiiiae always smder and more densely arranged than ranids, the most commonly confused taxa; few taxa with 1-3 anterior rows with medial gaps distal to nonbroken rows. Hemiphractinae: CO: 5/64. GR: Panama-South America. Cryptoba~rachus: CO: 3. GR: Colombia. EMG: endotroph. Fhctonotus: CO: 5. GR: Brazil-Venezuela-Trinidad. EMG: endotroph; exotroph: arboreal. LTRF: 010. OA: typical but lacking labial teeth; anteroventral. MP: dorsal gap: uniserial. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: ventral. UJ: medium, wide arc. LJ: narrow, wide U-shaped. DE: low, originates near dorsal td-body junction, tip round. BS: oblong/depressed. CP: iightly pigmented. TI,: 18/31. NO: exotrophic larvae are morphologicdy similar to those of nidicolous endotrophs. CI: Dueliman and Gray (1983); Heyer et al. (1990); and Weygoldt (1989). Gastrotheca: CO: 44. GR: Panama-Brazil. EMG: exotroph: lentic: benthic; endotroph. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: ventrolaterdy. DEM: absent. NA: equidistant between snout and eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: wide, open V-shaped. DF: low, originates near dorsal taii-body junction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 40-60136. CI: Wassersug and Dueliman (1984). Hemiphractus: CO: 5. GR: Panama-Bolivia. EMG: endotroph. Stgrania: CO: 7. GR: Venezuela-Guyana. EMG: endotroph. Hylinae: CO: 26/488. GR: Eurasia and North, Central, and South America. A h s : CO: 2. GR: eastern North America. EMG: exotroph: lentic: benthic. LTRP: 2(2)/ 2. OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: few laterdy. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, wide smooth arc. LJ: medium, U-shaped. DF: low, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: dorsum of t d muscle banded; taii tip usudy black. TL: 40-50136. NO: spiracular tube long. CI: Altig (1970); Orton (1952); A. H . Wright and Wright (1949). Anotheca: CO: 1. GR: Mexico-Panama. EMG: exotroph: lentic: arboreal. LTRF: 2(2)/ 2. OA: typical: nearly terminal. MP: complete: uniserial. SUP: few laterdy. DEM: absent. NA: about equidistant between snout and eye. VT: d m a l . EP: dorsal. SP: sinistrai. UJ: wide, long shdow arc, finely serrate. LJ: wide, open V-shaped. DE: low, originates near dorsal taii-body junction, tip rounded. BS: ovalJdepressed. CP: uniformiy purpiish brown. TI,: 45/38. NO: gut not coiled. CI: Robinson (1961); E. H . Taylor (1954). Apa~mphenodon:CO: 2. GR: Brazil-Venezuela. EMG: lentic: benthic. LTRF: 2(1)/56(1). OA: typical: anteroventral. MP: wide dorsal gap: mdtiserial. SUP: s m d group ventrolaterdy. DEM: absent. NA: -. VT: mediai. EP: lateral. SP: sinistral. UJ: -. LJ: -. DF: low, originates anterior to the dorsal td-body junction, tip pointed. BS: ovoid in dorsal view. CP: dark brown with irregular dark spots; taii muscle banded dorsdy. TL: 47/36. CI: H. R. de Silva (personal communication). Aplastodiscus: CO: 1. GR: Brazil. EMG: exotroph: lentic: benthic. LTRF: 2(2)/4(1). OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: lateral. SP: sinistral. UJ: medium, straight medial edge. LJ:medum, U-shaped. DE: low, originates at dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: distal one-third of taii darkened; anterior part of t d muscle with lateral stripe. TL: 68/38. CI: Caramaschi et al. (1980). Argenteohyu: CO: 1. GR: Argentina-Paraguay. EMG: exotroph: lentic: benthic. LTRF: 2(2)/4(1). OA: typicai: anteroventral. MP: medium dorsal gap: biserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open V-shaped. DF: somewhat arched, originates on body, tip pointed. BS: oval/depressed. CP: uniformiy brown. TL: 42/31. CI: De S (1983).

Ami~

RONALD ALTIG AND ROY W. McDIARMID

Duellmanohyla

Hyla; H. bromeliacia

Hyia; H. picadoi

Hyia; H. sarayauensiJ

Hyla; H. armata

Cal'tahyh: CO: 1. GR: Jamaica. EMG: exotroph: lentic: arboreal. LTRP: 110. OA: typical: anteroventral.MP: narrow dorsal gap: uniserial. SUP: throughout lower labium. DEM: absent. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, serrated edge a smooth arc. LJ: medium, open V-shaped. DF: somewhat arched, originates on body, tip pointed. BS: oval/depressed. CP: uniformly brown. TL: 42/31. CI: E. R. Dunn (1926). Cmythomantis: CO: 1. GR: Brazii. Tadpole unknown. Duellmanohyh: CO: 8. GR: Mexico-Costa Rica. EMG: exotroph: lotic: adherent. LTRF: 3(1,3)/3[1], 2(2)/3,2(2)/2. OA: typical: ventral. MP: complete: small and uniseriai. SUP: scattered throughout disc. DEM: midventral. NA: small, nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, coarsely serrate, srnooth arc to slightly indented medially. LJ: medium, coarsely serrate, V-shaped. DF: low, originates near dorsal tail-body junaion, tip rounded. BS: oval/depressed. CP: largely brown, some rnotthg on tail. TL: 33/26. NO: tooth rows much shorter than transverse diameter of oral disc. CI: J. A. CampbeU and Srnith (1992); D u e h a n (1970). Hyh: CO: 289. GR: North, Central and South America-Eurasia. EMG: exotroph: lotic: arboreal, benthic, adherent, clasping, suctorial; lentic: benthic, nektonic, arboreal. LTRF: many variations between 010 and 17/21; 2/3 most comrnon. OA: typical: terminal, anteroventral,ventral; reduced to various degrees. MP: dorsal gap, dorsal and ventral gaps: uni-, bi-, and multiserial. SUP: few to many laterally, distinct rows above upper and below lower tooth rows, absent. DEM: absent. NA: variou positions. VT: dextral, rnedial. EP: dorsal, lateral. SP: sinistral. UJ: many variations. LJ: many variations. DF: very low to very high, originates anterior to, near (most cornmon), or posterior to dorsai taii-body junaion, tip pointed to round. BS: oval to oblong/depressed, cornpressed or equidimensional. CP: variable. TL: 20-60/36. NO: a number of species groups have distinaive rnorphologies. CI: Caldweli (1974); D u e h a n (1970, 1978); Duellman and Altig (1978); Duellman and Fouquette (1968); D u e h a n and Tmeb (1989); Peixoto and Da Cmz (1983); J. M. Savage (1980b). Nyctimuntis: CO: 1. GR: Ecuador. Tadpole unknown. Osteocephalus: CO: 14. GR: Guianas-Venezuela-Amazon Basin-Argentina. EMG: exotroph: lentic: benthic, arboreal. LTRF: 2(2)/5(1), 3(2-5)/5(1). OA: typical: anteroventrai. MP: wide dorsal gap: biserial. SUP: few laterally. DEM: absent. NA: equidistant between eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow, wide smooth arc. LJ: narrow, V-shaped. DF: low, originates near dorsal taii-body junaion, tip rounded. BS: oval/depressed. CP: uniformly dark. TL: 49/37. CI: Duellman (1978); Duellman and Lescure (1973); Tmeb and Duellman (1970). Osteopilus: CO: 3. GR: Greater Antiiies-Florida. EMG: exotroph: lentic: benthic, arboreal. LTRF: 2(2)/4(1). OA: typical: anteroventral, ventral. MP: dorsal gap: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextrai, rnedial. EP: lateral, dorsal. SP: sinistral. UJ: medium, edge forms smooth arc. LJ: medium, deep rnedial indentation. DF: low, originates near dorsal taii-body junaion, tip slightly pointed. BS: oval/depressed. CP: uniformly dark or pale. TL: 37/41. CI: Duellman and Schwar~ (1958); Lannoo et al. (1987). Phtynohyas: CO: 5. GR: Mexico-Argentina. EMG: exotroph: lentic: nektonic, benthic, arboreal. LTRF: 4(1-2,4)/6(1), 3(1,3)/5(1),2(2)/4[1]. OA: typical: anteroventral. MP: rnedium dorsal gap: biserial. SUP: few ventrolaterally. DEM: absent. NA: nearer eye than snout. VT: rnedial. EP: lateral. SP: sinistral. UJ: medium, edge describes a smooth arc, coarsely serrate. LJ: medium, open U-shaped. DF: rnedium, originates on body, tip pointed. BS: oval/compressed. CP: taii rnucle striped. TL: 30-38/36. NO: unusual LTRF shared with TrachcephczLs;Pyburn (1967) shows a rnedial constriaion in the upper jaw sheath; Zweifel (1964) shows a rnediaiiy broken upper sheath; apparently random breaks, not gaps as noted above, cornrnon in lower rows. CI: Pyburn (1967); Zweifel (1964). Phyllodytes: CO: 7. GR: Trinidad-Brazil. EMG: exotroph: lentic: benthic, arboreai. LTRF: 2(2)/3. OA: typical: anteroventral. MP: wide dorsal gap: uni- and biserial. SUP: few laterally. DEM: absent. NA: nearer snout than eye. VT: rnedial. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. JJJ: rnedium, open U-shaped. DF: low, originates

DIVERSITY

near dorsal taii-bodyjunction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 29/36. CI: Bokermann (1966); F. M. Clarke et al. (1995); Giarretta (1996). Plectrobyla: CO: 16. GR: Mexico-Guatemala. EMG: exotroph: lotic: clasping. LTRF: 2(2)/3. OA: typicai: ventral. MP: complete: uni- and biserial. SUP: row above A-1 and below P-3, scattered laterally. DEM: absent. NA: small, nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, uniform arc, uniform graded series of serrations, hypertrophied serrations in specific sections. LJ: wide, open U-shaped. D E low, originates near dorsal tail-body junction, tip round. BS: oval/depressed. CP: uniformiy dark, some mottiing on tail. TL: 35-43/34. CI: J. A. Campbeii and Kubin (1990); Dueiiman (1970). Pseudacris: CO: 13. GR: North America. EMG: exotroph: lentic: nektonic, benthic. LTRF: 2(2)/3[1]. OA: typical: anteroventral.MP: medium dorsal gap: uni- and biserial, sometimes narrow ventral gap. SUP: few laterally. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: lateral. SP: sinistrai. narrow to medium, wide arc, sometimes medial zone straight. LJ: medium, V-shaped. D E medium to high, originates near dorsal tail-body junction, on body, tip pointed. BS: oval/compressed. CP: muted mottling, stripes or bicolored on tail. TL: 25-45/36. CI: Aitig (1970); D u e h a n (1970); A. H. Wright and Wright (1949). Pternobyla: CO: 2. GR: southern Arizona-Mexico. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3. OA: typical: anteroventral. MP: dorsal gap: uniserial dorsolaterally, biserial elsewhere. SUP: few laterally. DEM: absent. NA: small, equidistant between eyes and snout. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, edge a smooth arc. LJ: medium, open U-shaped. DF: high, originates on body, tip somewhat pointed with slight flagellum. BS: oval/compressed. CP: pale dorsolateral stripes. TI,: 38/36. CI: R. G. Webb (1963). Ptycbobyla: CO: 10. GR: Mexico-Panama. EMG: exotroph: lotic: benthic, suctorial. LTRF: 4(1,4)/7(1), 4(4)/7(1), 4(4)/6(1). OA: typical: nearly ventral. MP: complete: biserial. SUP: many lateray. DEM: absent. NA: small, nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. U J: medium, edge a smooth arc. LJ: medium, V-shaped. DF: low, originates near dorsal taii-body junction, tip pointed to rounded. BS: oval/ depressed. CP: dark, some minor mottling, especially on tail. TL: 33-39125-31. CI: J. A. Campbeii and Smith (1992); D u e h a n (1970). Scarthyla: CO: 1. GR: Peru. EMG: exotroph: lentic: nektonic. LTRF: 2(1-2)/2(1). OA: typical: nearly terminal. MP: wide dorsal and ventral gaps: uniserial. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: lateral. SP: sinistral. UJ: medium, prominent arch, coatsely serrate. LJ: wide, open U-shaped, coarsely serrate. DF: low, originates posterior to dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: tail h s with prominent serial blotches that look like incomplete bands. TL: 28/36. NO: see Scinax. CI: D u e h a n and De S (1988). Scinm: CO: 81. GR: Mexico-Uruguay. EMG: exotroph: lentic: nektonic, benthic. LTRF: 2(2)/3[1.]. OA: typical: anteroventral, almost terminal. MP: narrow dorsal gap: uni- and biserial; rostrata group: ventral gap 2 extent of short P-3 that sits upon an armiike derivative of the lower labium. SUP: few laterally. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, smooth arc or with medial convexity, sometimes coarsely serrate. LJ: medium to wide, V to open U-shaped. DF: low, originates near dorsal taii-body junction, tip pointed, or quite high, originates anterior to dorsal td-body junction, tip pointed, sometimes with flagellum. BS: oval/depressed or compressed. CP: iightly pigmented, eye iine common. TL: 20-30136. NO: see Scartbyla; at least rostrata and wbra groups with distinctive morphology of the oral apparatus. CI: D u e h a n (1970); Hero and Mijares-Urrutia (1995); Heyer et al. (1990); McDiarmid and Aitig (1990). Smilisca: CO: 6. GR: southern Texas-northern South America. EMG: exotroph: lentic: benthic; lotic: adherent. LTRF: 2(2)/3. OA: typical: anteroventral, ventral. MP: dorsal gap, complete: uni- and multiserial. SUP: many laterally. NA: nearer eye than snout. VT: dextral. EP: lateral, dorsal. SP: sinistral. UJ: medium to wide, edge a smooth arc or indentation medially. LJ: medium to wide, open U-shaped. D E low, slightly arched, originates near dorsal td-body junction or onto body, tip pointed. BS: oval/siightly de-

v:

Scinax; S. su~dlata

Scinm; S. staufiri

RONALD ALTIG AND ROY W. McDIARMID pressed. CP: uniformly brownish; pale dorsolateral stripe on body; dorsurn of tail muscle lightiy banded. TL: 2 2 4 0 / 3 0 4 0 . CI: Dueilman (1970). Spbaenmbyncbus: CO: 11. GR: Trinidad, Orinoco and Amazon basins of northern South America. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1).OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: few lateraiiy and ventrolateraiiy. DEM: absent. NA: smaii, nearer snout than eye. VT: medial. EP: lateral. SP: sinistral. UJ: wide, slight medial convexity. LJ: wide, very open U-shaped. DP: low, originates near dorsal tailbody junction, tip pointed. BS: oval/depressed. CP: stripes below eye and on mil. TL: 57/35. CI: Bokermann (1974); Da Cniz and Peixoto (1980a); Heyer et al. (1990). Tepuibyla: CO: 7. GR: eastern Venezuela. NO: no tadpoles known for the 0. rodnauezi group that was recently elevated to a genus distinct from Osteocepbalus. Tracbycepbalus: CO: 3. GR: Brazii-Ecuador. EMG: exotroph: lentic: nektonic. LTRF: 4(1-2,4)/6(1). OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: patch ventrolateraiiy. DEM: absent. NA: small, nearer snout than eye. VT: medial. EP: lateral. SP: sinistral. UJ: medium, forms smooth arc. LJ: mediurn, open V-shaped. DF: high, originates on body, tip pointed. BS: oval/cornpressed. CP: body and tail striped. TL: 381 34. NO: shares uncommon LTRF with Pbqmbyas. CI: McDiarmid and Altig (1990). Triprion: CO: 2. GR: western and southern Mexico. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3. OA: typical: anteroventral. MP: mediurn dorsal gap: biserial. SUP: few lateraiiy. DEM: absent. NA. nearer eye than snout. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, smooth arc. LJ: wide, broadly U-shaped. DE: low, originates at dorsal tailbody junction, tip rounded. BS: oval/slightly depressed. CP: udormly brown. TL: 27/ 30. CI: Duellman (1970). Xenobyla: CO: 2. GR: Brazii. EMG: exotroph: lentic: nektonic. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: absent. DEM: absent. NA: ncarer tip of snout than eye. VT: -. EP: lateral. SP: sinistral. UJ: medium, with moderate medial convexity LJ: wide, broadly U-shaped. DF: high, originates directly behmd plane of eye, tip with prominent flageiium. BS: oval/cornpressed. CP: bold bands on tail, prominent eye stripe. TL: 32/33. CI: Izecksohn (1997). Pelodryadiae: CO: 4/143. GR: Australia-New Guinea. Cyciurana: CO: 12. G k Australia. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1). J OA: typical: anteroventral. MP: wide dorsal gap: biserial. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, V-shaped. DF: low, originates near dorsal td-body junction, tip pointed. BS: oval/depressed. CP: unicolored to faintiy mottled or blotched. TL: 40-70/42. CI: Watson and Martin (1973). Litmia: CO: 104. GR: Australia-New Guinea. EMG: exotroph: lentic: benthic; lotic: benthic, suctorial. LTRF: 213, 2(2)/3[1], 2(2)/2. OA: typical: anteroventral, ventral. MP: narrow to wide dorsal gap: uni- and biserial; complete: biserial. SUP: absent, few lateraiiy, many lateraiiy. DEM: absent. NA: nearer eye than snout and reverse. VT: dextrai. EP: lateral, dorsal. SP: sinistral. UJ: narrow to wide, smooth arc, slight medial convexity. LJ: narrow to wide, open U-shaped. DF: low, originates near dorsal tail-body junction, tip pointed to rounded. BS: oval/depressed or compressed. CP: variable. TL: 30-60/36. NO: species groups not well defined (e.g., Barker and Grigg 1977; King 1981; Tyler 1968; Tyler and Davies 1978). CI: Watson and Martin (1973). Nyctimystes: CO: 24. GR: northern Australia-New Guinea. EMG: exotroph: lotic: suctorial. LTRF: 2/3. OA: typical: ventral. MP: complete: uniserial. SUP: radiaiiy oriented, elongate papillae across lower labium distal to P-3. DEM: absent. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, medial modification for suctorial existence. LJ: wide, open U-shaped. DF: low, originates posterior to dorsal tail-body junction, tip round. BS: oval/depressed. CP: muted mottiing. TL: 33/36. CI: Davies and Richards (1990); Tyler (1989). Peiudryas: CO: 3. GR: Australia. EMG: exotroph: lentic: benthic. LTRF: 2(1-2)/3. OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: absent. DEM: absent. NA: equidistant between eye and snout. VT: dextral. EP: lateral. SP: sinistral. UJ: med i u , smooth arc, coarsely serrate. LJ: medium, open V-shaped, coarsely serrate. DF: low, originates near or slightly posterior to dorsal tail-body junction, tip pointcd. BS: oval/depressed. CP: tail striped. TL: 41/41. NO: see Litoria. CI: Tyler et al. (1983).

Cycloma

DIVERSITY

Phyllomedusinae: CO: 6/48. G R Mexico-Bolivia. Agalychnis: CO: 8. G R Mexico-Ecuador. EMG: exotroph: lentic: suspension-rasper, nektonic, arboreai. MP: medium dorsai gap, complete: biseriai. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: dextrai. EP: lateral. SP: low on left side without mediai wall. UJ: medium to wide, smooth arc, slight mediai convexity. LJ: medium, open U-shaped. DF: low, originates near dorsai tail-body junction, tip rounded. BS: oval/ compressed. CP: lightly pigmented. TL: 30-60136. NO: ventrai fin often higher than dorsai fin. CI: Donneliy et ai. (1987); Duellman (1970); Hoogmoed and Cadle (1991). Hylomantis: CO: 2. G R Brazii. EMG: exotroph: lentic. NO: similar to tadpoles ofPhasmabyla (0.L. Peixoto, personai communication). Pachymedusa: CO: 1. G R western Mexico. EMG: exotroph: lentic: suspension-rasper. LTRE: 2(2)/3(1). OA: typicai: anteroventrd. MP: wide dorsai gap: uniserial. SUP: few bordering marginal papillae. DEM: absent. NA: nearer snout than eye. VT: dextrai. EP: laterai. SP: low on left side without mediai wall. UJ: medium, smaii serrations on smooth, broad arc. LJ: medium, open V-shaped. DF: medium, originates at dorsai taiibody junction, tip pointed. BS: ovai/compressed. CP: bluish gray in Me, tan in preservative. TL: 45/34. CI: Duellman (1970). Phasmubyla: CO: 4. G R Brazii. EMG: exotroph: lentic: neustonic. LTRP: 1/2(1). OA: typicai: umbelliform. MP: complete, absent: minute. SUP: scattered, some large and oriented radial to mouth. DEM: dorsai, dorsai and ventrai. NA: equidistant between eye and snout. VT: dextrai. EP: laterai. SP: low on left side without mediai waii. UJ: medium, prominent mediai convexity. LJ: medium, open V-shaped. BS: oval/compressed. DF: low, originates near or anterior to dorsai taii-body junction, tip pointed. CP: body uniformly paie, tail s i d a r or with distai portion darker. TL: 39/36. NO: ventrai h extends onto abdomen and higher than ventrai fin; vent tube not associated with ventrai fin. CI: Da Cruz (1980,1982); Heyer et ai. (1990). Phlynomedusa: CO: 5. G R Brazii. EMG: exotroph: lentic: suspension-rasper.LTRP: 2/ 3(1), 2(2)/3(1). OA: typicai: anteroventrd. MP: complete: bi- or multiserial. SUP: few to many lateraiiy. DEM: absent. NA: nearer snout than eye. VT: dextrai. EP: lateral. SP: low on left side without mediai waii. UJ: medium, smooth arc. LJ: medium, open Vshaped. DP: low, originates near dorsai taii-body junction, tip pointed. BS: ovai/compressed. CP: uniformiy paie. TL: 3046/28-38. NO: ventrai h often higher than dorsai h.CI: Da Cruz (1982); Heyer et ai. (1990); Izecksohn and Da Cruz (1976). Phyllomedusa: CO: 28. G R Costa Rica-Argentina. EMG: lentic: suspension-rasper. LTRE: 2(2)/3/(1). OA: typicai: anteroventrai.MP: medium dorsai gap: uni- and biseriai, sometimes with ventrai gap. SUP: few lateraiiy. DEM: absent. NA: nearer snout than eye. VT: dextrai. EP: laterai. SP: low on left side without mediai wall. UJ: medium, smooth arc with straight area medially. LJ: medium, V-shaped. BS: oval/compressed. DF: low, originates anterior of dorsai taii-body junaion, taii pointed with flagellum. CP: umformiy paie; diflse yeliow, orange to black area common in anterior part of lower h.TL: 30-50133-37. NO: ventrai h higher than dorsai fin and extends anteriorly onto abdomen. CI: Cei (1980); Da Cruz (1982); Duellman (1970). HYPEROLIIDAE: CO: 191232. G R sub-Saharan Africa-Madagascar. EMG: exotroph: lentic: benthic, nektonic; lotic: benthic. LTRP: variations between 010 and 513; 113 most commoniy. OA: typicai: anteroventrai,terminal. MP: medium to wide dorsai gap, sometimes narrow ventrai gap: uni- and biseriai. SUP: absent, few laterally, throughout extent of marginal papdlae. DEM: absent. NA: often small, usually nearer snout than eye. VT: dextrai, mediai. EP: laterai, dorsai. SP: sinistrai. UJ: narrow to wide and massive, smooth arc or mediai convexity, coarse serrations in some. LJ: narrow to wide, U- or V-shaped. DF: low, originates near dorsai taii-body junction, tip pointed to rounded or very high, originates anterior to dorsai td-body junaion, tip pointed as flagellum. BS: oval/depressed and compressed. CP: variable -usuaiiy rather uniform, may have striped taii or tip black. TL: 25-85/36. Hyperoliinae: CO: 121161. G R Africa-Madagascar-Seychelie Islands. AcanthiXalus: CO: 1. G R Cameroon-Zaire. EMG: exotroph: lentic arboreai. LTRE: 4(24)/3. OA: typicai: subterminai. MP: wide dorsai and narrow ventrai gaps: biseriai. SUP: abundant centripetally around lower labium. DEM: absent. NA: nearer snout than eye. VT: mediai. EP: dorsai. SP: sinistrai. UJ: wide, large mediai convexity. LJ: wide,

Phyllomedusa

Amnthhalw

RONALD ALTIG AND ROY W. McDIARMID

open U-shaped. DF: low, originates near dorsal taii-body junction, tip rounded. BS: oval/ depressed. CP: udormly dark. TL: 52/33. CI: Lamotte et al. (1959a). Af7walus: CO: 28. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRE O/ 0, 011. OA: typical: nearly terminal. MP: wide dorsal gap: uni- and biserial ventraiiy, large, widely spaced. SUP: few rnidventrdy. DEM: absent. NA: very smaii, nearer snout than eye, or reverse. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, smooth arc or with medial convexity, serrations often coarse. LJ: wide, smooth arc. DF: low, originates near dorsal tail-body junction, tip pointed. BS: oblongldepressed. CP: uniformly dotted. TL: 30-65136. NO: ecomorphologicalguild poorly known, may in fact be occur in midwater or near the surface. CI: Drewes et al. (1989); Pickersgiii (1984); Schgtz (1967, 1975). Aiextmoon: CO: 1. GR: Cameroon. EMG: exotroph: lotic: benthic. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: single row from tip of A-1. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, broad arc with medial area straight, coarsely serrate. LJ: medium, open Ushaped. DF: low, originates near dorsal taii-body junction, tip slightly pointed. BS: oval/ depressed. CP: uniformly brown, margins of fins largely unpigmented. TL: 29/33. CI: Amiet (1974b). Arleguinus: CO: 1. GR: Cameroon. EMG: exotroph: lentic: benthic. LTRF: 113. OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: many throughout the extent of the marginal papillae. DEM: absent. NA: nearer snout then eye. VT: dextral. EP: lateral. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsal td-body junction, tip pointed. BS: oval/depressed. CP: black stripe the length of dorsal haif of taii musde and body bordered above and below with pale stripes; dorsum of taii musde white. TL: 43/36. NO: adult pattern appears before metamorphosis. CI: Amiet (1976). Callkalus: CO: 1. GR: Rwanda-Zaire. Tadpole unknown. Chlorolius: CO: 1. GR: Cameroon-Gabon. EMG: exotroph: lentic: benthic. LTRE 1/ 3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: single row dosely adjacent to marginal papillae laterdy and ventraiiy. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, slight convexity mediaiiy, coarsely serrate. LJ: medium, V-shaped. DF: low, originates near dorsal td-body junction, tip pointed. BS: oval/depressed. CP: uniform body pigmentation with pale lines from snout to eyes; distal haifof tail lightly pigmented. TL: 38/36. CI: Amiet (1976). Chrysobatrachus: CO: 1. GR: Zaire. Tadpole unknown. Cwtothylux: CO: 2. GR: Cameroon-Zaire. EMG: exotroph: lentic: benthic. LTRF: 1/ 3. OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: single row extending from A-1. DEM: absent. NA: nearer snout than eye. VT: -. EP: dorsal. SP: sinistral. UJ: medium, serrated edge describes a smooth arc. LJ: narrow, open U-shaped. DF: low, originates near dorsal td-body junction, tip pointed. BS: oblongldepressed. CP: body brown, td faintly striped. TL: 22/26. CI: Larnotte et ai. (1959b). Hetwixalus: CO: 7. GR: Madagascar. EMG: exotroph: lentic: benthic. LTRF: 1/3(1). OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: few lateraiiy and row inside marginal papiilae of lower labium. DEM: absent. NA: nearer snout than eye. VT: medial. EP: lateral. SP: sinistral. UJ: medium, wide, smooth arc, coarsely serrate. LJ: medium, open V-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/compressed. CP: uniform body, blotched t d . TL: 35/36. CI: Arnoult and Razarihelisoa (1967); Blommers-Schlosser (1982); Blommers-Schlosser and Blanc (1991). Hyprolius: CO: 116. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 1/3[1], 213. OA: typical: anteroventral. MP: wide dorsal gap: uni- and biserial, ventral papillae sometimes elongate. SUP: absent. DEM: absent. NA: aperture margin sometimes papiilate. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, smooth arc or slight medial convexity, serrations coarse. LJ: wide, open U-shaped. DF: low, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: variable: unicolored or striped tail, t tip black, fin margins dark. TL: 30-50136. NO: P-3 usuaiiy much shorter d than P-2. CI: Amiet (1979); Channing and De Capona (1987); Lamotte and Perret (1963~); Schigtz (1967,1975).

DIVERSITY

Kassinula: CO: 1. GR: Zambia-Zaire. Tadpole unknown. Neswnixalus: CO: 2. GR: Sao Tom. Tadpole unknown. Kassininae: CO: 5/21. GR: sub-Saharan Africa. Kasrina: CO: 13. GR: sub-Saharan Africa. EMG: exotroph: lentic: nektonic. LTRF: l/ 2(1) or 113. OA: typical: anteroventral. MP: dorsal gap: wide; ventral gap: narrow or absent. SUP: numerous lateraiiy. DEM: absent. NA: closer to snout than eye. VT: medial. EP: lateral. SP: sinistral. UJ: massive, smooth but prominent arc. LJ: massive robust, very widely V-shaped, at least medially. DF: very hgh, originates weii anterior to dorsal tail-body junction, tip pointed. BS: oval/compressed. CP: clear parts of taii fins often red or orange. TL: to 83/36. NO: keratinized area on lower disc below jaw sheath sometimes present. CI: Largen (1975); Schigtz (1975); Wager (1986). Opistbothyfw: CO: 1. GR: Cameroon-Zaire. EMG: exotroph: lotic: benthic. LTRF: O/ O. OA: typical: reduced. MP: wide dorsai gap: uniserial and large. SUP: absent. DEM: absent. NA: tiny, nearer the snout than the eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: very narrow, smooth arc. LJ: wider than upper, open V-shaped. DF: low, originates near dorsal tad-body junction, tip slightly pointed. BS: ovalJdepressed. CP: body brown, fins and taii muscle mottled; throat dark. TL: 26/36. CI: Arniet (1974a). Parwssina: CO: 2. GR: Ethiopia. EMG: exotroph: lentic: nektonic. LTRF: 1/3(2). OA: typicai: anteroventral. MP: wide dorsai gap, very narrow ventral gap where P-3 positioned. SUP: many lateraiiy. DEM: absent. NA: -. VT: medial. EP: lateral. SP: sinistral. UJ: massive, serrated edge describes smooth arc. LJ: massive, open V-shaped. DE: tall, highIy arched, originates weli anterior of the dorsal tail-body junction, tip pointed. BS: oval/compressed. CP: uniformly dark on body, taii dark in distai third. TL: 54/36. NO: keratinized area posterolateral to lower jaw sheath. CI: Largen (1975). Phlyctimantis: CO: 4. GR: Ivory Coast-Tanzania. EMG: exotroph: lentic: benthic. LTRF: 1/3[1-21. OA: typical: anteroventral. MP: wide dorsal gap: biserial. SUP: numerous throughout extent of marginal papiiiae. DEM: absent. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, serrated margin describes a fairly uniform arc. LJ: medium, open U-shaped. DF: low, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: uniformly pale with blotches. TL: 49/34. NO: P-3 exceptionally short. CI: Guib and Lamotte (1958); Schi~tz (1975). S e m v w ~ l u s CO: 1. GR: South Africa. EMG: exotroph: lentic: nektonic. LTRE 1/ : 3(1). OA: typical: anteroventral. MP: dorsal gap: biserial. SUP: few lateraiiy. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: lateral. SP: sinistral. UJ: wide, coarsely serrate. LJ: massive, open U-shaped. DE: high, originates on body, tip pointed. BS: ovai/compressed. CP: distinct dark stripe on taii muscle. T, 58/36. NO: P-3 excepI: tionally short. CI: Wager (1986). Leptopelinae: CO: 1/49. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 3-5/3[1], usually 2-n rows on upper labium broken medially. OA: typical: anteroventral. MP: wide dorsal gap: uniserial dorsaiiy, biserial ventrally. SUP: absent. DEM: absent. NA: smaii, nearer snout than eye. VT: dextral. EP: dorsai. SP: sinistral. UJ: wide, medial indentation. LJ: wide, V-shaped. DP: low, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: uniformly dark. TL: 49/36. Leptopelis: CO: 49. CI: Drewes et ai. (1989); Lamotte and Perret (1961b); Lamotte et ai. (1959b); Schi0tz (1967, 1975). Tachycneminae: CO: 111. GR: Seycheile Islands. EMG: exotroph: lotic: benthic. LTRF: 3(211)/3. OA: typicai: anteroventral. MP: wide dorsai gap: uniseriai midventraiiy. SUP: -. DEM: absent. NA: very smaii, nearer to snout than eye. VT: dextrai. EP: dorsai. SP: sinistral. UJ: wide, broad arch. LJ: medium, V-shaped. DP: low, originates distal to body. BS: oval/depressed. CP: uniformly paie brown, venter clear in preservative. TI,: 44/27. TachynemZr: CO: 1. GR: Seycheile Islands. CI: specimens. LEIOPELMATIDAE: CO: 113 GR: New Zeaiand. EMG: endotroph. Leiopelma: CO: 3. GR: New Zeaiand. EMG: endotroph. LEPTODACTYLIDAE: CO: 481850. GR: southern United States-South ArnericaCaribbean. EMG: exotroph: lentic: various guilds; lotic: various guilds; endotroph. LTRF: many variations between 010 and 919, 213 most common. OA: typical: anteroventrai or aii

LeptopeLis

RONALD ALTIG AND ROY W. McDIARMID keratinized mouthparts absent. MP: medium to wide dorsal gap, complete, narrow ventral gap: uni- and biserial. SUP: absent, few laterally or dorsolateraiiy. DEM: absent, lateral. NA: nearer snout than eye or reverse. VT: medial, dextral or sinistral. EP: dorsal. SP: sinistral except in Lqt'dubatrachus (dual/lateral). UJ: variable, narrow to wide, usually a smooth arc, incised or not. LJ: variable, narrow to wide, U to V-shaped. DF: low, originates near, slightiy anterior to or weii posterior to the dorsal tail-body junction, tip rounded to pointed. BS: round, oval to oblong/depressed. CP: variable but uniformly dark is cornmon. TL: 15-1301 36. Ceratophryinae: CO: 6/33. G R South America. Ceratophlys: CO: 8. G R South America. EMG: exotroph: lentic: benthic, often carnivorous and spends considerable time swimming in midwater. LTRF: 6(3-6)/7(3), 9(7-9)/ 9(1-3) but teeth often lost. OA: typical: nearly terminal. MP: complete: large and sparsely distributed. SUP: absent. DEM: absent. NA: nearer snout than eye, or reverse. VT: medial, dextral or sinistral. EP: dorsal. SP: sinistral. UJ: wide, medial notch. LJ: wide, medial convexity complimentary to upper sheath. DE: low, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: mostiy uniformly dark. TL: 40-70136. NO: jaw musculature hypertrophied. CI: Cei (1980); D u e h a n (1978); Honegger et ai. (1985). Champhlys: CO: 1. G R Argentina. EMG: exotroph: lentic: benthic. LTRF: 1/2(1). OA: typical: anteroventral. MP: complete: uniserial, widely spaced. SUP: absent. DEM: lateral. NA: large, nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, open V-shaped. DF: low, originates at dorsal td-body junction, tip pointed. BS: oval/depressed. CP: uniformly gray, tail muscle bicolored. TL: 40136. NO: unique flap projects upward from tip of snout. CI: Faivovich and Carrizo (1992). Lepidubatvachus: CO: 3. G R southern South Arnerica. EMG: exotroph: lentic: carnivore. LTRF: 010. Ok. slitlike mouth exceptiondy large and terminal. MP: complete: uniserial. SUP: absent. DEM: absent. NA. nearer eye than snout. VT: medial. EP: dorsal. SP: dual and lateral, but not homologous to spiracles of pipids and rhinophrynids. UJ: few, widely spaced jaw serrations without basai sheaths on one or both jaws. LJ: see upper jaw. DF: low, originates near dorsal taii-body junction, tip broadly pointed. BS: round/strongly depressed, widest near plane of eyes. CP: unormly gray or faintiy mottied. TL: 40-60136. NO: Cei (1980) states that L. llanensis has one upper and one lower tooth row; based on our observations of L. 1 4 , we note that these are actually isolated jaw serrations, tooth rows are absent; this condition probably occurs in aii three species. CI: Cei (1980); Ruibal and Thomas (1988). M ~ * c y ~ e ' n w ~ ~ CO: :1. GR: Brazii. EMG: exotroph: lentic: benthic. LTRF: 2(2)/ ttus 3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: absent. DEM: lateral. NA: equidistant between eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: very narrow, wide, smooth arc, serrations lacking. LJ: very narrow, open U-shaped, serrations laterally. DF: low, originates at dorsal taii-body junction, tip pointed. BS: oblong/depressed. CP: brownish-gray, reticulated with nonpigmented areas, dorsurn of tail muscle banded, edges of fins with black areas. TL: 35/30. CI: Abravaya and Jackson (1978). Oduntophrynus: CO: 8. G R eastern and southern South America. EMG: exotroph: lentic: benthic, lotic: benthic. LTRF: 2(2)/3[1.]. OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextrai. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, open U-shaped. DF: low originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: body dark, taii uniform to blotched. TL: 35-45/36. CI: Cararnaschi (1979); Cei (1980). Proceuatuphlys: CO: 12. G R Brazii-Paraguay. EMG: exotroph: lentic: benthic. LTRF: 2/3(1), 2(2)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: absent, few anterolateral. DEM: lateral, two ventrolateral. NA: large, not round, nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, open V-shaped. DF: low, originates near tail-body junction, tip rounded. BS: oval/depressed. CP: tail muscle weakly banded, body and fins speckied. TL: 33/36. CI: Da Cruz and Peixoto (1980b); Izecksohn et al. (1979). Hylodinae: CO: 3/28. G R eastern and southern South America. Crosso~Lus: CO: 9. G R southern South America. EMG: exotroph: lotic: benthic.

Lepiriobatracbus

DIVERSITY

LTRF: 3(1,3)/5,2(2)/3[1]. OA: typicai: anteroventrd. MP: medium dorsai gap: uniseriai. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: dorsai. SP: sinistrai. UJ: wide, uniform arc and coarsely serrate. LJ: wide, open U-shaped, coarsely serrate. DF: low, originates near dorsai taii-body junction, tip rounded. BS: oval/ depressed. CP: body uniformiy dark, taii and fins with dark specks. TL: 52/37. CI: Caramaschi and Kisteumacher (1989); Caramasch and Sazima (1985); Cei (1980). Hylodes: CO: 15. GR: eastern South America. EMG: exotroph: lotic: benthic. LTRP: 2(2)/3[1], 2(2)/4. OA: typicai: anteroventrai. MP: medium dorsai gap: uniseriai. SUP: absent, few lateray. DEM: lateral. NA: nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc but coarsely serrate, mediai serration hypertrophied. LJ:wide, open U-shaped, coarsely serrate. DF: low, originates near dorsai taii-body junction, tip pointed. BS: ovai/depressed. CP: uniformiy dark. TL: 36/25. CI: Bokermann (1966); Heyer et ai. (1990); Sazima and Bokermann (1982). Mguelosia: CO: 4. GR: Brazil. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1). OA: typicai: anteroventrd. MP: wide dorsal gap: sma and uniserial. SUP: few lateray near emargination and below P-3. DEM: lateral. NA: sma, nearer snout than eye. VT: dextrai. EP: dorsal. SP: sinistrai. UJ: wide, edge describes smooth arc. LJ: wide, V-shaped. DF: low, originates near dorsai tail-body junction, tip pointed. BS: oval/depressed. CP: uniforrnly dark. TL: 128125. CI: Giarretta et ai. (1993); Heyer et ai. (1990). Leptodactylinae: CO: 101121. GR: southem Texas-Chle-Antilies. Adenomera: CO: 6. GR: South America east of Andes. EMG: endotroph. NO: De la Riva (1995), based on observations in Bolivia, noted that a species ofAdenomera are not entirely endotrophic; if the larvae of some forms gain access to free water, they can develop as typical lentic-inhabiting, leptodactylid tadpoles and resemble the tadpoles of some Leptodactylus spp., LTRF 2(2)/3(1). CI: De Ia Riva (1995); Heyer and Silverstone (1969). Edahhinu: CO: 2. GR: Colombia-Peru. EMG: exotroph: lentic: benthic. LTRF: 2(2)/ 3(1). OA: typicai: anteroventral. MP: dorsal gap: biseriai. SUP: few dorsolaterdy. DEM: absent. NA: equidistant between eye and snout. VT: dextrai. EP: dorsal. SP: sinistrai. UJ: narrow, wide uniform arc. LJ: medium, open U-shaped. DF: low, originates near dorsai taii-body junction, tip rounded. BS: ovai/depressed. CP: uniformly pale brown. TL: 14/25. CI: De Ia Riva (1995); Schiuter (1990). Hydroluetare: CO: 1. GR: Colombia-Peru. Tadpole unknown. Leptodaqlus: CO: 50. GR: southern Texas-Mexico-South America-West Indies. EMG: exotroph: lentic: benthic. LTRF: 213, 2(2)/3[.L]. OA: typicai: anteroventral. MP: medium to wide dorsai gap: uniserial, narrow ventral gap rarely. SUP: absent. DEM: laterai, absent. NA: nearer snout than eye. VT: medial. EP: dorsai. SP: sinistrai. UJ: narrow to medium, wide smooth arc. LJ: narrow to medium, U- or V-shaped. DF: low, originates Lept0-h~; L. melanonotw near dorsai tail-body junaion, tip rounded to pointed. BS: ovai/depressed. CP: often uniformiy dark. TL: 25-60136. NO: tadpoles of 4-5 species groups can be recognized by features more subtle than presented here. CI: Cei (1980); Heyer (1973a); Sazima ~eptoakty1w; pentaductyLus L. and Bokermann (1978). Lithodytes: CO: 1. GR: Venezuela-Ecuador-Peru. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1). OA: typicai: anteroventrai. MP: wide dorsai gap. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: mediai. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc. LJ: narrow, wide, open U-shaped. DF: low, originates near dorsai taii-body junction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL:47/34. NO: large lower labium extends far beyond lower tooth row. CI: Lamar and Wild (1995); Regos and Schiuter (1984). Paratelmutobius: CO: 3. GR: Brazil. EMG: exotroph: lotic or lentic: benthic. LTRF: 2(2)/3(1). OA: typicai: anteroventrai. MP: wide dorsal gap: uniseriai midventray. SUP: absent. DEM: lateral, absent. NA: nearer eye than snout. VT: dextrai, mediai. EP: dorsai. SP: sinistrai. UJ: narrow, edge a smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsai tail-body junaion, tip bluntly pointed. BS: oval/depressed. CP: uniformiy paie brownish, belly lightly pigmented, fins nonpigmented to brownish, tail muscle blotched. TL: 26-32/37. CI: Cardoso and Haddad (1990); Giarretta and Castanho (1990). Physaluemus: CO: 37. GR: Mexico-South Arnerica. EMG: exotroph: lentic: benthic.

RONALD ALTIG AND ROY W. McDIARMID

LTRF: 2(2)/2,2/3(1),2(2)/3(1).OA: typical: anteroventral. MP: wide dorsal gap: uniserial, sometimes ventral gap. SUP: few laterdy. DEM: lateral. NA: equidistant between snout and eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: medium, open V-shaped. DF: low, originates near dorsal td-body junction, tip pointed. BS: oval/ depressed. CP: uniformly dark, mottied. TL: 23/38. NO: some (e.g., l?petersi) with prominent, white glands on body. CI: Barrio (1953); Heyer et al. (1990). Pieurodema: CO: 12. GR: southern South America. EMG: exotroph: lentic: benthic. LTRF: 2(2)/2, 2(2)/3[1]. OA: typical: anteroventral. MP: wide dorsal gap: uniserial, ventral gap sometimes. SUP: rare lateraiiy. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, U-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/dcpressed. CP: uniformly dark, tail muscle mottied. TL: 25/36. CI: Cei (1980); Leon-Ochoa and Donoso-Barros (1970); Peixoto (1982). Pseudopaludicoiu: CO: 8. GR: southern South America. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3. OA: typical: anteroventral. MP: wide dorsal and narrow ventral gaps: uniserial. SUP: row(s) ventrolaterdy. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, wide smooth arc. LJ: medium, wide Ushaped. DF: low, originates near or anterior to dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: uniformly dark. TL: 22/36. CI: Barrio (1953); Cei (1980). Vanwlinius: CO: 1. GR: Ecuador-Peru. EMG: exotroph: lentic: benthic. LTRF: 213. OA: typical: anteroventral. MP: dorsal gap: uniserial. SUP: -. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: narrow. LJ: narrow. DF: low, originates near dorsal tail-bodyjunction, tip rounded. BS: oval/depressed. CP: uniformly tan or brown. TL: 25/30. CI: Duellman (1978). Telmatobiinae: CO: 291668. GR: central to southern South America. Adelopbqvne: CO: 2. GR: Guiana Shield. EMG: endotroph. Alsodes: CO: 11. GR: Chile-Argentina. EMG: exotroph: lotic: benthic. LTRF: 2(2)/ 3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: few dorsolateraiiy and row(s) distal to P-3. DEM: lateral. NA: equidistant between eye and snout, nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, open V-shaped. DF: low, originates distal to dorsal td-body junction, tip broadly rounded to pointed. BS: oval/depressed. CP: uniformly dark, few s m d spots on t d . TL: 60-75136. CI: Diaz and Valencia (1985). Atelognathus: CO: 7. GR: Chile-Argentina. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1).OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: few lateraiiy. DEM: lateral. NA: equidistant from eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, medial zone of serrate edge straight. LJ: wide, open V-shaped. DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: brown, sparsely pigmented, belly transparent. TL: 57/36. CI: Cei (1964, 1980). Atqpqphlynus: CO: 1. GR: Colombia. Tadpole unknown. Baqvcholos: CO: 2. GR: Ecuador-Brazil. EMG: endotroph. Batrachqphqvnus: CO: 2. GR: Peru-Bolivia. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: biserial. SUP: throughout extent of marginal papillae. DEM: absent. NA: -. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, open U-shaped. DF: low, originates at dorsal td-body junction, tip broadly rounded. BS: oval/depressed. CP: uniformly dark, smaii spots on fins. TL: 67/26. CI: Sinsch (1990). Bamachyla: CO: 3. GR: ChdeArgentina. EMG: exotroph: lentic: benthic. LTRF: 2(2)/ 3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: dorso- and ventrolaterdy. DEM: lateral. NA: equidistant between eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open V-shaped. DF: lou; originates near dorsal tail-body junction, tip roiinded. BS: oval/depressed. CP: uniformly dark, t d speckled. TL: 27-37/33. NO: early stages develop like nidicolous endotrophs. CI: Cei (1980); Cei and Capurro (1958); Formas and Pugn (1971). Caudiverbera: CO: 1. GR: Chile. EMG: exotroph: lotic: benthic. LTRF: 2(2)/3(1).OA: typical: anteroventral. MP: medium dorsal gap: uniserial. SUP: few laterdy. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow, edge describes smooth arc. LJ: narrow, open U-shaped. DF: low, originates at dorsal tail-body

DIVERSITY

junction, tip pointed. BS: ovai/depressed. CP: paie green in life, distai third of tail dark. TL: 95/37. NO: Cei (1962) shows a LTRF of 3(2)/3(1) and p-2 and P-3 with nonrnedial gaps. CI: Diaz and Vaiencia (1985); Jorquera and Izquierdo (1964). Cromductylodes: CO: 3. GR: Brazil. EMG: exotroph: lentic: arboreai. LTRF: 2(2)/3. OA: typicai: anteroventral. MP: wide dorsai gap: uniserial. SUP: absent. DEM: laterai. NA: nearer snout than eye. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: medium, small but prominent mediai convexity, basai margin aiso with convexity. LJ: medium, open Vshaped. DF: low, originates near dorsai tail-body junction, tip round. BS: oblong/depressed. CP: uniforrnly dark. TL: 19/35. CI: Peixoto (1981). Cycloramphus: CO: 25. GR: Brazil. EMG: exotroph: lotic: semiterrestriai; endotroph. LTRF: 213, 2(1-2)/3(1), 2/3(1), 2(1)/3(1). OA. typicai: anteroventral, ventrai. MP: wide dorsai gap: uniseriai. SUP: absent. DEM: absent. NA: see notes. VT: mediai. EP: dorsai. SP: sinistral. UJ: medium, strongly arched. LJ: medium, strongly arched. DF: low, originates weli posterior of dorsai tail-body junction, tip pointed. BS: oblong to oval/depressed. CP: umform to mottled, tail muscle may be slightly banded. TL: 2 0 4 0 1 36. NO: Heyer et ai. (1990) reported nares absent in C. boraceiensis but seem apparent in figure; Heyer (1983a, b) did not locate a spiracle on C. boraceiensis or C. v a h ; may have parts of abdominal w d expanded as a free flap. CI: Heyer (1983a, b). Dirchidodagylus: CO: 2. GR: Venezuela. EMG: endotroph. Eeutherodactylus: CO: 515. GR: North, Central, and South America-Caribbean. EMG: endotroph. Euparkerella: CO: 4. GR: Brazil. EMG: endotroph. Eupsophw: CO: 8. GR: Chile-Argentina. EMG: endotroph. Geobatrachus: CO: 1. GR: Colombia. EMG: endotroph. Holoaden: CO: 2. GR: Brazil. EMG: endotroph. Hylorina: CO: 1. GR: Chile-Argentina. EMG: exotroph: lentic: benthic. LTRP: 2(2)/ 3(1). OA: typicai: anteroventral. MP: wide dorsal gap: uniserial. SUP: few centripetally arranged throughout extent of marginai papiilae. DEM: laterai. NA: equidistant between snout and eyes. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: wide, serrated edge alrnost transversely straight. LJ: wide, open U-shaped. DF: somewhat arched, originates near dorsai tail-body junction, tip broadiy rounded. BS: oval/depressed. CP: uniform with dark spots throughout t d . TL: 51/38. CI: Barrio (1976); Formas and Pugn (1978). Insuetophlynw: CO: 1. GR: Chile. EMG: exotroph: lotic: benthic. LTRF: 2(2)/2(1-2) or' less. OA: typicai: ventrai, jaw sheaths and labial tooth rows reduced. MP: medium dorsai gap: uniserial, papillae very s m d . SUP: absent. DEM: absent. NA: s m d , nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: s m d , weakly developed but serrate. LJ: similar and U-shaped. DF: low, originates at dorsai td-body junction, tip broadiy rounded. BS: ovai/depressed. CP: greenish brown in life, irregularly mottled. TL: 61/37. NO: lower tooth rows sometimes absent, one drawing of oral variation shows complete marginal papiiiae, laterai line visible. CI: Diaz and Vaiencia (1985); Formas et al. (1980). Ischnucnema: CO: 3. GR: Amazonia-Brazil. EMG: endotroph. Limnumedusa: CO: 1. GR: Brazil-Argentina. EMG: exotroph:. LTRF: 2(2)/3(1). OA: typicai: anteroventral. MP: medium dorsai gap: uniserial. SUP: numerous aiong d marginal papiiiae. DEM: laterai. NA: nearer eye than snout. VT: mediai. EP: dorsai. SP: sinistrai. UJ: wide, uniform arc. LJ: medium, open U-shaped. DF: low, originates near dorsai tail-body junction, tip pointed. BS: ovai/depressed. CP: body uniform, tail boldiy spotted. TL: 42/34. CI: Cei 1980; Gehrau and De S (1980). Phlynupus: CO: 20. GR: Andean South America. EMG: endotroph. Phyllonaster: CO: 4. GR: Pem-Ecuador. EMG: endotroph. Phyzelaphlyne: CO: 1. GR: Brazil. EMG: endotroph. Scythrophlyr: CO: 1. GR: Brazil. Tadpole unknown. Somuncuria: CO: 1. GR: Argentina. EMG: exotroph: lotic: benthic. LTRF: 2(2)/3(1). OA: typicai: anteroventral. MP: wide dorsai and smaer ventral gaps: biseriai. SUP: absent. DEM: lateral. NA: equidistant between eye and snout. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: wide, smooth arc, or perhaps a slight mediai concavity in margin. LJ: wide, open U-shaped. DF: low, originates near dorsai tail-body junction, tip rounded. BS: ovai/depressed. CP: body dark, tail blotched. TL: 46/36. CI: Cei (1980).

Insuetophvynw

RONALD ALTIG AND ROY W. McDIARMID

Thoropa

Telmatobius: CO: 35. GR: Ecuador-Argentina. EMG: exotroph: lentic: benthic; lotic: benthic. LTRF: 2(2)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: uni- and biserial. SUP: few dorsolateraiiy, lateraliy and distal to P-3. DEM: lateral. NA: equidistant between eye and snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, edge a smooth arc. LJ: wide, open U-shaped. DF: low, originates near dorsal taii-body junction, tip rounded to pointed. BS: oval/depressed. CP: uniform to muted motthng. TL: 60100137. CI: Diaz and Valencia (1985); Lavilla (1984a, 1985,1988). Telmatobufo: CO: 3. GR: Chile. EMG: exotroph: lotic: suctorial. LTRP: 2/3(1). OA: typical: ventral. MP: complete: uniserial. SUP: muitiple rows centripetaiiyto ali marginal papillae except at lateral margins of disc. DEM: lateral. NA: nearer eye than snout. VT: dextral, vent flap underiies tube. EP: dorsal. SP: sinistral. UJ: medium, open V-shaped. LJ: medium, open V-shaped, "internal borders smooth" (Diaz et al. 1983). DF: low, originates weii posterior of dorsal td-body junction, tip broadly pointed. BS: oval/depressed, may narrow appreciably posterior to oral disc. CP: dark brown, orange protuberances over body dorsum. TL: 107140. CI: Diaz et al. (1983); Formas (1972). Thopa: CO: 5. GR: Brazil. EMG: exotroph: lotic: serniterrestrial. LTRP: 2(1-2)/3(1). OA: typical: ventral. MP: wide dorsal gap: uniserial. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, strongly arched and coarsely serrate. LJ: wide, strongly arched. DF: low, originates on taii muscle weii posterior to dorsal taii-body junction, almost absent in some, tip rounded. BS: obIongJdepressed. CP: uniform dark to mottled; taii muscle banded. TL: 20-30136. NO: some have parts of lateral margins of abdominal wali expanded as a free flap; nidicolous endotrophs may be involved (Bokermann 1965). CI: Caramaschi and Sazirna (1984). Z1u:haenus: CO: 3. GR: Brazil-Argentina. EMG: endotroph.
MEGOPHRYIDAE: CO: 9/79. GR: northern India-China-Philippines. EMG: exotroph: lotic: clasping, neustonic, benthic; endotroph. LTRF: 010 to 717. OA: typical: anteroventral, ventral; umbeliiform: oral disc bi-triangular. MP: narrow dorsal gap, complete, absent: uniserial. SUP: scattered near mouth, few lateraliy, some arranged radialiy to mouth. DEM: absent, dorsal and ventral. NA: equidistant between eye and snout, nearer eye than snout or reverse. VT: dextral, medial. EP: dorsal, lateral. SP: sinistral but weii below longitudinal axis. UJ: absent, medium to wide, smali medial convexity or not. LJ: absent, medium to wide, Uor V-shaped. DF: low, originates near or posterior to dorsal tail-bodyjunaion, tip rounded to pointed. BS: oval to elongate/depressed. CP: muted mottiing to unicolored. TL: 25-75/36. Atympanophys: CO: 1. GR: China. Tadpole unknown. Bv1u:hytamophys: CO: 1. GR: Burma-China-Thdand. Tadpole unknown. Leptubvachela: CO: 7. GR: Borne+Natuna Islands. EMG: exotroph: lotic: benthic. LTRF: 010. OA: typical: ventral. MP: narrow dorsal gap: uniserial. SUP: scattered near mouth. DEM: dorsal and ventral. NA: nearer snout than eye, one medial projection. VT: dextral. EP: dorsal. SP: sinistral but weii below longitudinal axis. UJ: medium, smali medial convexity. LJ: medium, open U-shaped. DF: low, especialiy in proximal third, originates near dorsal tail-body junction, tip round. BS: oblong/depressed. CP: uniformly dark. TL: 3045/36. CI: Inger (1983,1985). Leptobrachium: CO: 10. GR: Indonesia-China-Philippines. EMG: exotroph: lotic: benthic. LTRP: 3(1-3)/4(1-3) to 7(2-6)/6(1-5). OA: typical: anteroventral to ventral. MP: complete: uniserial. SUP: few lateraliy. DEM: absent. NA: equidistant between eye and snout, 1-3 medial projeaions. VT: dextral. EP: dorsal. SP: sinistral but weii below longitudinal axis. UJ: wide, smooth arc. LJ: wide, wide V-shaped. DF: low, originates near dorsal tail-body junaion, tip pointed to rounded. BS: oval to oblongJdepressed. CP: uniforndy dark. TL: 30-50136. NO: LTRF very variable, integumentary glands sometimes present, submarginal papillae commonly have labial teeth. CI: Berry (1972); Inger (1983, 1985). Leptohh: CO: 7. GR: Burma-China-Borneo. EMG: exotroph: lotic: suctorial, clasping. LTRF: 5(2-5)/3(1-2). OA: typical: ventral. MP: narrow dorsal gap or complete: uniserial. SUP: scattered above and below mouth, probably rudimentary tooth ridges. DEM: dorsal and ventral. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistrai but weii below longituduial axis. UJ: wide, smooth arc. LJ: wide, open U-shaped. DF: low, particuiarly in first fourth, originates distal to dorsal td-body junction, tip

L.eptobracheUa

Lqtobrnchium

DIVERSITY

rounded. BS: oblong/very depressed. CP: uniformiy brown. TL: 25-50136. NO: integumentary glands dorsolateraiiy. CI: Inger (1966). MgupInys: CO: 25. GR: Southeast Asia-China-Philippines-Greater Sundas. EMG: exotroph: lotic: neustonic; endotroph. LTRF: 010. OA: umbelliform: bi-triangular. MP: slight crenulations to none. SUP: scattered throughout oral disc, some arranged radiaiiy to mouth. DEM: absent. NA: nearer snout then eye. VT: medial. EP: lateral. SP: sinistral but well below longitudinal axis. UJ: absent, minor keratinization. LJ: minor keratinization. DF: low, originates near or posterior to dorsal td-body junction, tip pointed. BS: oblong/depressed. CP: beily striped in some. TL: 30-50136. CI: Inger (1985); C.C. Liu and Hu (1961); Y and Fei (1996). e Ophryophryne: CO: 3. GR: Vietnam-China. Tadpole unknown. Scutigev: CO: 24. GR: India-China. EMG: exotroph: lotic: benthic, clasping. LTRF: 5(2-5)/4(14), 5(2-5)/6(1-5), 6(2-6)/6(1-5), 7(2-7)/7(1-6). OA: typical: anteroventrai. MP: narrow dorsal gap: uniserial, dorsal gap absent: uniserial. SUP: few lateray. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral but well below longitudinal axis. UJ: very wde, smooth arc, very coarsely serrate. LJ: wide, wide U-shaped, coarsely serrate. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 60-75136. NO: C.-C. Liu (1950) showed labial teeth on submarginal papdae. CI: C.-C. Liu (1950). Vibrissaphora: CO: 1. GR: southern Cima. EMG: exotroph: lotic. LTRF: 6(2-6)/4(1), 6(2-6), 6(1-5), A-1 exceptionaiiy short. OA: typical: anteroventral. MP: narrow dorsal gap: uniserial. SUP: few ventrolaterally. DEM: absent. NA: nearer snout than eye. VT: -. EP: dorsal. SP: sinistrai but weil below longitudinal axis. UJ: very wide, narrow Ushaped, very coarsely serrate. LJ: very wide, wide U-shaped, coarsely serrate. DF: low, originates near dorsai tail-bodyjunction, tip rounded. BS: oval/depressed. CP: uniformiy dark body, fins with coarse blotches. TL: -. CI: C.-C. Liu and Hu (1961). MICROHYLIDAE: CO: 651327. GR: North and South America, Mrica, India-KoreaAustropapuan region. EMG: exotroph: lentic: suspension feeder, psarnmonic; endotroph. LTRF: 010. OA: keratinized structures absent (except in Otophyne and Scaphiophlyne); any resemblance to an oral disc absent (except in umbelliformMicvohyla, Nelsonuphryne, Otophryne, and Scaphiophryne), oral apparatus usuaiiy involves semicircular to straight-edged oral flaps pendant over terminal mouth, few species with umbelliform oral disc. MP: not applicable except- Nelsonuphyne and some endotrophs. SUP: not applicable except perhaps in Nelsonin upbryne. DEM: not applicable. NA: usuaiiy absent until metamorphosis but positioned nearer snout than eye. VT: medial, dextral. EP: lateral, dorsal. SP: midventral on chest, abdomen or near vent, appears sinistrai in diflerent configurations in Otophryne and Stereocyclops. UJ: usuaiiy absent but Scaphwpbryne has unpigmented jaw sheaths, Otuphryne has hypertrophied serrations without a basal sheath. LJ: as upper jaw. DF: low, originates near dorsal tail-body junction, sometimes with flagellum. BS: round to oval/depressed. CP: dorsum uniform, sometimes with sagittal line, venter sometimes mottled or striped. TL: 15-35/36. Asterophryinae: CO: 8/49. GR: New Guinea. Asteruphrys: CO: 1. GR: New Guinea. EMG: endotroph. Barygenys: CO: 7. GR: New Guinea. EMG: endotroph. Callulops: CO: 13. GR: New Guinea. EMG: endotroph. Hylophorbus: CO: 1. GR: New Guinea. EMG: endotroph. Mantuphryne: CO: 3. GR: New Guinea. EMG: endotroph. Pherohapsis: CO: 1. GR: New Guinea. EMG: endotroph. Xenoba~acbus: CO: 17. GR: New Guinea. EMG: endotroph. Xenurhzna: CO: 6. GR: New Guinea. EMG: endotroph. Brevicipitinae: CO: 5/19. GR: eastern and southern Afiica. Balebrevimps: CO: 1. GR: Ethiopia. EMG: endotroph. Breuiceps: CO: 13. GR: southern f i c a . EMG: endotroph. Callulina: CO: 1. GR: Tanzania. EMG: endotroph. Probreuiceps: CO: 3 . GR: Tanzania-Zimbabwe. EMG: endotroph. Spelaeuphryne: CO: 1. GR: Tanzania. EMG: endotroph. Cophylinae: CO: 7/36. GR: Madagascar. Anuhnthyla: CO: 4. GR: Madagascar. EMG: endotroph.

320

RONALD ALTIG AND ROY W. McDIARMID

Cophyh: CO: 1. GR: Madagascar. EMG: endotroph. Madecmsophryne: CO: 1. GR: Madagascar. EMG: endotroph. Plittypelis: CO: 9. GR: Madagascar. EMG: endotroph. Plethhntohyla: CO: 14. GR: Madagascar. EMG: endotroph. Rhombophryne: CO: 1. GR: Madagascar. EMG: endotroph. StumpffG: CO: 6. GR: Madagascar. EMG: endotroph. Dyscophinae: CO: 219. GR: Madagascar and Southeast Asia. Calluella: CO: 6. GR: Southeast Asia. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: straight-edged upper lip without medial notch above terminal mouth. MP: absent SUP: absent. DEM: not applicable. NA: absent until metamorphosis VT: medial. EP: lateral. SP: midventral below gut UJ: absent LJ: absent DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: pale brown, translucent, distal part of t d dark. TL: 24/31. NO: ventral fin with lobe near body, hangs in midwater at a 45" angle. CI: C.-C. Liu and Hu (1961); M. A. Smith (1917). Dyscophus: CO: 3. GR: Madagascar. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: straight-edged upper lip without medial notch above terminal mouth. MP: absent SUP: absent. DEM: not appiicable. NA: absent until metamorphosis VT: medial. EP: lateral. SP: midventral below gut UJ: absent LJ: absent DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: uniformly pale melanic pigment, sometimes more intense on distal part of t d . TL: 36/36. CI: BlomrnersSchlosser (1975); Blomrners-Schlosser and Blanc (1991); Glaw and Vences (1992); Pintar (1987). Genyophryninae: CO: 8/89. GR: Australia-New Guinea-Philippines. Albmicus: CO: 3. GR: New Guinea. EMG: endotroph. Aphantophryne: CO: 3. GR: New Guinea. EMG: endotroph. Choemphryne: CO: 1. GR: New Guinea. EMG: endotroph. Cophixalus: CO: 28. GR: New Guinea-northeastern Australia. EMG: endotroph. Copiula: CO: 5. GR: New Guinea. EMG: endotroph. Genyophryne: CO: 1. GR: New Guinea. EMG: endotroph. Oreophryne: CO: 24. GR: Philippines-Celebes-New Guinea. EMG: endotroph. Sphenuphryne: CO: 24. GR: New Guinea-northeastern Australia. EMG: endotroph. Melanobatrachinae: CO: 314. GR: India-Tanzania. Hoplophryne: CO: 2. GR: Tanzania. EMG: exotroph: lentic: arboreal, probably spends considerable time out of water. LTRF: 010. OA: absent to transversely straight upper lip above terminal mouth. h W not appiicable. SUP: not applicable. DEM: not appiicable. NA: absent. VT: medial. EP: dorsal. VT: medial. SP: midventral on belly. UJ: absent. LJ: absent. DF: low, originates near dorsal tail-body junaion, tip pointed. BS: oval/ depressed. CP: uniform pale pigmentation. TL: 30136. NO: hgerlike projeaion at anterolateral extent of abdomen. CI: Noble (1929). Melunobatrachus: CO: 1. GR: southern India. EMG: endotroph. Parhoplophryne: CO: 1. GR: Tanzania. EMG: endotroph. Microhylinae: CO: 291106. GR: North and South America and Southeast Asia. Ahhstes: CO: 1. GR: Venezuela. EMG: endotroph. A l ~ ~ i uCO: 1. GR: Peru. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: s: immense, paired, quadrangular oral flaps pendant over terminal mouth. MP: not applicable. SUP: not appiicable. DEM: not appiicable NA: absent. VT: dextrai. EP: lateral, fully visible from below. SP: midventral, near vent. UJ: absent. LJ: absent DF: low, originates near dorsal td-body junction, tip pointed. CP: generdy dark with middorsal stripe. TL: 59/36. CI: Wild (1995). Arcovomw: CO: 1. GR: Brazil. Tadpole unknown. Chaperina: CO: 1. GR: Borneo. EMG: exotroph: lentic: suspension feeder. LTRE 010. OA: s m d semicircular oral flaps with smooth edges pendant over terminal mouth. MP: not applicable SUP: not appiicable. DEM: not appiicable. NA: absent. VT: medial. EP: lateral. SP: midventral on abdomen. UJ: absent. LJ: absent. DF: low, originates near dorsal td-body junction, tip pointed. BS: rounded/depressed. CP: uniformly black. TL: 25/42. CI: Inger (1956, 1966, 1985). Chiawwcletr: CO: 15. GR: Panama-South America. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: s m d semicircular oral flaps with smooth edges pendant over

DIVERSITY

terminal mouth. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral on lower abdomen. UJ: absent. LJ: absent. DF: low, originates at dorsal taii-body junaion, tip pointed with flageiium. BS: rounded/depressed. CP: unifordy pale. TL: 17/36. CI: Schiter and Salas (1991). Cteqhryne: CO: 2. GR: Colombia-Brazii. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: straight-edged upper lip without oral flaps above terminal mouth. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral. UJ: absent. LJ: absent DF: low, originates at dorsal tail-body junaion, tip round. BS: rounded/depressed. CP: unifordy pale. TL: 12/25. CI: Schiter and Salas (1991). Dmypops: CO: 1. GR: Brazii. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: labial flaps reduced without medial notch. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates at dorsal taii-body junction, tip rounded without flageiium. BS: oval/depressed. CP: unifordy dark, beliy spotted and banded. TL: 52/37. CI: Da Cruz and Peixoto (1978). Demzatonotus: CO: 1. GR: Brazil-Paraguay. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: semicircular oral flaps with smooth edges pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral below vent. UJ: absent. LJ: absent. DF: moderately high, originates at dorsal taii-body junction, tip rounded as flageiium. BS: rounded/depressed. CP: unifordy dark. TL: 36/35. NO: tail sheath (see text) present. CI: Cei (1980); Laviiia (1992b); Vizotto (1967). Elachistocleis: CO: 4. GR: Panama-Argentina. EMG: exotroph: lentic: suspension feeder. LTRP: 010. OA: semicircular oral flaps with smooth edges pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal taii-body junaion, tip pointed. BS: rounded/depressed. CP: unifordy pale, tail muscle striped. TL: 15-22/36. CI: J. D. Wdiams and Gudynas (1987). Gascrophryne: CO: 5. GR: United States-Costa Rica. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: semicircular oral flaps with smooth edges pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates ne'ar dorsal tail-body junaion, tip pointed. BS: rounded/depressed. CP: uniform dorsaliy, sagittal stripe, beliy mottled or blotched. TL: 35/36. CI: Donneliy et ai. (1990a); C. E. Nelson and Altig (1972); C. E. Nelson and Cueliar (1968); Orton (1952); A. H. Wright and Wright (1949). Gas~rophrynoides: CO: 1. GR: Borneo. EMG: exotroph: lentic: arboreal (probably sti feeds as a suspension feeder). LTRP: 010. OA: greatiy reduced straight-edged oral flap above terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral on gut. UJ: absent. LJ: absent. DF: low, originates near dorsal taii-body junaion, tip round. BS: rounded/depressed. CP: unifordy dark. TL: -. CI: tentative identification based on Inger (1985:38). Glyphoglosus: CO: 1. GR: Burma-Thailand. EMG: exotroph: lentic: suspension feeder. LTRP: 010. OA: straight-edgedupper lip without oral flaps above terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral below gut. UJ: absent. LJ: absent. DF: low, originates near dorsal td-body junaion, tip pointed. BS: oval/depressed. CP: pale brown, translucent, distal part of tail dark. TL: 40136. NO: M. A. Smith (1917) reports data on the nares as if they were present, hangs in midwater at a 45" angle. CI: M. A. Smith (1917). Hamptophryne: CO: 1. GR: northern and western Arnazon Basin-Guianas-Bolivia. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: semicircular oral flaps with smooth to slightiy crenulate edges pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal taii-body junaion, tip pointed. BS: rounded/depressed. CP: dark, taii striped. TL: 32/36. CI: Dueilman (1978); Schiter and Salas (1991). Hyophryne: CO: 1. GR: Brazii. Tadpole unknown.

Hamptophryne

RONALD ALTIG A N D ROY W. McDIARMID

Microhyla

Nelsonoplnyne

Hypopachus: CO: 2. GR: southern United States-Costa Rica. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: semicircular oral flaps with scalioped-papiilate edges pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal tail-body junaion, tip pointed. BS: rounded/ depressed. CP: dark reticulations, lightiy striped tail. TL: 21/36. NO: tadpoles described by E. H . Taylor (1942) as H . alboventer likely belong to a species of Gastrophyne. CI: C. E. Nelson and Cuellar (1968); Stuart (1941). Kduphynus: CO: 10. GR: southern China-Java-Phdippines. EMG: endotroph Kkloula: CO: 10. GR: Korea-Sri Lanka. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: straight-edged upper lip without oral flaps above terminal mouth. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal tail-bodyjunaion, tip pointed. BS: rounded/depressed. CP: uniform, tail striped, terminal part of tail may be unpigmented. TL: 25-30136. CI: Kirtisinghe (1958); C.-C. Liu (1950). Metaphynella: CO: 2. GR: Malaysia-Borneo. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: dorsoterminal, no oral flaps, 2-3 papiilae at lower jaw. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: dorsal. SP: midventral on chest. UJ: absent. LJ: absent. DF: low, originates slightiy posterior of dorsal tail-body junaion, tip round. BS: oval/depressed. CP: uniforrnly brown. TL: 32/36. NO: Berry (1972) reports open nares from stage 26 onward. CI: Berry (1972). Micyohyh: CO: 24. GR: Sri Lanka-Japan-China-Southeast Asia. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: smali to large semicircular oral flaps with smooth edges pendant over terminal mouth, umbeiliform taxa have oral disc with radialiy arranged submarginal papiilae, and formed almost entirely of lower labium. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral on belly or near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal tail-body junaion, tip pointed with or without a flagellum. BS: round to oval/ depressed. CP: dark, belly sometimes mottle or banded, tail striped. TL: 20-30136. CI: Chou and Lin (1997); Heyer (1971); Inger (1985); Inger and Frogner (1979). Myersielh: CO: 1.GR: eastern Brazil. EMG: endotroph. Nelsonophyne: CO: 2. GR: Costa Rica-western Ecuador. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: labial flaps absent, upper lip derivative forms large lateral flaps, smali lower labium. MP: absent, although some papillae midventrally resemble marginal papiilae. SUP: absent. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral on gut. UJ: absent. LJ: absent. DF: low, originates near dorsal td-body junaion, tip rounded. BS: oval/depressed. CP: uniformly dark. TL: 44/34. CI: Donneily et al. (1990a). Otophlyne: CO: 3. GR: northern South America. EMG: exotroph: lotic: psamrnonic. LTRF: 010. OA: oral disc derivatives reduced to lateral, verticaliy oriented flaps, nascent tooth ridges present below lower jaw but no teeth present. MP: not applicable. SUP: not applicable. DEM: not applicable NA: absent VT: medial. EP: lateral. SP: appears sinistral but starts development midventrally, spiracular tube very long and free from body wall. UJ: isolated, hypertrophied jaw serrations without basal sheath. LJ: similar to upper jaw. DF: low, originates near dorsal tail-body junaion, tip rounded. BS: oblong/ depressed. CP: uniformiy pale brown. TL: 36/36. CI: J. A. Campbeil and Clarke (1988); Pyburn (1980); Wassersug and Pyburn (1987). Phynelh: CO: 1. GR: Malaysia-Sumatra. EMG: endotroph. Ramanelh: CO: 8. GR: India-Sri Lanka. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: smooth-edged, entire, smali oral flap slightiy pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DF: low, originates near dorsal tail-body junction, tip rounded. BS: rounded/depressed. CP: uniform. TL: 27/32. NO: Kirtisinghe (1958) reports open nares. CI: Kirtisinghe (1958). Relictovomer: CO: 1. GR: Panama-Colombia. Tadpole unknown. Stereocyc&ps: CO: 1. GR: Brazil. EMG: exotroph: lentic: suspension feeder. LTRF: 010.

DIVERSITY

OA: semicircular oral flaps with smooth edges pendant over terminal mouth. M E not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral but opens to lefi of the plane of the ventral fin along with vent. UJ: absent. LJ: absent. DE: low, originates near dorsal tail-body junction, tip pointed. BS: round/depressed. CP: uniform (see note). TL: -. NO: I. Grifliths and De Carvalho (1965) discuss polymorphisms of a middorsal stripe and spiracular tube length; tail sheath (see text) present. CI: I. Grifliths and De Carvalho (1965). Synapturanus: CO: 3. GR: Colombia-Guianas-Brazil. EMG: endotroph. Syncope: CO: 2. GR: eastem Ecuador-Peru. EMG: endotroph. Uperodon: CO: 2. GR: India-Sri Lanka. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA. smooth-edged entire (no medial notch) oral flap pendant over terminal mouth. MP: not applicable SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral near vent. UJ: absent. LJ: absent. DE: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: uniform or mottled with striped tail. TI,: 28/36. NO: Bhaduri and Daniel (1956) and Mohanty-Hejrnadi et al. (1979) report open nares. CI: Kirtisinghe (1957). Phrynomerinae: CO: 115. GR: sub-Saharan Africa. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: no keratinized mouthparts, entire (without medial notch) oral flap with smooth edges pendant over terminal mouth. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: rnidventral, below vent. UJ: absent. LJ: absent. DE: low, originates near dorsal tail-body junction, flagellurn present. BS: round/depressed. CP: uniformly pale, tail fins may be darker. TL: 37/36. Phryvwmantzs: CO: 5. CI: Lamotte (1964); Wager (1986). Scaphiophryninae: CO: 2/10. GR: Madagascar Parauxophyh: CO: 1. GR: Madagascar. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: semicircular oral flaps with smooth edges pendant over terminal mouth. MP: absent. SUP: absent. DEM: not applicable. NA: absent. VT: medial. EP: lateral. SP: midventral, on abdomen. UJ: absent. LJ: absent. DE: low, originates near dorsal tail-body junction, tip pointed. BS: round/depressed. CP: dark dorsaily with pale spot on snout and over lungs. TL: 28/36. NO: looks like a microhyline tadpole. CI: BlommersSchlosser and Blanc (1991); Glaw and Vences (1992). Scaphiophryne: CO: 9. GR: Madagascar. EMG: exotroph: lentic: suspension feeder, may rasp some materials, sometimes carnivorous. LTRF: 010. OA: s m d oral disc with papillae and unpigmented, serrated jaw sheaths, labial teeth lacking. MP: complete. SUP: few dorsaily and ventraily. DEM: absent. NA: absent. VT: medial. EP: dorsal. SP: midventrai at center of beiiy. UJ: wide, uniform arc, unpigmented. LJ: wide, open U-shaped, unpigmented. DE: low, originates near dorsal tail-body junction, tip pointed. BS: oval/ depressed. CP: very lightly pigrnented, translucent. TL: 30136. NO: only rnicrohylid tadpole with fdiy formed jaw sheaths (see Otophryne). CI: Blornmers-Schlosser (1975); Blommers-Schlosser and Blanc (1991); Glaw and Vences (1992); Wassersug (1984). MYOBATRACHIDAE: CO: 231118. GR: Australia-New Guinea. EMG: exotroph: lentic: benthic, carnivore, nektonic; lotic: benthic, clasping; endotroph. LTRF: many variations between 010 and 813. OA: typical: anteroventral, ventral. MP: complete, mediurn to wide dorsal gap, narrow to wide ventrai gap. SUP: absent, throughout extent of marginal papiiiae, below P-3, few lateraily. DEM: absent, lateral. NA: nearer eye than snout or reverse. VT: dextral, medial. EP: dorsal. SP: sinistral. UJ: narrow to wide, smooth arc. LJ: narrow to wide, U- to V-shaped. DE: low, originates near dorsal tail-body junction, high and originates anterior to dorsal tail-body junction, tip rounded to pointed. BS: oval/depressed. CP: uniform to variou muted patterns. TL: 15-70136. Lirnnodynastinae: CO: 11151. GR: Australia-New Guinea. Adehtus: CO: 1.GR: eastern Australia. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/ 3(1). OA: typical: anteroventral. MP: mediurn dorsal gap: uniserial. SUP: throughout extend of marginal papillae. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow, wide smooth arc. LJ: narrow, open V-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: uniform. TL: 33/33. CI: Watson and Martin (1973).

Adehtus

324

RONALD ALTIG AND ROY W. McDIARMID

Helezbporus: CO: 6. GR: Australia. EMG: exotroph: lentic: benthic. LTRF: 8(2-8)/3(1), 6(2-6)/3(1), 5(3-5)/3(1), 5(2-5)/3(1). OA: typical: anteroventral. MP: medium dorsal gap: uniseriai. SUP: row below P-3. DEM: lateral. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: udormiy dark. TL: 30-45136. CI: Davies (1991); A. K. Lee (1967). Kyawanus: CO: 3. GR: eastern Australia. EMG: endotroph. Lechriodus: CO: 5. GR: eastern Australia-New Guinea. EMG: exotroph: lentic: carnivore. LTRF: 6(2-6)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: biserial. SUP: throughout extent of marginal papillae. DEM: lateral. NA: nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, wide smooth arc. LJ: medium, open V-shaped. DF: low, originates near dorsal taii-body junction, tip broadly rounded. BS: oval/depressed. CP: uniformly dark. TL: 33/34. CI: Watson and Martin (1973). Limmdynastes: CO: 12. GR: Australia-New Guinea. EMG: exotroph: lentic: benthic; lotic: benthic. LTRF: 6(2-6)/3(1), 5(2-5)/3(1), 4(3-4)/3(1), 4(2-4)/3(1), 2(2)/3(1). OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: few lateraiiy. DEM: lateral. NA: nearer snout than eye or reverse. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsal taiibody junction, tip pointed. BS: oval/depressed. CP: uniform to muted mottling. TL: 35-70136. CI: Tyler et al. (1983). Mgiistobtis: CO: 1. GR: northwestern Australia. EMG: exotroph: lotic: clasping. LTRF: 6(3-6)/3. OA: typical: anteroventral. MP: narrow dorsai gap: uniserial ventraiiy and biserial dorsolateraiiy. SUP: few lateraiiy and below lower tooth rows. DEM: absent. NA: nearer eye than snout. VT: mediai. SP: sinistrai. EP: dorsai. UJ: medium, edge describes smooth arc. LJ: medium, open V-shaped. DF: low, originates near dorsal taiibody junction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 52/42. CI: Tyler et al. (1979). Mkophyes: CO: 6. GR: eastern Australia. EMG: exotroph: lotic: benthic; lentic: benthic. LTRF: 6(2-6)/3(1). OA: typical: anteroventral. MP: complete: uniserial. SUP: few lateraiiy. DEM: lateral. NA: nearer eye than snout. VT: dextral. SP: sinistral. EP: dorsal. UJ: medium, smooth arc. LJ: medium, very wide U-shaped. DF: low, originates near dorsai taii-body junction, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 60-70137. NO: 1-5 accessory rows may be present anterior of lateral ends of lower rows. CI: Davies (1991); Watson and Martin (1973). Neobatrachus: CO: 10. GR: Australia. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/ 3(1). OA: tjpical: anteroventral. MP: dorsal gap: uniserial. SUP: absent. DEM: laterai. NA: nearer eye than snout. VT: dextrai. EP: dorsal. SP: sinistral. UJ: medium to wide, smooth arc. LJ: medium to wide, open V-shaped. DF: low, originates near dorsal taiibody junaion, tip rounded. BS: oval/depressed. CP: uniformiy dark. TL: 40-50136. CI: Davies (1991); Watson and Martin (1973). Notaden: CO: 4. GR: Australia. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/3(1), 2(2)/2(1). OA: typical: anteroventral. MP: wide dorsal and ventral gap: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: uniform. TL: 25/24. CI: E Slater and Main (1963); Tyler et al. (1983). Philoria: CO: 1. GR: northeastern Australia. EMG: endotroph. Rhwbatrachus: CO: 2. GR: northeastern Australia. EMG: endotroph. Myobatrachinae: CO: 12/67. GR: Australia-New Guinea. Arenophlyne: CO: 1. GR: Western Australia. EMG: endotroph. Assa: CO: 1. GR: northeastern Australia. EMG: endotroph. Blyobatrachus: CO: 1. GR: Tanzania. EMG: endotroph. Crinia: CO: 14. GR: Australia-New Guinea. EMG: exotroph: lentic: benthic; lotic: benthic. LTRF: 2(2)/3(1), 2/3(1), 2(2)/2. OA: typicai: anteroventral. MP: wide dorsal gap and narrow to wide ventral gap: uniserial. SUP: absent. DEM: laterai, absent. NA: nearer snout than eye. VT: dextral. EP: dorsai. SP: sinistral. UJ: medium wide smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsai taii-body junction, tip

DIVERSITY

rounded. BS: ovai/depressed. CP: dark, minor mottling. TI,: 15-20136. CI: Littlejohn and Martin (1965); Watson and Martin (1973). Geocrinia:CO: 7. GR: southern Austraiia. EMG: exotrophic: lentic: benthic. LTRF: 010, 2(2)/3[I.]. OA: typicai: anteroventral. MP: wide dorsai and narrow ventrai gaps: biserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc with mediai part straight. LJ: medium, open U-shaped. DF: low, originates near dorsai taii-body junction, tip round. BS: ovai/depressed. CP: uniformiy dark. TL: 22/30. CI: Littlejohn and Martin (1964); Watson and Martin (1973). Met-nia: CO: 1. GR: southwestern Australia. Tadpole unknown. Myobacrmhus: CO: 1. GR: Western Australia. EMG: endotroph. Paracrinia: CO: 1. GR: eastern Australia. EMG: exotroph: lentic: nektonic. LTRF: 2(2)/2. OA: typicai: anteroventrai. MP: medium dorsai gap: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistral. UJ: medium, smooth arc. LJ: narrow, open U-shaped. DF: high, originates anterior to dorsai tail-body junction, tip pointed. BS: oval/depressed. CP: uniformiy paie. TL: 34/34. CI: Watson and Martin (1973). Aeudophryne: CO: 10. GR: eastern Austraiia. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3[:1.]. OA: typicai: anteroventral. MP: wide dorsai gap and narrow ventrai gap: uniserial. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc. LJ: medium, open U-shaped. DF: low, originates near dorsai tail-body junction, tip pointed. BS: ovai/depressed. CP: uniformly dark. TL: 20-25136. CI: Watson and Martin (1973). Spicospina: CO: 1. GR: southwestern Australia. Tadpole unknown. CI: J. D. Roberts et ai. (1997). Taua!&yLus:CO: 6. GR: eastern Australia. EMG: exotroph: lotic: benthic. LTRF: 010, 2(2)/3(1). OA: typical: anteroventral, ventral. MP: complete: uniserial. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistral. UJ: very narrow, smooth arc, coarsely serrate. LJ:very narrow, open U-shaped, coarsely serrate. DF: low, originates near dorsai taii-body junction, tip round. BS: ovai/depressed. CP: uniformly dark. TL: 15-25/36. CI: Retaick and Hero 1998; Watson and Martin (1973). Uperoieia: CO: 23. GR: northeastern Australia-New Guinea. EMG: exotroph: lentic: benthic. LTRF: 1/3,2(2)/3[1]. OA: typical: anteroventral. MP: wide dorsai and ventrai gaps: uniserial. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistral. UJ: wide, smooth arc with slight mediai convexity. LJ: medium, open U-shaped. DF: low, originates near dorsai taii-body junction, tip slightly rounded. BS: ovai/depressed. CP: uniformly dark. TL: 31/36. CI: Tyler et al. (1983). PELOBATIDAE: CO: 3111. GR: Europe-North America. EMG: exotroph: lentic: benthic, carnivore. LTRF: 414 to 616, A-1 usuaily very short. OA: typical: anteroventral, aimost terminal. MP: narrow dorsai gap: uniserial. SUP: absent, few lateraily. DEM: absent. NA: nearer snout then eye. VT: mediai. EP: dorsai. SP: low on lefi side, sometimes aimost midventral. UJ: wide, smooth arc, uniess a carnivore and then with a mediai convexity. LJ:wide, smooth arc, uniess a carnivore and then with a mediai concavity. DF: low, originates near dorsai taiibody junction, tip bluntly pointed. BS: oblong/depressed. CP: uniformiy dark or paie, depending on environmental conditions. TL: 20-80136. NO: cannibal morphotypes of Spea have highly modified keratinized mouthparts and jaw muscuiature. Pelobates: CO: 4. GR: Europe. EMG: exotroph: lentic: benthic. LTRP: 4(24)/5(1-3). OA: typical: anteroventral. MP: narrow dorsai gap: uniserial. SUP: row throughout extent of marginal papillae. DEM: absent. NA: nearer snout then eye. VT: medial. EP: dorsai. SP: low on lefi side. UJ: medium, smooth arc. LJ: wide, open U-shaped. DF: low, originates near dorsai taii-body junction, tip bluntly pointed. BS: ovai/globuiar. CP: muted mottling. TL: 40-60136. CI: Bouienger (1892); Lanza (1983). ScaphiUpus: CO: 3. GR: North America. EMG: exotroph: lentic: benthic. LTRF: 6(4-6)/6(1), 4(34)/4(1-2). OA: typical: anteroventral. MP: narrow dorsai gap: uniserial. SUP: few lateraily. DEM: absent. NA: nearer snout then eye. VT: mediai. EP: dorsai. SP: low on lefi side. UJ: narrow, smooth arc. LJ: medium, open U-shaped, basai part may appear striated. DF: low, originates near dorsai taii-body junction, tip bluntly

scaphiupw

RONALD ALTIG AND ROY W. McDIARMID

pointed. BS: ovai/depressed. CP: uniformly dark. TL: 20-35136. CI: Aitig (1970); A. H. Wright and Wright (1949). Spea: CO: 4. GR: North America. EMG: exotroph: lentic: benthic, facultative carnivore. LTRF: 4(2-4)/4(1-3), 4(3-4)/4(1-2). OA: typical: aimost terminal. MP: narrow dorsai gap: uniseriai. SUP: few lateray, absent. DEM: absent. NA: nearer snout then eye. VT: mediai. EP: dorsai. SP: low on lefi side. UJ: typicai: wide, smooth are; carnivore: mediai convexity. LJ: typical: wide, V-shaped; carnivore: mediai indentation, basai pan- may appear striated. DF: low, originates near dorsai tail-body junction, tip bluntiy pointed. BS: oval/globular. CP: uniformly dark or pde. TL: 40-80136. NO: usuay has keratinized knob on the roof of the buccai cavity, LTRF vanable. CI: Aitig (1970); A. H. Wright and Wright (1949). PELODYTIDAE: CO: 112. GR: southwestern Europe-Caucasus. EMG: exotroph: lentic: benthic. LTRP: 4(34)/5(1-3). OA: typicai: anteroventral. MP: dorsai gap: wide. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: medial. EP: dorsai. SP: sinistral. UJ: narrow, uniform arc. LJ: narrow, wide U-shaped. DF: mediurn, onginates near dorsai tailbody junction. BS: oval/depressed. CP: uniformly dark. TL: 26/36. Pelodytes: CO: 2. CI: Boulenger (1892); Lanza (1983). PIPIDAE: CO: 5/28. GR: South America and sub-Saharan Africa. EMG: exotroph: lentic: suspension feeder, carnivore; endotroph. LTRF: 010. OA: oral disc or keratinized mouthparts absent, slitiike mouth, terminal or nearly so, sometimes with large barbels, highly derived protrusible mouthpan-s of microphagous carnivore. MP: not appiicable. SUP: not appiicable. DEM: not applicable. NA: much nearer snout than eye. VT: mediai. EP: lateral. SP: dual/laterai. UJ: not applicable. LJ: not appiicable. DF: low, originates near dorsal tail-body junction, tip pointed, often with flageiium that undulates continudy. BS: oval/ depressed. CP: uniformly iightiy pigmented, ofien translucent. TL: 30-60136. NO: low ventrai fin extends onto abdomen, ofien as a pronounced lobe, most resemble rhinophrynid tadpoles. Pipinae: CO: 117. G R South America. EMG: exotroph: lentic: suspension feeder; endotroph. LTRF: 010. OA: absent, mouth transverse and siitiike. MP: not applicable. SUP: not appiicable. DEM: not appiicable. NA: near mouth. VT: mediai. EP: lateral. SP: dual/ lateral. UJ: not appiicable. LJ: not applicable. DF: low, originates near dorsai tail-body junction, tip pointed as a flageilum. BS: ovai/depressed. TL: 40-50136. NO: barbels absent, ventral fin extends onto abdomen as a lobe. Pipa: CO: 7. G R Panama-northern South America. CI: Gines (1958); F. Schutte and Ehrl (1987); O. M. Sokol(1977a). Xenopodinae: CO: 4/21. G R sub-Saharan Africa. Hymenochims: CO: 4. G R Cameroon-Zaire. EMG: exotroph: lentic: carnivore. LTRF: 010. OA: absent, protrusible, rounded opening, opens dorsay when at rest. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: nearer mouth than eye. VT: medial. EP: lateral. SP: dual/laterai. UJ: not applicable. LJ: not applicable. DF: low, originates near dorsai tail-body junction, tip bluntiy pointed BS: ovai/depressed. TL: 12-21/34. CI: O. M. Sokol(l959, 1962). Pseu&ymenochims: CO: 1. GR: western Africa. EMG: exotroph: lentic: carnivore. LTRF: 010. OA. protrusible, rounded opening, opens dorsay when at rest. MP: not appiicable. SUP: not applicable. DEM: not applicable. NA: small, nearer snout than eye. VT: medial. EP: lateral. SP: dual/laterai. UJ: not applicable. LJ: not appiicable. DF: low, originates posterior to dorsal tail-body junction, tip rounded. BS: ovai/depressed. CP: uniformly paie. TL: 19/34. NO: I. Griffiths (1963) reported fine "denticles" on the mouthparts. CI: Lamotte (1963). Szlurana: CO: 2. GR: western Africa. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: single barbel at corners of terminal, slitiike mouth. MP: not appiicable. SUP: not applicable. DEM: not applicable. NA: near snout, transversely elliptical. VT: medial. EP: lateral. SP: dual/lateral. UJ: not applicable. LJ: not applicable. DF: very low, originates near or posterior to dorsai tail-body junction, tip pointed as a constantly undulating flagellum. BS: ovai/depressed. CP: uniformly paie. TL: 35-60136. NO: barbel with extrinsic musculature and cartilage support, ventral fin extends onto abdomen as a rounded lobe; floats head-downward in water. CI: O. M. Sokol (1977a).

Peodytes

Hymenocbirus

DIVERSITY

327

Xenopus: CO: 14. GR: sub-Saharan Africa. EMG: exotroph: lentic: suspension feeder. LTRF: 010. OA: single barbel at corners of terminal, slitlike mouth. MP: not applicable. SUP: not applicable. DEM: not applicable. NA: near snout, transverseiy elliptical. VT: medial. EP: lateral. SP: dual/lateral. UJ: not applicable. LJ: not applicable. DF: very low, originates near or posterior to dorsal tail-body junction, tip pointed as a constantly undulating flagellum. BS: oval/depressed. CP: uniformly pale. TL: 35-60136. NO: barbel with extrinsic musculature and cartilage support, ventral fin extends onto abdomen as a rounded lobe; floats head-downward in water. CI: Arnoult and Lamotte (1968); Nieuwkoop and Faber (1967); Rau (1978); Vigny (1979).
PSEUDIDAE: CO: 215. GR: South America. EMG: exotroph: lentic: benthic, nektonic. LTRF: 2/3[1], 2(2)/3. OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: few lateraily, absent. DEM: absent. NA. nearer snout than eye. VT: medial, sinistral. EP: lateral. SP: sinistral. UJ: medium to wide, medial edge arced or straight. LJ: medium, open U-shaped. DF: hgh, originates well anterior of dorsal tail-body junction or low, originates weii anterior of dorsal td-body junction, tip rounded to pointed. BS: oval/compressed. CP: often brightly patterned and prominent ontogenetic change. TL: 34-230136. Lysapsus: CO: 3. GR: South America east of Andes. EMG: exotroph: lentic: benthic. LTRP: 2(2)/3. OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: few lateraliy. DEM: absent. NA: nearer snout than eye. VT: sinistral. EP: lateral. SP: sinistral. UJ: wide, medial edge straight. LJ: medium, open U-shaped. DF: low, originates weli anterior of dorsal tail-body junction, tip pointed. BS: oval/compressed. CP: entire t d boldly banded. TL: 34/37. NO: tail tip black, black bar at base of tad muscle and h, perhaps only case reported of a sinistral vent tube being characteristic of a species. CI: Kehr and Basso (1990). Pseudis: CO: 2. GR: South America. EMG: exotroph: lentic: nektonic. LTRF: 2/3(1). OA: typical: anteroventral. MP: medium dorsal gap: biserial. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: medial. EP: lateral. SP: sinistral. UJ: medium, smooth are. LJ: medium, open U-shaped. DF: high, originates near or anterior to plane of eyes, tip rounded. BS: oval/compressed. CP: young tadpole boldly banded, older ones unicolored dark. TL: 230136. NO: metamorphc atrophy of fins precedes that of tad myotomes. CI: De S and Lavilla (1997); Dixon et al. (1995); Kenny (1969a). RANIDAE: CO: 451635. GR: widespread. EMG: exotroph: lentic: several gudds; lotic: several gudds; endotroph. LTRF: many variations benveen 010 and 918, 213 and n/3 most common. OA: typical: anteroventral, ventral to almost terminal, oral disc reduced and LTRF absent. MP: wide dorsal gap, dorsal and ventral gaps: uni- and biserial. SUP: absent, few lateraily. DEM: lateral, absent. NA: nearer eye than snout or reverse. VT: dextral, medial. EP: dorsal, lateral. SP: sinistral. UJ: narrow to wide, smooth arc, mediai convexity. LJ: narrow to wide, U- to V-shaped. DF: low, originates near dorsal tail-body junctions, very low, originates on tail muscle, tip pointed to round. BS: oval/depressed. CP: mostly muted tones of mottling. TL: 30-100136. Petropedetinae: CO: 13/94. GR: Africa. Anhydrophiyne: CO: 1. GR: South Africa. EMG: endotroph. Archroleptella: CO: 3. GR: South Africa. EMG: endotroph. Archroleptzdes: CO: 2. GR: Tanzania-Kenya. EMG: exotroph: lotic: semiterrestrial. LTRF: 3(1-3)/3(1). OA: typical: ventral. MP: wide dorsal gap: biserial ventrally. SUP: absent. DEM: absent. NA: smail, nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, strongly arched, medial convexity. LJ: similar to upper sheath. DF: very low, originates on tail muscle, tip round. BS: oval/depressed. CP: mottled. TL: 301 36. NO: eyes bulge noticeably, hind legs develop precociously relative to front legs and general development. CI: Drewes et ai. (1989). Cacostemum: CO: 7. GR: southern Africa. EMG: exotroph: lentic: benthic. LTRF: 2(2)/3(1), 3(2-3)/3,4(2-4)/3 (1).OA: typical: anteroventrai. MP: wide dorsal gap: uniserial. SUP: few in row laterally. DEM: lateral, absent. NA: equidistant or nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, medial convexity, coarsely serrate. LJ: wide, open V-shaped. DF: low, originates near dorsal tail-body junction, tip pointed to rounded. BS: oval/depressed. CP: uniformly dark. TL: 28-50136. CI: De Villiers (1929a); Power (1927a); Van Dijk (1966).

Xenopus

Pseudis

RONALD ALTIG AND ROY W. McDIARMID Dimorphognathus: CO: 1. GR: Cameroon-Gabon. Tadpole unknown. Ericabatracbus: CO: 1. GR: Ethopia. EMG: Tadpole unknown. Microbatracbella: CO: 1. GR: South Africa. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/3(1). OA: typical, anterovenual. MP: wide dorsal and narrow venual gaps: uniseriai. SUP: few lateraiiy in a row. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium, mediai convexity. LJ: medium, open U-shaped. DF: low, originates at dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: brown dorsaiiy, white ventraiiy. TL: 25/36. CI: Van Dijk (1966); Wager (1986). Natalobatrachus: CO: 1. GR: South Africa. EMG: exouoph: lotic: benthic. LTRF: 5(2-5)/3. OA: typical, anteroventral. MP: wide dorsal gap: biserial. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinisuai. UJ: wide, prominent medial convexity. LJ: medium, open V-shaped. DF: low, originates at dorsal tailbody junction, tip rounded. BS: oval/depressed. CP: pale gray. TL: 41/36. CI: Van Dijk (1966); Wager (1986). Nothqhlyne: CO: 1. GR: Malawi-Mozambique. Tadpole unknown. Petmpedetes: CO: 7. GR: Cameroon-Sierra Leone. EMG: exotroph: lotic: suctorial, semiterrestrial. LTRF: 3(1-3)/3(1), 3(3)/3(1), 415, 7(5-7)/6(1). OA: typical: venual. MP: wide dorsai gap: uniserial, large folds in lateral part of disc. SUP: two rows below distal most lower tooth row. DEM: lateral. NA: nearer eyes than snout. VT: medial. EP: lateral. SP: sinisual. UJ:wide, smooth arc, broken into two pieces connected by a narrow part. LJ: wide, open V-shaped. DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: uniformly dark. TL: 35/36. NO: upper rows much longer than lower rows in R natatoz CI: Lamotte et al. (1959a); Lamotte and ZuberVogeli (1954b). Phlynobatrachus: CO: 67. GR: sub-Saharan Africa. EMG: exouoph: lentic: benthic. LTRF: 112, 113, 1/4(1), 2(2)/2. OA: typicai: anterovenual. MP: medium dorsal gap: uniserial, ventral papdlae sometimes elongate. SUP: row throughout lower labium. DEM: lateral, absent. NA: nearer snout than eye. VT: mediai. EP: dorsai. SP: sinisual. UJ: wide, prorninent medial convexity. LJ: wide, open U-shaped. DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: dark. TL: 20-351 36. CI: Lamotte and Dzieduszycka (1958); Van Dijk (1966). Phlynodon: CO: 1. GR: Cameroon-Fernando P. EMG: endotroph. Poyntonia: CO: 1. GR: South Africa. EMG: exouoph: lentic: benthic. LTRF: 112 or 1(2)/2. OA: typical. MP: wide dorsal gap UP: lateral patch. DEM: absent. NA: smaii, about equidistant between eye and snout. VT: dexual. EP: dorsal. SP: sinisual. UJ: wide, forms shaiiow arch with a mediai convexity in serrated margin. LJ: V-shaped. DF: low, originates posterior of dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: pale brown with dark speckles. TL: 32/26. NO: skin glands prominent throughout dorsum, margin of vent tube slightly scaiioped. CI: Channing and Boycott (1989). Raninae: CO: 321541. GR: widespread. Amolops: CO: 24. GR: northern India-China-Southeast Asia-Indonesia. EMG: exotroph: lotic: gastromyzophorous. LTRF: 4-8/34, usuaiiy with three complete distai rows on upper labium and a medial gap in first lower row. OA: typical: suctorial. MP: wide dorsal gap. SUP: absent. DEM: lateral. NA: closer to eye than snout. VT: medial. EP: dorsai. SP: sinistral. UJ: medium, wide U-shaped. LJ: medium, wide U-shaped. DF: low, originates near or posterior to dorsai tail-body junction, tip rounded. BS: oval/ depressed. CP: muted mottling. TL: 50-70136. NO: postorbital and posteroventral integumentary glands, some species with keratinized spindes in the skin. CI: Inger (1985); Inger and Gritis (1983). Aubria: CO: 2. GR: Cameroon-Zaire. EMG: exotroph: lentic: benthic. LTRF: 5(3-5)/ 3. OA: typical: anteroventral. MP: wide dorsal gap: mdtiserial lateraiiy and uniserial ventraiiy. SUP: absent. DEM: absent. NA: nearer snout than eye. VT: dextrai. EP: dorsal. SP: sinisual. UJ: medium, smooth arc. LJ: medium, V-shaped. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: totaiiy black. TL: 40136. NO: school in compact balls. CI: Schiotz (1963). Batrachylodes: CO: 8. GR: Solomon Islands. EMG: endotroph. Ceratobatrachus: CO: 1. GR: Solomon Islands. EMG: endotroph.

DIVERSITY

Chaparana: CO: 6. GR: India-Chna. EMG: exotroph: lotic: clasping. LTRE 8(3-8)/ 3. OA: typicai: ventrai. MP: wide dorsai gap: uniserial. SUP: few ventrolaterdy and in a row distai to P-3. DEM: absent. NA. -. VT: dextral. EP: dorsai. SP: sinistral. UJ: medium, wide smooth arc. LJ: wide, V-shaped. DF: low, originates near dorsai taii-body junction. BS: ovai/depressed CP: uniformiy dark. TL: -. CI: C.-C. Liu and H u (1961). Conraua: CO: 6. GR: tropical sub-Saharan Africa. EMG: exotroph: lotic: clasping. LTRF: 7(5-7)/7(1), 8(5-8)/9, 14(7-14)/12(1-2). OA: typicai: ventral. MP: narrow dorsai gap: biserial at least ventrdy. SUP: abundant lateray and distai to posterior rows. DEM: lateral. NA: nearer eye than snout. VT: dextrai. EP: dorsal. SP: sinistrai. UJ: wide, mediai edge straight, exceptionay long processes laterally. LJ: medium, V-shaped. DF: low, originates near dorsai taii-body junction, tip rounded. BS: ovai/depressed. CP: body dorsum spotted, dorsum of taii muscle lightly banded. TL: 45-60134. CI: Lamotte et ai. (1959a); Lamotte and Zuber-Vogeli (1954a); Sabater-Pi (1985). Discodebs: CO: 5. GR: Solomon Islands. EMG: endotroph. Euhyjlossa: CO: 1. GR: Thaiiand. Tadpole unknown. Euphhctis: CO: 4. GR: Arabia-Sri Lanka-Malaysia. EMG: exotroph: lentic: benthic. LTRF: 1/2[1]. OA: typicai: anteroventral. MP: wide dorsal arid medium ventrai gaps: uniseriai. SUP: few laterdy. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsai. SP: sinistral. UJ: medium, edge a smooth arc. LJ: medium, V-shaped. DF: low, originates near dorsai taii-body junction, tip pointed. BS: ovai/depressed. CP: tail and body striped. TL: 35/36. CI: Annandaie and Rao (1918); Kirtisinghe (1957). Hildebrantia: CO: 3. GR: Ethiopia-Mozambique. EMG: exotroph: lentic: benthic. LTRF: 012. OA: typicai: anteroventrd. MP: wide dorsai gap: uniserial, widely spaced. SUP: absent. DEM: absent. NA: equidistant between snout and eye. VT: mediai. EP: dorsai. SP: sinistrai. UJ: wide, massive with mediai convexity. LJ: wide, massive, open U-shaped. DF: low, originates near dorsai taii-body junction, tip pointed and attenuate. BS: ovai/depressed. CP: uniformiy dark. TL: 95/36. CI: Van Dijk (1966); Wager (1986). H~lobatrachus:CO: 5. GR: Angola-Ethiopia-southern Asia-Chia. EMG: exotroph: lentic: carnivore; lotic: carnivore. LTRF: 4(34)/4(1-2), 5(3-5)/5(1). OA: typicai: anteroventral. MP: complete: uniseriai. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextrai. EP: dorsal. SP: sinistrai. UJ: wide, prorninently medidy incised. LJ: wide, prominently medidy incised. DF: low, originates near dorsai taii-body junction. BS: ovai/depressed. CP: uniformly brown. TL: 43/30. NO: M. A. Smith (1917) discussed variations and nonconcordance among sources that suggest other species are involved. CI: Annandaie and Rao (1918); Lamotte and Zuber-Vogeli (1954a); M. A. Smith (1917). Huia: CO: 4. GR: China-southeastern Asia-Greater Sundas. EMG: exotroph: lotic: gastromyzophorous. LTRP: several variations between 518 arid 1116. OA: typicai: suaoriai. MP: usudy wide dorsai gap. SUP: absent. DEM: absent. NA. nearer eye than snout. VT: mediai. EP: dorsai. SP: sinistrai. UJ: wide, M-shaped. LJ: wide, V-shaped. DF: low, originates on taii posterior to dorsai taii-body junction, tip pointed. BS: oval/ depressed CP: dark gray with paie spots, taii dark with paie spots at margins. TL: 681 39. NO: ventrai h originates on tail muscle, transverse keratinized patch on roof of sucker, clusters of glands on body, buccai papiiiae may be distinctive among Amo@, Huia, and Meriscogenys (Inger 1985). CI: Inger (1966); Yang (1991) Indirana: CO: 9. GR: India. EMG: exotroph: lentic: semiterrestriai. LTRP: 5(2-5)/ 3(1). OA: typicai: ventrai. MP: wide (entire upper labium) dorsai gap: uniserial. SUP: none. DEM: absent. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: wide, strongly arched. LJ: strongly arched. DF: low, originates on taii muscle, tip rounded. BS: ovai/depressed. CP: taii muscle banded. TL: 31/33. CI: Annandaie (1918); Dubois (1985). Ingerana: CO: 8. GR: western China-Thaiiand-BornewPhilippines. EMG: endotroph. Lanzarana: CO: 1. GR: Somalia. Tadpole unknown. Lzmnonectes: CO: 62. GR: China-Southeast Asia-Phiiippines. EMG: exotroph: lotic: benthic; lentic: benthic. LTRF: 1/3(1), 1/4(1), 2(2)/3[1]. OA. typical: anteroventrd. MP: wide dorsai gap: papiiiae often with elongate basal areas, narrow ventral gap some-

RONALD ALTIG AND ROY W. McDIARMID

times present. SUP: few lateraiiy. DEM: absent. NA. nearer snout than eye or reverse. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow to mediurn, smooth arc or slight medial convexity. LJ: narrow to medium, open U-shaped. DF: low, originates near dorsal tail-body junction, tip rounded to pointed. BS: ovalJdepressed. CP: variable, un&orm to prorninently spotted/mottied. TL: 2540136. CI: Inger (1954, 1966, 1985). Menjtogenys: CO: 9. GR: Borneo. EMG: exotroph: lotic: gastromyzophorous. LTRF: several variations between 616 and 817. OA: typical: suctorial. MP: complete: uniserial, reduced to crenulations dorsaiiy, lateral papillae may be larger than elsewhere. SUP: few laterdy. DEM: absent, but folds for expansion of the disc present. NA: nearer eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: broadly divided, constructed of few cellular palisades (= ribbed). LJ: ribbed and sometimes broken mediaiiy. DF: low, originates near dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: mottled, ins with lide pigment. TL: 3040/36. NO: ventral in originates on tail muscle, slun glands present (e.g., behind eye, midlaterdy, sometimes on tail); keratinized areas in roof of sucker. CI: Inger (1985); Yang (1991). MicnXalus: CO: 6. GR: India. Tadpole unknown. Nannupbrys: CO: 3. GR: Sri Lanka. EMG: exotroph: lotic: semiterrestrial. LTRF: 2(2)/ 3(1). OA: typical: ventral. MP: wide dorsal gap: uniserial, very s m d ventrdy. SUP: absent. DEM: absent. NA: nearer eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, strongly arched, coarsely serrate. LJ: wide, strongly arched, coarsely serrate. DF: very low, originates well posterior to dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: mottied. TL: 27/38. NO: ventral fin originates on tail muscle, eyes bulge noticeably, hind legs develop precociously relative to front legs and general development. CI: B. T. Clarke (1983); Kirtisinghe (1958). Nannorana: CO: 3. GR: China. EMG: exotroph: lotic: benthic. LTRF: 3(2-3)/3(1). OA: typical: anteroventral. MP: wide dorsal gap: uniserial. SUP: throughout lateral and ventrolateral areas. DEM: lateral. NA: equidistant between snout and eye. VT: sinistral. EP: dorsal. SP: sinistral. UJ: wide, edge a smooth arc. LJ: wide, U-shaped. D E low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed. CP: uniformly dark. TL: 36/36. CI: C.-C. Liu and H u (1961). Nyctibatracbus: CO: 10. GR: India. EMG: exotroph: lotic: benthic. LTRF: 010. OA: typica: anteroventral. MP: complete: uniserial. SUP: scattered throughout disc. DEM: dorsolateral and midventra. NA: nearer to eye than snout. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow, smooth arc. LJ: narrow, open U-shaped. DF: low, originates posterior of dorsal tail-body junction, tip bluntly pointed. BS: oval/depressed. CP: dark mottling, tail muscle banded. TL: 52/36. NO: see Rao (1923). CI: Pillai (1978). Occiqyga: CO: 2. GR: Southeast Asia-Java. EMG: exotroph: lentic: carnivore. LTRF: 010. OA: oral disc and papdiae reduced to fleshy rim, jaw sheaths recessed. MP: dorsal gap in fleshy, nonpapdlate rim, smaii rounded flap middorsdy. SUP: absent. DEM: absent. NA: equidistant between eyes and snout. VT: medial. EP: dorsal. SP: sinistral. UJ: -. LJ: distinctly U-shaped deeply semilunar in shape. DF: anterior portion prominently high, drops abruptly with remainder low, tip pointed. BS: oval/depressed. CP: black stripe through eye, tail variegated. TL: 30-50136. CI: Inger (1985); M. A. Smith (1916b). Paa: CO: 26. GR: India-Southeast Asia-China. EMG: exotroph: lotic: clasping. LTRF: 4(2-4)/3(1), 7(3-7)/3(1). OA: typical: anteroventral. MP: medium dorsai gap: uniserial. SUP: laterdy and below P-3. DEM: absent. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, smooth arc. LJ: wide, smooth arc, open U-shaped. DF: low, originates near dorsal tail-body junction, tip round. BS: oval/depressed. CP: body uniformly dark, tail iighter with flecking. TL: 55/36. CI: Annandaie (1912); C.-C. Liu and H u (1961). Palmatorappia: CO: 1. GR: Solomon Islands. EMG: endotroph. Pbrynoyghssus: CO: 9. GR: Southeast Asia-East Indies. EMG: exotroph: lentic: carnivore. LTRF: 010. OA: oral disc and papiiiae reduced to fleshy rim, jaw sheaths recessed. MP: dorsai gap in fleshy, nonpapillate rim. SUP: absent. DEM: absent. NA: equidistant between eyes and snout. VT: medial. EP: dorsal. SP: sinistral. UJ: -. LJ: deeply semilunar in shape. DF: low, originates posterior to dorsal tail-body junction, tip pointed. BS: oval/depressed. CP: black marks form scdoped appearance along margins of fins.

DIVERSITY

TL: 51/36. NO: snout noticeably pointed, lower lip sometimes visible from above, othenvise similar to Occzhzy~a. Inger (1985); M. A. Smith (1916b). CI: Pl~ztymantis: CO: 39. G R Fiji-New Guinea-Phiiippines-China. EMG: endotroph. Ptychdna: CO: 40. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/2, 3(2)/2, 2(2)/2, 112. OA: typical: anteroventral. MP: medium to wide dorsal gap: uni- and biserial. SUP: few lateraliy, row below distai most lower row. DEM: absent. NA: nearer snout than eye, or reverse. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, medial convexity. LJ: wide, open U-shaped. DF: low with slight arc, originates near dorsal tail-body junction, tip pointed. BS: ovalJdepressed. CP: muted but variable. TL: 30-55136. CI: Lamotte et al. (1958); Lamotte et al. (1959b); Lamotte and Perret (1961a); Van Dijk (1966); Wager (1986). Pyxicephalus: CO: 2. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 5(3-5)/3. OA: typical: anteroventral. MP: wide dorsal gap: multiserial laterally, biserial ventraliy. SUP: few laterally DEM: lateral. NA: near eye than snout. VT: medial. EP: dorsal. SP: sinistral. UJ: wide, medial convexity. LJ: medium, open U-shaped. DF: medium with slight arc, originates near dorsal tail-body junction, tip bluntly pointed. BS: ovalJdepressed. CP: uniformly dark. TL: 70136. NO: schools, sometimes in midwater and with other species. CI: Power (1927a); Van Dijk (1966). Rana: CO: 224. GR: widespread. EMG: exotroph: lentic: benthic; lotic: benthic, clasping, gastromyzophorous; endotroph. LTRF: many variations between 112 and 918,213 is common in North America, n/3 common elsewhere. OA: typical: anteroventral, ventral. MP: dorsal gap: usualiy uniserial, more rarely a narrow ventral gap, ventral papillae may be elongate. SUP: few laterally, absent. DEM: lateral, absent. NA: nearer eye than snout or reverse. VT: dextral. EP: dorsal. SP: sinistral. UJ: narrow to wide, smooth arc, medial part of edge straight. LJ: narrow to wide, narrow V-open U-shaped. DF: low, originates near dorsal tail-body junction, tip pointed to round. BS: ovalJdepressed. CP: variable. TL: 30-100136. NO: a number of species groups have distinctive morphologies. CI: Boulenger (1892); Hilhs (1982); Hillis and Frost (1985); Hdhs and De S (1988); Lanza (1983); Scott and Jemings (1985); A. H. Wright and Wright (1949); Zweifel (1958). Staurois: CO: 3. G R Borneo-Philippines. EMG: exotroph: lotic: burrower. LTRF: 2(2)/9(1). OA: typicai: ventral. MP: wide dorsal and narrow ventral gap: uniserial dorsolaterally, biserial ventraliy. SUP: few dorsolateraliy. DEM: lateral. NA: smali, much nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: narrow, wide arc. LJ: medium, open V-shaped. DF: low, originates near dorsal tail-bodyjunction, tip rounded. BS: ovalJdepressed. CP: very lightly pigmented, skin appears shiny. TL: 32/31. NO: convergent on the general centrolenid morphology, eyes particularly smali. CI: Inger and Tan (1990); Inger and Wassersug (1990). Taylorana: CO: 2. GR: India-Vietnam-Java. EMG: endotroph. Tomoptema: CO: 13. GR: India-sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/3, 4(2-4)/3. OA: typical: anteroventral. MP: wide dorsal gap: biserial, except uniserial midventraiiy. SUP: few lateraliy. DEM: lateral. NA: nearer eye than snout or reverse. VT: dextral. EP: dorsal. SP: sinistral. UJ: wide, massive with medial convexity. LJ: wide, massive, open U-shaped. DF: low, originates at dorsal tail-bodyjunction, tip pointed. BS: ovalJdepressed. CP: uniforrnly dark. TL: 3644136. CI: GriUitsch et al. (1988); Van Dijk (1966); Wager (1986). RHACOPHORIDAE: CO: 121292. G R Africa-India-southeastern Asia. EMG: exotroph: lentic: benthic, arboreal, nektonic; lotic: benthic, clasping, suaorial; endotroph. LTRF: many variations between 110 and 813; n/3 most common. OA: typical: anteroventral or terminal. MP: wide dorsal gap, sometimes narrow ventral gap or complete: uni-, bi-, and multiserial. SUP: absent, lateraliy or below P-3, row(s) centripetal to marginal papiiiae. DEM: lateral, absent. NA: nearer snout than eye or reverse. VT: dextral, medial. EP: dorsal, lateral. SP: sinistral. UJ: variable including wide, medial margin straight or smooth arc coarsely serrate or not, absent. LJ: medium to wide, U- to V-shaped, coarsely serrate or not, sometimes smali medial convexity. DF: low to moderately high, originates near dorsal tail-body junction, tip rounded to pointed. BS: round, oblong or ovalJdepressed. CP: variable. TL: 25-90136. Buergeriinae: CO: 114. G R China-Japan. EMG: exotroph: lentic: benthic; lotic: benthic.

Wchadena

RONALD ALTIG A N D ROY W. McDIARMID

Mantidaqlus; Mantidaqlus sp.

Mantidaqlus; M. 0pqat-m

LTRF: 5(2-5)/34(1), 6(3-6)/4(1). OA: typicai: anteroventrai.MP: wide dorsai gap: uniseriai. SUP: row(s) centripetal to most marginal papillae. DEM: lateral. NA. nearer snout than eye or reverse. VT: dextrai. EP: dorsai. SP: sinistrai. U J:wide, mediai margin straight, coarsely serrate. LJ: medium, V-shaped, coarsely serrate. DF: low, originates near dorsai taii-body junction, tip pointed. BS: oblong/depressed. CP: iightly pigmented with flecks and spots on tad. TL: 30-44136. Buegeria: CO: 4. GR: China-Japan. CI: Maeda and Matsui (1989); Okada (1931); Utsunomiya and Utsunomiya (1983). Mantellinae: CO: 2/68. GR: Madagascar. Mantella: CO: 10. GR: Madagascar. EMG: exotroph: lentic: benthic, arboreai. LTRF: 3(2-3)/3,5(2-5)/3[1]. OA: typicai: anteroventrai.MP: wide dorsai gap: uniseriai dorsolaterdy, biseriai ventrolaterdy. SUP: absent. DEM: lateral. NA: nearer snout than eye. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc. LJ: medium, open V shaped. DF: low, tip rounded. BS: ovai/depressed. CP: uniforrnly dark. TL: 20-30136. CI: Arnoult (1965); Blommers-Schlosser and Blanc (1991); Glaw and Vences (1992). Mantidactylus: CO: 58. GR: Madagascar. EMG: exotroph: lentic: benthic, arboreai; lotic: benthic, psarnmonic, umbeiiiform; endotroph. LTRF: a number of variations between 010 and 613. OA: typicai: anteroventrai, terminal, ventrai. MP: wide dorsai gap: uniseriai, s m d ventrai gap less common, complete. SUP: few laterdy, row below P-3. DEM: lateral, absent. NA: dose to snout then eye and reverse. UJ: absent, narrow to wide, smooth arc, sometimes with mediai convexity, serrations fine to coarse, one species with no basai sheath and serrations elongated into series of spikelike projections. LJ: narrow to wide, V-shaped, spikelike projections. DF: tip pointed, sometimes proximai part low with a prominent increase in height more posteriorly. BS: ovai/depressed. CP: quite variable, often uniformiy dark. TL: 25-70136. NO: M. lugubrzs without labiai teeth but pigmented stniaures in rows on lower labium look iike labiai teeth. CI: Arnoult and Razarihelisoa (1967); Blommers-Schlosser and Blanc (1991); Glaw and Vences (1992). Rhacophorinae: CO: 91220. GR: Africa-Madagascar-Seycheiie Islands-India-southeastern Asia. Aglrptodactylus: CO: 1. GR: Madagascar. EMG: exotroph: lentic: benthic. LTRF: 6(2-6)/3(1). OA: typical: anteroventrai. MP: wide dorsai gap: uniseriai. SUP: absent. DEM: lateral. NA: nearer eye than snout. VT: dextrai. EP: dorsai. SP: sinistral. UJ: wide, mediai edge straight. LJ: wide, U-shaped. DF: low, originates near dorsai taii-body junction, tip round. BS: ovai/slightly depressed. CP: paie with reticulations on taii. TL: 25/36. CI: Blommers-Schlosser and Blanc (1991); Glaw and Vences (1992). Boophis: CO: 37. GR: Madagascar. EMG: exotroph: lotic: benthic, clasping, suctoriai; lentic: benthic. LTRF: 3(2-3)/3, 5(2-5)/3(1), 6(2-6)/3[1], 6(3-6)/3(1), 7(2-7)/3(1), 75-7)/3, 8(3-8)/3,8(5-8)/3. OA: typicai: anteroventrai, ventrai. MP: wide dorsai gap, sometimes narrow ventrai gap, uncommonly complete: uni- and biseriai. SUP: common laterdy and below P-3. DEM: lateral, absent. NA: various sites. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: medium to wide and various shapes, absent. LJ: medium to wide, various shapes. DF: low, originates near dorsai taii-body junction, tip pointed to rounded. BS: ovai/depressed. CP: variable mottled. TL: 30-90136. CI: BlommersSchiosser (1979b); Blommers-Schiosser and Blanc (1991); Glaw and Vences (1992). Chimiafus: CO: 8. GR: Southeast Asia. EMG: exotroph: lentic: arboreai. LTRF: 5(2-5)/3, 4(24)/3(1), 212. OA: typicai: terminal or anteroventrai. MP: wide dorsai gap: multi- or uniseriai midventrdy, narrow ventral gap. SUP: absent. DEM: lateral, absent. NA: nearer snout than eye. VT: dextrai. EP: dorsai. SP: sinistrai. UJ: medium, massive construction, smooth arc. LJ: medium, V-shaped. DF: low, originates at dorsai tail-bodyjunction, tip slightly rounded. BS: ovai/depressed. CP: siightly speckled on taii, uniformiy dark, distal part of tail dark. TL: 30141. NO: C. ezjinderi oophagous. CI: Bourret (1942); Heyer (1971); Kam et ai. (1996); Kurarnoto and Wang (1987); Utsunomiya and Utsunomiya (1983). Chiromantis: CO: 3. GR: sub-Saharan Africa. EMG: exotroph: lentic: benthic. LTRF: 3(2-3)/3, 5(2-5)/3. OA: typicai: anteroventrai. MP: wide dorsai gap, narrow ventrai gap: uniseriai dorsolaterdy and biseriai ventrdy. SUP: absent. DEM: lateral. NA. s m d , nearer snout than eye. VT: mediai. EP: dorsai. SP: sinistrai. UJ: medium, smooth arc. LJ: narrow, V-shaped. DF: low, originates near dorsai taii-body junction, tip pointed.

DIVERSITY

BS: oval/depressed. CP: u d o r d y paie with blotches on taii. TL: 45-55/34. CI: Lamotte and Perret (1963a); Van Dijk (1966). NyctkaLus: CO: 5. GR: Greater Sundas-Philippines. EMG: exotroph: lentic: arboreai. LTRE 5(2-5)/3, 5(3-5)/3. OA: typicai: anteroventrai. MP: wide dorsai gap: biseriai. SUP: absent. DEM: lateral. NA: equidistant between eye and snout, smaii medial projection. VT: medial. EP: dorsal. SP: sinistral. UJ: medial convexity, coarsely serrate. LJ: medlurn, V-shaped, coarsely serrate. DF: low, originates near dorsal mil-body junction, tip rounded. BS: oval/depressed. TL: 43/26. CP: unifordy purplish brown. CI: Alcala and Brown (1982); Inger (1966, 1985). Philautus: CO: 87. GR: India-China-Philippines. EMG: exotroph: lentic: arboreal; endotroph. LTRP: 110. OA: typical: terminal. MP: complete: biserial. SUP: -. DEM: lateral. NA: tiny, nearer snout than eye. VT: medial. EP: dorsal. SP: sinistrai. UJ: wide, without serrations. LJ: wide, coarsely serrate. DF: low, originates near dorsal tail-body junction, tip rounded. BS: oval/depressed, widest near plane of eyes, snout truncate. CP: unifordy paie brown. TL: 29/37. CI: Wassersug et ai. (1981a). Polpedutes: CO: 19. GR: India-China-Japan-southeastern Asia-Phhppines. EMG: exotroph: lentic: benthic. LTRF: 4(24)/3[1.], 5(2-5)/3[:1]. OA: typical: anteroventral. MP: wide dorsal gap: biserial, narrow ventral gap. SUP: few lateraiiy, throughout extent of marginal papillae, absent. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsai, lateral. SP: sinistral. UJ: medium, high smooth arc, medial convexity. LJ: medium, open V-shaped. DF: moderately high, originates near dorsal td-body junaion, tip pointed. BS: oval/depressed. CP: variable. TL: 2545136. NO: Inger (1966) shows tail sheath (see text chap. 3) in l?macrotis and l? otiiuphus. CI: Inger (1985). Rhacuphoms: CO: 51. GR: India-China-Celebes Islands. EMG: exotroph: lentic: benthc, nektonic. LTRF: 4(2-3)/3(1), 5(2-5)/3[1]. OA: typicai: anteroventral. MP: wide dorsal gap: biserial, sometimes s m d ventral gap. SUP: absent, few laterally. DEM: lateral. NA: nearer snout than eye. VT: dextral. EP: dorsal. SP: sinistral. UJ: medium to wide, smooth arc or smaii medial convexity. LJ: medium to wide, V-shaped. DF: low to medium, originates near dorsal taii-body junction, tip pointed. BS: oval/depressed. CP: usudy unicolored, often pale. TL: 3545136. CI: Inger (1966, 1985). Tbelodema: CO: 9. GR: Burma-Malaysia. EMG: exotroph: lentic: arboreal. LTRF: 4(24)/4(1). OA: typical: nearly terminal. MP: wide dorsal gap: biserial. SUP: absent. DEM: lateral. NA: s m d , nearer snout than eye. VT: medial. EP: dorsal. SP: sinistral. UJ: medium, medial convexity, coarsely serrate. LJ: wide, more finely serrate, open Ushaped. DF: medium, erupts abruptly from near dorsal td-body junction, tip broadly rounded. BS: round/depressed. CP: unifordy dark. TL: 28/28. CI: Wassersug et ai. (1981a). RHINODERMATIDAE: CO: 112. GR: southwestern South America. Rhinoakma: CO: 2. GR: southwestern South America. EMG: endotroph. RHINOPHRYNIDAE: CO: 111. GR: Texas-Costa Rica. EMG: exotroph: lentic: suspension feeder. LTRP: 010. OA: terminal, wide, slitlike mouth with multiple barbels. MP: not appiicable. SUP: not applicable. DEM: not applicable. NA: smaii, nearer eye than snout. VT: medial. EP: lateral. SP: dual/lateral. UJ: not applicable. LJ: not appiicable. DF: low, originates near dorsal tail-body junction, tip bluntly pointed. BS: oval/depressed. CP: unifordy pale or dark. TL: 50136. NO: papilla-like structure at syrnphysis of infrarostral cartilages; looks grossly like a %nopus tadpole. Rhin@hrvnus: CO: 1. GR: Texas-Costa Rica. CI: Orton (1943): E. H. Tavlor (1942). SOOGLOSSIDAE: CO: 213. GR: Seycheiie Islands. Nesomantis: CO: 1. GR: Seycheiie Islands. EMG: endotroph. Sooglosus: CO: 2. GR: Seycheiie Islands. EMG: endotroph.

Polrpedates

fiivwpbrynus

334

RONALD ALTIG A N D ROY W. McDIARMID

Alphabetical List of Representative Morphotypes Presented in the Text: Genus, Guild, Species Drawn, and Source (s)
Acanthkai~(Arboreal Type 2); A. spznosus; after Larnotte et al. (1959a) Acris (Benthic);A.~lyllus; USNM 332799 Adelotus (Benthic); b r h ; after Watson and Martin (1973) A. Afnxalus (Nektonic); Afnxalus sp.; composite of USNM 308801-03 Ansonia (Clasping);A. lon&&ita; USNM 306204 Ascaphus (Suctorial);A. wei; USNM 297184 Astylosteemus (Adherent);Astylosteemw sp.; USNM 306967 Atelupus (Gastromyzophorous); A. pachydemzus; USNM 286455 Bombina (Benthic); B. bombina; composite of cultured B. orientalis and after Angel (1946), Boulenger (1892), and Lanza (1983) Boophis (Suctorial); Boophis sp.; USNM field number 107433A B u . (Benthic);B. woodhousiz; USNM 320297 Cardw~lossa(Benthic); C. ~racilis;composite of USNM 306968 and C. sp. USNM 306969 Ceratophlys (Carnivorous Type 1); C. comuta; USNM 299936 Chawphlys (Benthic); C. pierottiz; after Faivovich and Carrizo (1992) Cochranella (Fossorial); C. jranulosa; composite of USNM 29 1045 and after C. uranoscopa of Heyer (1985) Colostethus (Benthic);C. hinztatis; USNM 167515 Colostethus (Neustonic); C. flotatur; composite of USNM 203704-07 Cyclorana (Benthic); C. cultripes; after Watson and Martin (1973) Dendrobates (Arboreal Type 2); D. guinguivittatus; USNM 269075 Discglossus (Benthic);D. pictus; after Boulenger (1892) Duellmanohyla (Adherent);D. salvavida; USNM 304979 Hamptophlyne (Suspension-feeder Type 1); H. boliviana; composite of USNM 298826,299953, and 299957 Hehioporus (Benthic); H. australiacus; after Watson and Martin (1973) Hehophlyne (Suctorial);H. regir; USNM 300543 Hemzsus (Benthic); H. Juttatum; composite of USNM 308795 and308798 Hyla (Suctorial);H. amutta; USNM 346278 Hyla (Arboreal Type 1); H. bromeliacia; USNM 304997 Hyla (Arboreal Type 2) ;H. picadoi; USNM 331374 Hyla (Macrophagous);H. sarayacuensis; USNM 313525 Hymnochirus (Carnivorous Type 2); H. curttpes; USNM 140275 Insuetophrynus (Benthic); I. acarpicus; composite after Diaz and Valencia (1985) and Formas et al. (1980) Kassina (Nektonic); K. senegalensis; composite of USNM 308809 and Wager (1986) Lepzdobatrachus (Carnivorous Type 2); L. Ilanensis; composite of USNM 307185 and Cei (1980) Leptobrachella (Benthic);L.~raczlis; after Inger (1966) Leptobrachium (Benthic);Leptobrachium sp.; composite of L. hasselti and L. ngrops after Inger (1966) and Berry (1972) Leptodactylodon (Neustonic); Leptodactylodon sp.; USNM 306971 Leptodactylus (Benthic);L. melanonutus; USNM 304941 Leptodactylus (Benthic, sometimes carnivorous);L. pentadactylus; USNM 203730 Leptopelis (Benthic); L. natahnsis; composite of USNM 308812 and Wager (1986) Leptophlyne (Benthic);L. borbonica; after Berry (1972) Mdntidaclylus (Benthic); Mantidactylus sp.; USNM field number 107426A Mdntidactylus (Fossorial); M. femoralzs; USNM field nurnber 107404 Mantidactylus (Fossorial); M. lugubmr; USNM field number 107405 Mantidactylus (Neustonic);M. opiparus; USNM field number 11828 Megophlys (Neustonic);Megophlys sp.; USNM 292051 Microhyla (Neustonic); M. heymonsi; composite of USNM 206402-05 NelsonUphlyne (Suspension-feeder Type 1); N. atewimus; after Donneiiy et al. (1990) Otupblyne (Psamrnonic); 0.pyburnz; composite of USNM 304280 and after Wassersug and Pyburn (1987) Pelodpes (Benthic); E punctatus; composite after E. N. Arnold and Burton (1978), Boulenger (1892), Lanza (1983), and Noiiert and Noiiert (1992) Phasmahyla (Neustonic);E ~uttata; USNM 241258 Phyllomedusa (Suspension-rasper); E vazllanti; USNM 321149 Polfiedates (Nektonic); E mgmphala; USNM 3 13968 Pseudis (Nektonic); Eparadoxa; USNM 302522 Ptychadena (Benthic);Ptychadena sp.; composite of USNM 308829 and Wager (1986) RhinUphlynus (Suspension-feederType 2); R. dorsalis; composite of USNM 306925,307148, and 307157 Scaphwphryne (Suspension-feeder but may rasp or feed as a carnivore); Scaphiophlyne sp.; USNM field number 11691 Scaphwps (Benthic); S. couchii; USNM 320302 Scarthyla (Nektonic); S. ostinudactyla; USNM 266112 Schismademza (Benthic);S. carens; USNM 308793 Scinax (Nektonic); S. sugillata; composite of USNM 285280-88 Scinax (Nektonic); S. eimochroa; USNM 330834 Scinm (Nektonic);S. stauffm>composite of USNM 306917 and 306921 Stephopaedes (Arboreal Type 3); S. anotis; after Channing (1978) Thorupa (Semiterrestrial); ?: miliaris; USNM 209372 Trichobatrachus (Suctorial);?: robustus; composite of USNM 306974 and 306976 Werneria (Suctorial); Iilpeussi; reconstructed after dorsal and ventral views in Mertens (1938) Xen~pus (Suspension-feeder Type 2); X. l m s ; USNM 308822

DIVERSITY

335

Summary

Severai interesting things came out of our attempt to characterize the morphologicai variation represented among ali genera and families of anuran tadpoles. The use of a standard African: Sooglossidae, Heleophrynidae, Hemisotidae, Microhylidae, Arthroleptidae, Pipidae, Rhacophoridae, Bufonidae, Ranidae, format and terminology gave us a much better appreciation and ~ ~ ~ e o l i i d a e of the distribution of Iamal traits among living frogs. This Austropapuan:r Ranidae, Hylidae, and Myobatrachidae summation also uncovered the types of gaps in our knowl- Nearctic: Ascaphidae, Rhinophqmdae, Leptodactylidae, Microhyliedge. By our count, no information is available on the reprodae, Pelobatidae, Bufonidae, Ranidae, and Hylidae duaive mode for 29 genera ( 8%) of frogs, and only four Neotropical: Rhinophrynidae, Pseudidae, Pipidae, Ranidae, Microhylidae, Bufonidae, Centrolenidae, Dendrobatidae, Leptodactylidae, of these are reasonable candidates for endotrophy. Most and Hylidae (86%)of these 43 species are in monotypic genera, and a- Oriental: Bombinatoridae, Hylidae, Megophryidae, Microhylidae,Buing this gap has a high probability of uncovering different fonidae, Rhacophoridae, and Ranidae tadpole morphotypes. Although we currently do not know Palearaic: Pelodytidae, Pelobatidae, Hylidae, Bombinatoridae, Discoglossidae, Bufonidae, and Ranidae the total number of different tadpoles that have been described, we know that those of many species in genera from which some tadpoles are known remain to be discovered. Given past findings, we are confident that some of these undiscovered forms also will provide new insights into tadpole if there are unworked larvai coiiections in some European morphology. Whde this summation confirms that we know museums similar to the extensive holduigs of reptiles that something about the reproduaive mode of most frog spe- were amassed during periods of European colonial occucies, the inclusion of data for a given taxon does not imply pancy of tropical Africa. If not, efforts should be made to that comparable data are available for ali species in that genus overcome the logisticai and politico-economic problems that nor that the recorded data necessarily are representative of too often make field work in these areas so difficult and astadpoles of ali included species. Less than a third of ali tad- semble new colleaions that are direaed specificaliy at the poles have been described reasonably well, and the number tadpole fauna. To facilitate tadpole work there and in other of entries that have no data, including endotrophs and exo- regions of the world, we have summarized (table 12.4) the trophs, shows that much work remains. We have a reason- general distributions of anuran families as an aide to their ably good foundation in certain aspects of tadpole morpho- identification and an adjuna to the comparative study of logicai diversity, but there is much to learn. tadpole morphology. The foiiowing key ais0 grew out of the Another spin-off from t h s compilation was a better ap- summation and is intended not so much as an help to identipreciation of the geographical gaps in our knowledge. It fication (which it may be) but more as an iustration of the seems clear to us that of the tropical regions of the world distribution, including notable cases of convergence, of a where frog diversity is high, the tadpole fama of equatorial few major larvai traits among anuran families and subfamilAfrica stands out as one in need of much more study. ies. A taxa are not accounted for in every case, and the key Whether a broader sampling of lamal morphology would could have been compiled in several ways (i.e., the same or shed new light on ranoid phylogeny is debatable, but the different charaaers in different sequences). Even so, the confew glimpses we have had of morphologicai diversity and veyed message in each case would have been the same -taddevelopmental patterns characteristic of some arthroleptid poles in a few families can be keyed easily and those in antadpoles cry out for attention. We are not aware of much other large group of families cannot be separated based on current interest in tadpoles from tropical Africa and wonder these few charaaers.

Table 12.4 World ranges of anuran families with aotrophc tadpoles in mogeographical realms. Families are listed in approximate order of increasing number of species in each realm, and endemic families are in boldface

336

RONALD ALTIG AND ROY W. McDIARMID

An Illustrative Key to Families of Anuran Tadpoles Based on a Few Major Characters

SECTION 1. SPIRACLE A. Dual and lateral, centripetal waii absent B. Single, midventral on chest, abdomen, or near vent, centripetal waii absent except in cases where spiracle is near vent C. Single, low on left side, sometimes alrnost medial, centripetal waii often absent D. Single, sinistral, centripetal wa usuay present, at least as a ridge SECTION 2. BARBELS AND ORAL APPARATUS A. Single, long, motiie, at corners of mouth; oral disc, jaw sheaths, and labial teeth absent B. Multiple, short, nonrnotile, around mouth margin; oral disc, jaw sheaths, and labial teeth absent C. Barbels absent; oral disc and labial teeth absent; jaw sheaths present as few, spikelike projeaions

Seaion 2

Section 3 Seaion 4 Section 5

Pipidae: Pipinae and Xenopodinae Rhinophrymdae Leptodactylidae: Ceratophrymae (part)

SECTION 3. ORAL APPARATUS, NARES, AND SPIRACLE A. Oral apparatus consists of large suctorial disc, large platelike upper jaw sheath, smaii lower sheath, and some rows of biserial labial teeth; nares present; spiracle on chest Ascaphidae B. Oral apparatus consists of semicircular oral flaps overhanging mouth and no jaw sheaths or labial teeth; Microhylidae: Dyscophinae, Microhyiinae (part), nares absent; spiracle on abdomen or near vent Phyrnomerinae, and Scaphiophryninae (part) C. Oral appararus consists of large umbelliform oral disc derived mostly from the lower labium and no jaw sheaths or labial teeth; nares absent; spiracle on abdomen Microhylidae: Microhylinae (also see section 5) SECTION 4. MARGINAL PAPILLAE, TOOTH ROWS, A-1, AND LTRF A. Marginal papiae complete or nearly so; tooth rows biserial; A-1 aimost the same length as A-2; LTRF 213 Bombinatoridae and Discoglossidae B. Marginal papiae complete or with dorsal gap; tooth rows uniserial; A-1 almost the same length as A-2; LTRF 213 Hylidae: Phyiiomedusinae (part) C. Marginal papiae complete or nearly so (i.e., narrow dorsal gap); tooth rows uniserial; A-1 much shorter than A-2; LTRF > 213 Megophryidae (part), Pelobatidae D. Marginal papiiiae complete or with dorsal gap; tooth rows uniserial; A-1 aimost the same length as A-2; LTRF 5 613 Myobatrachidae: Myobatrachmae (part) SECTION 5. ORAL DISC, JAW SHEATHS, LABIAL TEETH, AND BELLY A. Oral disc reduced to rim around mouth, usuay with dorsal gap; jaw sheaths present; labial teeth present or not; Hylidae Hemiphractinae (part), Hylinae (part) and body wall unrnodiied Ranidae: Raninae (part) B. Oral disc present, typical size or larger than normal but Arthroleptidae: Astylosterninae (part), Dendrobatidae umbeliiform (upturned); jaw sheaths present or not; labial (part), Megophryidae (part), and Rhacophoridae: teeth present (uniserial) or not; body wa not modified Mantellmae (part; also see seaion 3) C. Oral disc normal; one or both jaw sheaths absent; labial teeth present; body wa unrnodified Section 6

D IVE RS ITY

D. Oral disc normal; both jaw sheaths present; labial teeth present; body w d unmodhed E. Oral disc normal; both jaw sheaths present; labial teeth present; body w d variously modified

Section 7 Section 8 Heleophrynidae Rhacophoridae: Mantellinae (part) Bufonidae (part), Hylidae: Hylinae and Pelodryadinae, Leptodactylidae: Teimatobiinae (part), Megopwdae (part), Myobatrachidae: Myobatrachinae, Ranidae: Raninae and Petropedetinae Bufonidae (part), Centrolenidae, Dendrobatidae (part), Hemisotidae, Hylidae: Hyhae, Pelodryadinae (part), and Phyiiomedusinae (part), Hyperoliidae, Leptodactylidae: Teimatobiinae (part), Myobatrachidae: Lirnnodynastinae and Myobatrachinae, Pelodyidae, Pseudidae, and Ranidae: Raninae and Petropedetinae (part)

SECTION 6. LTRF BALANCE VALUE A. LTRF imbalanced negatively (3114 = - 11) B. LTRF imbalanced positively (913 = +6) SECTION 7. ORAL DISC AND LTRP A. Oral disc enlarged, often with many marginal papiiiae smaii and closely spaced (i.e., rheophilous), may be complete or not; number of tooth rows 213-17121; lotic habitats B. Oral disc typical with larger, more widely spaced marginal papdiae, usudy with at least a dorsal gap (i.e., not rheophilous); number of tooth rows 2 113; leutic or lotic habitats

SECTION 8. BODY WALL MODIFICATION A. Beiiy modified into aaively operating sucker (i.e., center can be manipulated to form negative pressure between beiiy and substrate = gastromyzophorous) B. Lateral and posterior body waii developed into free flap that cannot be manipulated to form negative pressure C. Lateral body waii with various forms and degrees of lateral sac development D. Fleshy, hoiiow crown surrounding eyes and nares E. Transverse, fleshy flap behind eyes F. Fingerlike projection ventrolaterdy at junction of opercular and abdominal cavities

Bufonidae (part) and Ranidae: Raninae (part) Leptodactylidae: Teimatobiinae (part) Arthroleptidae: Astylosterninae (part) Bufonidae (part) Bufonidae (part) Microhyiidae: Melanobatrachinae can Museum of Natural History and De S. R. J. Richards suppiied a number of pertinent papers on the Austraiian fama at a moment's notice. Over the past thirty years the curators and stafat several museums have loaned specimens, permitted dissections and other speciai preparations, and coiiected or c d e d to our attention material that has noticeably advanced our research. Without the cooperation of these and others our understanding and awareness of tadpole morphology would have been noticeably lessened.

ACKNOWLEDGMENTS
Several persons helped with the preparation of this chapter in different ways. K. Spencer's sketches of many of the distinctive tadpole morphotypes enhanced the diversity presentation immeasurably. The sketch of Leptobrachella was redrawn frorn Inger (1966) and is reprinted by permission of Inger; the sketch of Nelsonophryne was redrawn from Donneiiy et ai. (1990a) and is reprinted courtesy of The Arneri-

The glossary includes terms that may have been deined or restriaed by the terminology suggested in this book or are not commonly found in other sources. Refer to chapters 2 and 3 for further discussion of many morphological terms. Pertinent citations are often presented. Terms within a definition that are defined elsewhere in the glossary are in boldface. Synonymous terms that are accepted in this book are presented in parentheses, and unaccepted ones are presented in brackets. Uniess noted othenvise, d staging notations pertain to Gosner (1960).

GLOSSARY

A-2 gap (Altig 1970) [medial i n t e ~ d space]. Medial gap or in second upper row in species with 2 upper rows; can be expanded to denote any row on either labium (e.g., A-4 gap, P-3 gap); see A-2 gap ratio and labial tooth row formula. A-2 gap ratio (Altig 1970). Length of either section of the actual row of teeth in row A-2 divided by the length of the medial gap between the two parts of the row; number > 1 indicates a gap smalier than the length of a row section and vice versa; can be calculated for any row that is broken medid y on either labium; see A-2 gap and gap. abdominal length (Carr and Altig 1991). Medial, straight line distance from base of vent tube to the spiracular w d ; this measurement is sometimes usem but has a large measurement error because neither landmark is consistentiy de&able. abdominal sucker (beily sudrer) [belly or sucker disc]. See beily sucker, also adhesive giand, belly flap, and gastromyzophorous. accessory labial tooth rows (R. G. Webb and Korky 1977). Short tooth rows oriented verticdy or transversely and often positioned near the lateral ends of tooth rows; occur norm d y in some ranids and pelobatids, infrequentiy in other taxa; this modified LTRF shows four accessory rows between solidi: 3(2)/4/3(1). acosmanura (Starrett 1973). Word synonym for Orton's (1953) Type 4 tadpole; ranoid of O. M. Sokol(1975). adaptive peak. Combination of dele frequencies or morphologies that leads to the locdy maximal fimess of a population. additive genetic variance. That fraction of total phenotypic variation caused by deles that are statisticdy additive in effea; can be aaed upon by natural selection; see nonadditive genetic variance. adherent. An ecomorphological guild that includes those lotic tadpoles that live in flowing water and have s m d , complete marginal papillae, LTRF commoniy 213, and inhabit faster water than tadpoles in the clasping guild, position maintenance via oral disc common to continuously; see table 2.2, figures in chap. 12, adherent, benthic, clasping, fossorial, gastromyzophorous, nektonic, neustonic, psammonic, rheophilous, semiterrestrial, and suctorial. adhesive gland [cement gland, oral or ventral sucker]. Transitory sticky gland of various shapes on the ventral surface of the head of an embryo or hatchling immediately posterior to the stomodeum or mouth; used to stabilize hatchlings before swimming abilities develop; comotations of "sucker" should not be used in this case. afferent nerve o r axon. Neuron that carries nerve im~ulses to the brain (= sensory) or into a brain nucleus; see efferent nerve or axon. aggregation [school]. A general term that describes a group of tadpoles that are congregated in response to some other faaor(s) than social interactions; qualities of a school (e.g., polarized orientation and coordu-iated movements) and the

340

GLOSSARY

reason for the congregation are not implied; see biosocial aggregation, school, and taxic aggregation. alar plate. Sensory portion of the brain stem, oriented vertically and located dorsal to the sulcus limitans along the lateral ventricular wail; see basal plate. allelopathy. Produaion of chemicals into the environment by one individual that results in a reduction in the fitness of other organisrns. aiiochthonous. Originating outside the systern; usually in reference to food rnaterials. amphigyrinid. Describes the condition of having dual, lateral spiracles as in pipids, rhinophrynids, and Lepidobatrachus (leptodactylid), although latter case not hornologous with former two (Ruibal and Thornas 1988); see laevogyrinid, mediogyrinid, paragyrinid, and sinistral. anamniote. Describes the condition of lacking an amnion; characteristic of fishes and amphibians. anlage (n). An ernbryological precursor. antagonistic pleiotropy. Negative genetic correlation between two traits caused by the fixation of pleiotropic genes that d e a both traits in the adaptive direction, leaving those pleiotropic genes that d e c t one trait in the adaptive direction and the other in the rnaladaptive direction to contribute the genetic variation and covariation. anus. See vent and vent tube. aortic arches. Aortic vessels parallel with the posterior visceral arches and connecting the heart with the dorsal aorta; the major vessels c q i n g oxygenated blood. AR-1 (Thibaudeau and Altig 1988). Expansion of the tooth row notation of Altig (1970) to designate the tooth ridge prior to tooth developrnent in developrnental studies; can be expanded to denote any tooth ridge. arboreal. An ecomorphological guild that includes those lentic tadpoles that are rnodified in one of several rnorphological patterns for living and feeding in water-filled phytotelmata or similar arboreal sites; see table 2.2, figures in chap. 12, benthic, carnivorous, macrophagous, nektonic, neustonic, suspension feeder, and suspension rasper. barbel [papia]. Fleshy, filiforrn projection(s) at the corners of the rnouth of pipid (long with internal support and rnusculature) and rhinophrynid (short, no internal support or rnusculature) tadpoles, although structures are not hornologous between the two families (Cannateila and Trueb 1988a); barbel at the syrnphysis of the lower jaw ofRhinophrynus tadpoles is probably not homologous with either of the others. basal plate. Motor portion of the brain stem oriented h o r i i n tally and located ventral to the sulcus limitans; see alar plate. behavioral fever. Elevation of the preferred body temperature of an ectotherm in response to infection or inoculation with pathogenic bacteria, vimes, or products of these organisms. behavioral thermoregulation. Regulation of body temperature (usually by an ectotherm) by rnovernent or changes of posture or position within or between therrnal regimes. belly flap. Out-folding of lateral, posterior, or both parts of the body wall to forrn a flap; probably used in attachment by semiterrestrial tadpoles but only by increasing surface area for cohesion; see table 2.2, figures in chap. 12, belly sucker, and gastromyzmphorous. beily sucker (abdominal sucker) [beily or sucker disc]. The rnodified belly of a gastromyzophorous tadpole, includes a raised rim, rnusculature to raise the roof of the sucker to actuaily forrn a negative pressure, and papillate and sometimes keratinized areas on the roof; see table 2.2, figures in chap. 12, belly flap, rheophilous, and suctorial.

benthic [benthonic]. An ecomorphological guild that includes those lentic or lotic tadpoles that rasp food from submerged surfaces with keratinized mouthparts; mostiy at or near the bottom, pools and backwaters in lotic sites, body depressed, eyes dorsal, fins low with rounded or slightiy pointed tip, dorsal fin originates at or near the dorsal tailbody junction; see table 2.2, figures in chap. 12, adherent, arboreal, carnivorous, clasping, fossorial, gastromyzmphorous, macrophagous, nektonic, neustonic, psammonic, rheophilous, semiterrestrial, suctorial, suspension feeder, and suspension rasper. bicolored. Two colored, usually in reference to a lateral view of a tail rnuscle that is dark dorsaily and pale ventrally. biosocial aggregation. A general term that describes a group of tadpoles that are congregated because of some social interaction, although qualities of a school (i.e., polarized orientation and coordinated movernents) are not implied; see aggregation, school, and taxic aggregation. birth. (1) Release of some forrn of an immature individual fiom the reproductive traa of its mother, including viviparous and ovoviviparous arnphibians. (2) By extension, used for the analogous release of a froglet frorn nonoviducal sites (e.g.,Assa, Gdctrotheca, Rheobatrachus, and Rhivwdemza) in a parent's body; see hatch. biserial. Describes structures occurring in two series at a site, as two rows of labial teeth per tooth ridge (e.g., some tooth rows in Ascaphw and ail tooth rows in bornbinatorids and discoglossids); see multiserial and uniserial. blotch. Contrasting, usuaily dark, irregular marks of various sizes and shapes with indistinct margins; larger and more irregular than a dot or spot; see mottled, punctate, and steilate. body. (1) [head-body] That part of a tadpole minus the t* a, defined for rneasurernent as the straight iine distance frorn the tip of the snout posteriorly to the body terminus; see body length. (2) Poorly distinguishable part of a labial tooth (Gosner 1959) between the head (working surface) and sheath (embedded in tooth ridge). body axis. Irnaginary iine in lateral view that extends from the tip of the snout to the body terminus; used as a reference for describing the relative locations of stmctures; see tail axis. body length [head-bodylength]. Straight iine, lateral measurernent frorn the tip of snout to the body terminus; see fig. 3.1 and tail length. body ratio (Altig 1970). Total length divided by body length; see tail length and tail ratio. body terminus. Intersection of the posterior body wail and the axis of the tail myotornes; most accurately describes body length and tail length arnong diverse taxa in contrast to any rneasurernent involving the vent tube; see chap. 3. branchial arch(es). The three gill-supportingvisceral arches. branchial food trap (Wassersug 1976a, b). Glandular region between the ventral velurn and Mter plates of larval Types 3 and 4; see chap. 5. buccal apparatus [oral apparatus]. AU structuresin the buccopharyngeal cavity taken as a unit; does not refer to any stmctures of the mouthparts. buccal papia(e). Collective terrn for ail papillae in the buccal cavity; see buccai apparatus. buccopharynx, anterior and posterior. Collective term for the buccal cavity (anterior) + pharynx (posterior). carnivorous. An ecomorphologicai guild that includes those lentic tadpoles that feed on rnacroinvertebratesand conspe-

GLOSSARY cific and heterospecific tadpoles, either rasping the prey apart or e n g u h g thern intact; keratinized mouthparts present and usuaiiy rnodified; does not include opportunistic scavenging of dead organisrns; see tables 2.2 and 10.2, figures in chap. 12, arboreai, benthic, macrophagous, nektonic, neustonic, suspension feeder, and suspension rasDer. cauda equina. The spinal nerves that exit the spinal cord near the sacrum and then continue posteriorly as a series of fibers. choana(e) (internal nares). Opening of the narial canal in the roof of the buccopharynx. chromatophore. Pigrnent-containing celis derived frorn neuroectoderm; pigrnents contained in specific organelies; see blotch, dot, melanin, melanophore, mottied, punctate, spot, and steiiate. Eiliary cushion (Wassersug 1976a, b; chap. 5). In larval Types 3 and 4, at the dorsolateral pharynx, structures hanging between the ilter plates and consisting of ciiiary celis and goblet cells originating frorn the esophagus; structuraiiy like the ciliary grooves and in close contact with thern; produces and transports rnucus with entrapped food particles to the esophagus. ciliary groove (Wassersug 1976a, b; chap. 5). A horizontal groove at the rnargin of the pharyngeal roof frorn the anterolateral corner of the ventral velum and eventuaiiy into the esophagus; positioned laterdy in pipids and more posteriorly in larval Types 3 and 4; histology and function resembles the ciliary cushions. dasping. h ecomorphological guiid that includes those lotic tadpoles that iive in flowing water and have marginal papiilae with an anterior gap, LTRs comrnonly 5 but as numerous as 818 and usuaiiy with anterior rows more numerou than lower rows (e.g., 913); inhabit rnedium to slow currents, position rnaintenance via the oral disc minor; see table 2.2 and figures in chap. 12, adherent, benthic, fossoriai, gastromyzophorous, nektonic, neustonic, psammonic, rheophilous, semiterrestriai, and suctorial. dearance time. Period of time between ingestion of food and elimination of feces derived frorn that food. cold hardening. Accomrnodation to a brief exposure of low ternperature near the lethal limit such that the individual is more resistant to subsequent exposures to such ternperatures; see heat hardening. complex life cycle. A life cycle in amphibians characterized by two forrns separated by an abrupt shifi (= metamorphosis) in behavior, ecology, rnorphology, and physiology. compressed [laterdy cornpressed, depressed, or flattened]. Describes an imaginary cross-sectional view of a structure that is higher than wide; usuaiiy used in reference to body shape; see cyiindrical, depressed, fusiform, and globular. concentration refuge. That concentration or density of food below which a tadpole ceases to feed and thus the rernaining prey survive and reproduce. wntraiaterai. Occurring on the opposite side of the body; see ipsilateral. wnversion efficiency. Ratio of rnass of food ingested to body rnass increase. craniai nerve. One of about 14 nerves connected to the telencephalon (CNs O-I), midbrain (CNs 11-IV), and hindbrain (CNs V-XII); absolute nurnber of cranial nerves is controversial; see spinai nerve. crenulate. Describes an edge with a wavy margin as would be produced by a series of weakly differentiated papiae; see emarginate and incised.

34 1

critical oxygen tension (P,). PO, below which the rate of O, consurnption of a rnetaboiic 0, regulator becornes dependent upon ambient PO,. critical thermal maximum (CT,,). Upper lirnit of the ternperature tolerance range at which locornotor activity becornes disorganized and the animal loses its abiiity to escape frorn conditions that wi prornptly lead to its death; see critical thermai minimum. critical thermai minimum ( a - ) . Lower limit of the ternperature tolerance range at which locornotor activity becornes disorganized and the animal loses its abiiity to escape frorn conditions that wi prornptly lead to its death; see critical thermai maximum. cusp. Projection(s) of various size, shape, number, and onentation that occur on the working end (= head) of a labiai tooth; does not refer to serrations on jaw sheaths; see incised. . -- . . cuspate. A terrn that has been used to describe jaw sheaths but should be abandoned (see incised) because of confusion with cusps of labiai teeth. cyiindricai. Describes an imaginary cross-sectional view of a structure that approaches equidirnensionality, therefore quasi-circular, neither depressed nor cornpressed; usudy used in reference to body shape; see compressed, depressed, fusiform, and globular. degree-day. Measure of cumulative potential biological activity, usudy a surn of daily rnean ternperatures. denticle. Unacceptable terrn that should be discarded in referente to the labiai teeth of tadpoles. depressed [dorsoventrdy cornpressed or dorsoventrdy flattened]. Describes an irnaginary cross-sectional view of a structure that is wider than high; usudy used in reference to body shape; see compressed, cyiindrical, fusiform, and globular. detritus. Dead organic rnatter. developmentai rate (DR). Change in forrn or stage through time. dextrai. To the right. Refers to a vent tube that opens to the right of the plane of the ventral h or tail rnuscle; ernergence frorn abdornen rnay be sagittal or not; see mediai and sinistrai. diaphragm (peribranchiai w d ) . The waii or septum separating the viscera frorn the buccopharynx of a tadpole. dserentiation rate. Generaiiy, the inverse of time to develop between two given stages. direct development. (1)Developmental rnode of a frog that results in a froglet that developed without a tadpole rnorphotype frorn an egg not intimately associated with a parent's body. (2) h ecomorphological guild that includes species that have direct development; see table 2.2 and figures in chap. 12; see endotrophy, exotrophy, exoviviparous, nidicolous, ovoviviparous, paraviviparous, and viviparous. dorsal. (1)Describes a category of eye position in which no part of the eye is induded in a dorsal silhouette of the tadpole (i.e., the entire eye is positioned medial to the lateral surface of the body). (2) h anatomical descriptor rneaning "toward top or dorsum" relative to a stated landrnark; see lateral. dorsai body terminus: See body terminus and tail-body junction. dot. Smaii, rounded, dark rnarking with discrete edges; see blotch, mottied, and spot. ecological species. A population of a species that is ewlogicaliy

342

GLOSSARY

Merent from the rest of the population; occupies a separate niche. ecomorphological guild. A categorization of a group of tadpoles from several taxa that share common morphological features that collectively suggest some sort of commonality in ecology; tadpoles with similar ecologies have similar morphologies regardless of their taxonomic relationships. ectodermal-endodermal transition zone. Zone where the epiderm from the gills transition zone overlays the pharyngeal endoderm of the anlagen of filter plates. ectotherm [cold-blooded, poikilotherm]. An organism that cannot retain and regulate metabolic heat, therefore, barring behavioral thennoregulation, the body temperature varies with the ambient temperature; see behavioral fever and behavioral thernioregulation. efferent nerve or axon. Neuron that projects away from (= motor) the brain or a brain nucleus; see afferent nerve or axon. elygium (Van Dijk 1966). A pigmented zone at the basal margin of the iris (= ocular elygium) or at the skin-cornea margin (= epidermal elygium); presumably shield the eye from excessive light; see u m b r a c u l u . emarginate [indented, Solded, folded]. Describes a margin of the oral disc (most commonly lateral) with a real indentation(~), one caused by folnot or bending; distinction should be noted between emarginations, folds (= overlapping areas) and tucks (= marginal retractions caused by tensions of connective tissue and labial musculature). embryo. An individual in any stage of development from fertilization until hatching; see froglet, hatchling, immature, juvenile, larva, metamorph, and tadpole. endotroph, endotrophy (Altig and Johnston 1989). General developmental mode that results in an embryo or larva which derives its developmental energy either entirely from vitellogenic yolk or other parentally produced material, sometimes is a nonfeeding tadpole; see table 2.2, direct development, exotrophy, exoviviparous, nidicolous, oviviparous, paraviviparous, and viviparous. exotroph, exotrophy (Altig and Johnston 1989). General developmental mode which results in a larva (tadpole) that feeds on various materials not derived from a parent or trophic eggs provided by subsequent ovulations of the parent, a feeding tadpole; see table 2.2, direct development, endotrophy, exoviviparous, nidicolous, ovoviviparous, paraviviparous, ar~d viviparous. exoviviparous (Altig and Johnston 1989). (1) Developnlental mode of some frogs in which a hatched individual moves from where the eggs were oviposited to another site (often associated with male parent) from which it is eventually "birthed." (2) An ecomorphological guild that includes species that have some form of exoviviparous development; see table 2.2 and figures in chap. 12; see direct development, endotmphy, exotrophy, nidicolous, ovoviviparous, paraviviparous, and viviparous. explosive breeding. All animals in a local population reproduce within a few days. fictive swimming. Involuntary swimming induced in laboratory animals in order to study kinematic interactions (e.g., coordination of muscle groups or the cycle of firing of motor neurons). Miform. Describes a long, narrow object (i.e., filament-like in form). (1) Often a projection that usually has uniform surfaces (e.g., shape of many buccal papillae). (2) Describes the shape of some melanophores.

fin [caudal or tail crest]. Unsupported flaps of epidermiscovered, loose connective tissue usually extending the length of the dorsal and ventral edge of the tail muscle. final thermal preferendum (FIT).Temperature ultimately selected by an individual regardless of prior thermal experience; usually measured within 24-96 hr after placement in a thermal gradient. flagellum (Altig and Johnston 1989) [filament]. Elongate posterior part of the tail that is often nonpigmented and ma!. move independently of the remainder of the tail. fluctuating selection. Adaptive value of a trait varies as a h c tion of varying abiotic and biotic conditions. froglet. A small, sexually immature frog produced by either metamorphosis of an exotrophic tadpole or by direct development. Metamorph is not applicable in the case of direct developing forms; see embryo, hatchling, immature, juvenile, larva, metamorphic, metamorphic climax, premetamorphosis, and prometamorphosis. fossorial. An ecomorphologicai guild that includes those lotic, fusiform tadpoles that have marginal papillae with an anterior gap, LTKF 2/3 or less, and inhabit leaf mats in slow water areas; position maintenance via oral disc absent; see table 2.2, figures in chap. 12, adherent, benthic, datping, semiterrestrial, g~stromyzophorous, nektomk, neustonic, psamrnonic, rheophilous, and suctorial. foliose. ~ o l i a i e or moss-like or $hape of individual or groups of chromatophores. fracture plane. Suture-Zike area in the body of a labial tooth where the head can break off of the sheath. The next replacement tooth (see tooth series) protrudes through the sheath of the old tooth. friction surface. Area that may be Gnely papillate, keratinized, or both on the roof of the befly sucker of some gastromyzophorous tadpoles. functional constraint. Limit to evolution by natural selection caused by physical or mechanical laws. funnel mouth. Has been used in reference to the upturned oral disc of some surface feeders (= umbelliform) tadpoles and the large, pendant (most easily visible in preserved specimens) oral discs of some lotic species; these usages should be abandoned to enhance clarification among the mouthparts, oral disc, and mouth, also see suctorial. fusiform [vermiform]. Describes the body shape of fossorial tadpoles, extremely depressed with dorsal eyes, long tail, and low furs; see compressed, cylindrical, depressed, and globular. ganglion. Enlargement in a nerve outside the central nervous system where pre- and postganglionic axons synapse or cell bodies of sensory nerves are located; see nucleus. gap. Literally a break or discontinuityin a linear structure; used in reference to breaks, usually medial, in a tooth row; see A-2 gap and A-2 gap ratio. gap ratio. See A-2 gap ratio. gape limited. Describes a predator that cannot eat prey above a maximum size set by the size of ~ t mouth; see size refuge. s gas exchange partitioning. Partitioning of 0, uptake and CO, elimination among available external gas exchange surfaces (e.g., lungs, gills, skin, etc.) or between media (air and water). gastromyzophorous. (1) Describes a tadpole that has the belly modified into an actual sucker (= forms negative pressure), like tadpoles of Amlops and Atelopus but not Tboropa and Cyclor~bus, position maintenance in fast and perhaps for turbulent water. An ecomorphological guild that includes those lotic tadpoles that have such a modified belly; see table

GLOSSARY

343

2.2, figures in chap. 12, adherent, beiiy flap, beiiy sucker, benthic, clasping, fossorial, nektonic, neustonic, psammonic, rheophiious, semiterrestrial, and suctorial. genetic correlation. Degree to which a gene(s) sirnultaneously influentes two or more traits. germ plasm. Hereditary material transmitted to the offspring through the germ ceils (as opposed to the inheritance of acquired characteristics); in anuran germ h e usually refers to a conspicuous juxtanuclear organeile aggregate consisting mostiv of mitochondria. giia. Supportiveceils of the nervous system, some ofwhich may provide nutrition to neurons or guide migrating neurons during development; may be 10-50 times more giial ceils than neurites in the brain, astrocytes, oligodendrocytes (form m y e h sheaths), and ependymal ceils (line central canal of brain). giobular. The body shape common to many tadpoles, rounded and more or less spherical or eiliptical; see table 2.2, figures in chap. 12, compressed, cylindrical, depressed, and fusiform. giomus. A number of glomeruli grouped together. growth rate (GR). Change in some measure of size over time. hatch. In amphibians, to escape from the jeily membranes that surround an amphibian embryo; in most cases, a hatchling subsequentiy develops into a tadpole, but in some endotrophs, a hatchling actively moves to another site in or on a parent either temporarily or until released (e.g., Asa; see birth) after further development, or the parent actively relocates the hatchling (e.g., Rhinoakma). In viviparous and ovoviviparous forms, hatching occurs in the oviduct. hatchling. Individual within a series of stages between hatching and a tadpole (stage 25); usually used in a generic sense to distinguish individuals in these ecologically unique developmental stages from an embryo or a tadpole; does not refer to a metamorph; see juveniie, hatch, and larva. head. (1) Cranial part of tadpole body from the snout to the pehbranchial wall or diaphragm. (2) The working surface of a labial tooth (Gosner 1959) that often bears cusps; see body and sheath. heat hardening. Accommodation to a brief exposure of high temperature near the lethal limit such that the organism is more resistant to subsequent exposures to such temperatura; see cold hardening. immature. A free-livingindividual prior to attainment of sexual maturity regardless of size or developmental stage; see embryo, froglet, hatchling, juveniie, larva, metamorph, and tadpole. incised [cuspate]. A descriptor of the shape of the cutting edge of jaw sheaths; with one or more pronounced convexities in the upper sheath and sometimes matched by a concavity in the lower sheath, not forming a uniform arc; see cusp, labial teeth, and serration. inclusive fimess. Genes contributed to the next generation by an individual indirectly by helping nondescendent kin, in effect creating relatives that would not have existed without the help of the individual. infralabial prominence (Thibaudeau and Aitig 1988). The Uor V-shaped medial protrusion of the lower jaw typical of many microhylid tadpoles. internarial distance. Transverse distance between centers of the external narial openings; see chap. 3. interorbital distance [interpupiary or interocular distance]. Transverse distance between the centers of the pupiis of the eyes; see chap. 3.

ipsilateral. Occurring on the same side of the body; see contralateral. jaws. Loose but inclusive term for both jaw sheaths and their supportive infrarostral and suprarostal milages; see chaps. 3 and 4. jaw sheath (Altig 1970) [beak, maxilla (for upper jaw sheath), horny beak or jaws, labial mandible (for either lower or both jaw sheaths), mandible (for lower jaw sheath), suprarostradont (on suprarostral cartilage), infrarostradont (on infrarostral cartiiage)]. Usually serrate, keratinized sheaths that overlie the infrarostral (lower; articulates posteriorly with Meckel's cartiiage) and suprarostral (upper, articulates or attached to chondrocranium) cartiiages that serve as cutting or abrasive feeding surfaces. Used as a separate term from "beak" because it is unlikely these struaures are homologs of other structures termed beaks in birds and turtles; these structures overiie entirely different supportive elements (i.e., not dermal bones); see chap. 3. juvenile. Postmetamorphic frog up to the time of attainment of sexual maturity; see embryo, froglet, hatchling, immature, larva, metamorph, ahd tadpole. juxtagiomerular apparatus. A unit within the kidney composed of the macula densa (= part of the dista1 tubule segment), the lacis ceils (= extraglomeruiar mesangium), and the epithelioid ceils (= granular ceils around the vas afferem). Kugelzelle. "ball cell" or "uniceilular gland" (chap. 5). Kupffer ceiis (Sternzellen). Phagocytic ceiis of the reticuloendothelial system between the hepatocytes. labia. see labium. labial flap (Altig 1970; oral flap) [oral apron]. Fleshy flaps of various shapes and sizes that overhang the mouth in microhyiid tadpoles; all keratinized structures absent; derivatives of upper labium (Thibaudeauand Altig 1988) of typical tadpoles; edges may be uniform or scalloped. labial papilla (oral papilla). Coilective term for reference to any marginal or submarginal papillae (but not buccal papillae) of the oral disc. labial teeth (Altig 1970) [denticle, dermal, or labial denticle, keratodont, labial odontoid]. Keratinized tooth derived from ceiis produced in a mitotic zone in the base of the tooth ridge; composed of a head of various shapes and usually with cusps, body, and sheath, see chap. 3. labial tooth row (LTR).Transverse h e a r array of labial teeth embedded in a tooth ridge; the erupted tooth usually has several replacement teeth, number, lengths, and distribution of rows vary on both labia; see labial tooth row formula. labial tooth row formula (LTRF)(Aitig1970). Fractional notation for designating the number of tooth rows on the upper (numerator) and lower (denominator) labia and the disposition of rows with medial gaps; rows on the upper (anterior, superior) labium are numbered from the lip margin toward the mouth (distal-to-proximal) and rows on the lower (posterior, inferior) labium are numbered from the mouth posteriorly (proximal-to-distal).Numbers in parentheses indicate rows with medial gaps, and numbers in brackets indicate variation in the presence of a medial gap. A formula of 5(2-5)/3[:1.] denotes 5 anterior rows (rows 2 through 5 have medial gaps) and 3 posterior rows [row 1 may or may not have a medial gap]; see accessory rows, labial tooth, and labial tooth row. labium [lip]. Fleshy, disc-shaped structures that more or less surround the mouth; usually with labial teeth arranged in

344

GLOSSARY

transverse rows; margins mostly free and papiilate; upper and lower labia together form the oral disc; see labial teeth, labial tooth row, oral apparatus, and oral disc. laevogyrinid (sinistral). Describes a tadpole having a single spiracle on the left side; see amphigyrinid, mediogyrinid, and paragyrinid. larva. Postembryonic, hatched, immature form of a normaiiy metamorphoSing (obligate process in most anurans) species-; a larva of an anuran is specificaiiya tadpole; lawiform adults (at least partial larviform morphology but sexually mature) occur in salamanders but not in anurans; see embryo, hatchling, juvenile, and metamorph. larval transport. Parental transport over various periods of time that re-locates exotrophic larvae from the site of oviposition of terrestrial eggs to some aquatic site; larvae usuaiiy on the back of the parent; does not include endotrophs. lateral. (1) Describes a category of eye position in which some part of the eye is included in a dorsal sihouette of the tadpole (i.e., the eyes protrude further lateraliy than the surrounding body surface). (2) An anatomical descriptor meaning "toward the side" relative to a stated landmark; ser dorsal. lateral line organ (neuromast). See neuromast. lateral l h e system. Ali the pressure-sensitive neuromasts distributed over the body in specific patterns, usuaiiy in lines; see lateral line organ and receptor unit. lateral process (Aitig 1970). Poorly defined lateral portion of the upper jaw sheath beyond which serrations are small to absent; probably not a working surface for feeding. lemmanura (Starrett 1973). Word synonym of Orton's (1953) Type 3 tadpole; discoglossoid of O. M. Sokol (1975). lentic. Limnological description of any nonflowing water system; see lotic. iimbic system. Portion of forebrain, including the amygdala, habenda, hippocampus, and preoptic areas, that may form the highest correlative center in the brain. limiting resource. Resource which is in criticaiiy short supply; an increase in the availability of a limiting resource shodd increase the size of a popdation. lingual papilia (Wassersug 1980). Specifically those buccai papiilae of various numbers, shapes, and sizes that rest on the lingual anlage. lotic. Limnological description of any flowing water system; see lentic. LTR. Abbreviation for labial tooth row. LTRF.Abbreviation for labial tooth row formula. lymphatic system. Lyrnph glands and vessels that function in imrnune responses. macrophagous. An ecomorphologicai guiid that includes those lentic tadpoles that presumably feed by taking larger bites (compared with smaiier particles generated by rasping tadpoles) of attached materiais on submerged substrates, sometimes at least facultatively oophagous; oral disc nearly terminal, jaw sheaths present, LTRF 010-011, marginal papiliae greatly reduced to absent; see table 2.2, figures in chap. 12, arboreal, benthic, carnivorous, nektonic, neustonic, suspension feeder, and suspension rasper. Magenhauptzelle. "stomach main cells" (chap. 5). Malthner ceiis. Paired ceiis in the basal plate of anamniotes that mediate fast start and stade responses, especially in aquatic animais; largest ceiis in the brain stem and persist after metamorphosis in terrestrial forms of amphibians. manicotto glandulare. Sleevelike glandular larval stomach of larval Types 3 and 4.

marginal papiila(e) (Aitig 1970) [labial, oral, lower, or upper festoon or fringe; buccal, dermal, or peribuccal papiilae; papillary border]. Describes any papiiia(e) anywhere on the margin of the oral disc; cornmonly appear as a complete series [circumoral], around margin or with dorsal [rostral gap], ventral [mental gap], or dorsal and ventral gaps. mature. Describes an individual that attained sexual maturity regardless of morphology; does not apply to a tadpole in latter stages of development; see froglet, hatchling, larva, and metamorph. medial. (1) Refers to or toward the sagittal line relative to a referente. (2) Describes a vent tube that opens parailel with the plane of the ventral fin or tail muscle; see dextral and sinistral, G. F. Johnston and Aitig (1986) discuss other variations of the vent tube. median ridge (Wassersug 1980). Transverse flaplike ridge of various shapes suspended from the roof of the buccal cavity; see buccal papiliae. mediogyrinid (midventral spiracle). Having a single spiracle anywhere on the midventd, sagittal line (e.g., ascaphids and microhylids); may be on the chest, abdomen, or ventral to the vent; see amphigyrinid, laevogyrinid, paragyrinid, and sinistral. melanin. Black or brown pigment that is a complex tyrosineprotein polymer; see chromatophore, melanophore, and melanosome. melanophore. Dermal or epidermal chromatophores of various shapes that produce brown to black pigments that give colorations by the presence of melanin; see blotch, dot, chromatophore, melanophore, melanosome, mottled, melanosome, spot, punctate, and stellate. melanosome (J. D. Taylor and Bagnara 1972). Pigrnentcontaining organeiie in a melanophore; see melanin. mesangium. Tissue between the glomerular/glomdar capillaries composed of mesangial cells and mesangial matrix. metabolic depression. A normaily large decrease in the standard metabolic rate that is usuaiiy triggered by some major change in physiological status, such as estivation or hibernation; caused by biochemical regulatory mechanisms that make this a controlled response. metabolic oxygen conformer. Animal whose rate of 0, consumption is independent of ambient PO, over some specified range of ambient PO,; see metabolic oxygen regulator. metabolic oxygen regulator. Animal whose rate of 0, consumption is dependent on ambient PO, over some specified range of ambient PO,; see metabolic oxygen conformer. metabolic scope. The degree that a baseline metabolic rate, either the standard metabolic rate (SMR) or resting metabolic rate (RMR), can be increased by strenuous activity; may be expressed in relative (factorial increase) or absolute (incremental increase) terms. metamorph. An immature frog that resdts from the metamorphosis-of an exotrophic tdpole; see embryo, hatchling, immature, iuveniie, larva, metamorphic, metamorphic . , ciimax, p.remetamo*hosis', and prometamorphosis. metamorphic. General term that describes either the process or an individual in the process of metamorphosis; see larva, hatchling, metamorph, metamorphic ciimax, premetamorphosis, prometamorphosis, and tadpole. metamorphic ciimax. That period when metamorphic changes are the most profound and most rapid; Gosner 42-46; see embryo, immature, juvenile, larva, metamorph, metamorphic, metamorphosis, premetarnorphosis, prometamorphosis, and tadpole.
<

GLOSSARY

345

metamorphosis. Process of a nidicolous endotrophic or exotrophic tadpole changing into a fioglet; most changes occur during stages 4 2 4 6 of Gosner (1960); other definitions (chap. 2) are based on endocrinological data; often used in synonymy with metamorphic climax; see metamorph, metamorphic, premetamorphosis, and prometamorphosis. minor metamorphosis. Metamorphic changes in morphology and ecology characterized by minimai variation between larvai and postlarvai stages (i.e., salamander metamorphosis as opposed to that of a frog). monoculture. A population containing oniy a single species, usually as a cultured population. mottled. A color pattern formed by haphazard arrays of usually dark streaks, lines, or blotches on a pale background or pale markings on a dark background; see blotch, dot, punctate, spot, and stellate. mouth [oral apparatus, oral disc, and mouthparts (in referente to mouth]. Refers oniy to that opening formed by the rupture of the oropharyngeai membrane to connea the buccopharynx with the exterior; does not refer to any mouthparts. mouthparts (oral apparatus). Colleaive term for all sofi and keratinized feeding structures largely external to the mouth (e.g., labia and all associated structures, tooth ridges, labial teeth, and jaw sheaths); does not refer to any strucnires in the buccopharynx. muciferous crypt [neuromast]. Refers to openings of unicellular glands but has also been used incorrectly in referente to neuromasts; should be abandoned because of these confusing usages. multiserial. Describes structures occurring in three or more series at a site, as tooth rows on a tooth ridge (e.g., some rows ofRrcaphus) or marginal papillae; see biserial and uniserial. muscle height (Altig and Johnston 1989). Maximum height (verticai distance) measured from the junction of the body wall and the ventral margin of the tail musculature; see chap. 3. muscle width (Altig and Johnston 1989). Maximum transverse width of the tail muscle measured in dorsai view at the same plane as muscle height. naris, nares (nostrii). Usually refers to the externai openings of the nasal canal; internal opening in the roof of the buccopharynx is the choana(e) or internai nares. nasolacrimal duct [lacrimal duct or gland, orbitonasal line]. A d u a connecting the corner of the eye with the naris; appears as a contrastingly pigmented line during its formation in later stages of a tadpole. nephrotome (nephromere, mesomere, intermediate plate). That part of the mesodermic somite that develops into excretory organs. negative Bohr shift. Increase in the affinity of hemoglobin (decreased P,,) associated with a decrease in blood pH, as opposed to the normally observed decrease in affinity (increased P,,) that accompanies an increase in the acidity of the blood. nektonic (pelagic). (1) An ecomorphological guild that includes those lentic or lotic tadpoles that rasp food from submerged surfaces with keratinized mouthparts and do so somewhere within the water column including quiet backwaters in lotic sites; body compressed, eyes lateral, fins high with pointed tip with or without a flagellum; dorsal fin ofien originates well anterior to dorsal tail-body junc-

tion; see table 2.2, figures in chap. 12, adherent, arboreal, benthic, carnivorous, clasping, fossorial, gastromyzophorous, macrophagous, neustonic, psammonic, rheophilous, semiterrestrial, suctorial, suspension feeder, and suspension rasper. neuromast [minute glands, lateral line pore, muciferous crypt]. Complex (see chap. 6), pressure-sensitiveorgans arranged in various patterns on the body and tail; collectively form the lateral iine system. Any definition using "pore" should not be used because the lateral iine system of amphibians does not indude pores or channels like in some fishes. neustonic (surface feeder). An ecomorphological guild that includes those lentic or lotic tadpoles that filter particles in or near the surface film with upturned (= umbelliform) oral disc with or without keratinized mouthparts; body depressed; eyes usually lateral; tail long with low fins; see table 2.2, figures in chap. 12, adherent, arboreal, benthic, carnivorous, clasping, fossorial, gastromyzophorous, macrophagous, nektonic, psammonic, rheophilous, semiterres&ial,suspension feedir, and suspension rasper. nidicolous. (1)Developmental mode of a frog whereby a tadpole morbhotype ofsome form develops gut the in&vidual does not feed; morphology varies from a fully fornled tadpole at one end of a continuum through many developmentai variations that deviate progressively from the developmentai pattern of a typical tadpole. (2) An ecomorphological guild that includes those species that have some form of nidicolous development; see table 2.2, figures in chap. 12, endotrophy, exotrophy, direct development, exoviviparous, ovoviviparous, paraviviparous, and viviparous. nonadditive genetic variance. That fraction of the total phenotypic variation primariiy caused by deles that display dominance and epistasis; see additive genetic variance. nucleus. (1) An organelle of a cell housing the genetic material and other structures. (2) Aggregation of cells within the brain that performs a similar function or operates toward a similar end; see ganglion. opercular tube or canal. Single or paired stmctures that conn e a the branchiai chamber to the externai opening of the spiracle in tadpoles with midventrai or ventrolaterai spiracles. opercular fold. Outgrowth from the hyoid arch in stages 20-24 that eventually forms a covering over the gills and associated structures and fuses with the body wall in patterns that produce several spiracular configurations; not a homolog of the piscine operculum or auditory elements; see amphigyrinid, laevogyrinid, mediogyrinid, paragyrinid, peribranchial chamber, and sinistral. operculum. (1) The completed covering of the buccopharyngeai area of a tadpole provided by the development of the opercular fold; not homologous with the piscine gill cover of the same narne; see amphigyrinid, laevogyrinid, mediogyrinid, paragyrinid, sinistral, and spiracle. (2) An auditory element in the fenestra ovalis; see chap. 4. oral apparatus (mouthparts) [buccai apparatus, oral sucker]. Collective term applicable to all sofi and keratinized structures mostly externai to the mouth, see barbel, labium, marginal papiae, mouth, oral disc, and submarginal papillae. oral disc (Altig 1970) [mouth disc, disk, or pad; oral sucker]. Composite structure composed of upper and lower labia, usually with transverse tooth ridges surmounted by rows of labial teeth; marginal papillae occur in various configura-

346

GLOSSARY

tions at the edges and submarginal papiiiae occur in various patterns on the face of the disc; does not refer to the mouth; a major part of the mouthparts; see oral apparatus. oral flap. See labial flap. oral hitching (Aitig and Brodie 1972). Describes the means of movement of rheophilous tadpoles by extension-retraaion cydes of the oral disc whereby attachrnent to the substrate is maintained; largely independent from feeding activities; probably gastromyzophorous forms do similar actions. oral papiiia(e). See labial papiila(e). oral sucker. A loose term for the oral disc of a suctorial tadpole; use for position maintenance in flowing water; see belly flap, beiiy sucker, funnel mouth, gastromyzophorous, mouthparts, oral apparatus, and oral disc. oral tube (Lavilla 1990). Describes the shape of a protruding oral apparatus at a specific point in the feeding cycle of a tadpole; probably not a speciic struaure. ovoviviparous. (1) Developmental mode of a frog whose endotrophic larva is retained within the female's reproductive tract and is born as a metamorph with no additional energy beyond viteiogenic yolk of the original egg being supplied by the mother. (2) An ecomorphological guiid that includes those species with ovoviviparous development; see table 2.2, figures in chap. 12, direct development, endotrophy, exotrophy, exoviviparous, nidicolous, paraviviparo&, and viGparous. PR-1. See AR-1. papiiia(e). (1) A fleshy projection(s) on the margin or face of the oral disc, most often circular in cross section and with a blunt tip; see barbel, marginal papiiiae, and submarginal papiiiae. (2) A fleshy projection in general, like the various structures on the floor and roof of the buccopharynx; see chap. 5. papiiiate, papiiiiferous. Describes structures or surface that have papiiiae; see barbel, buccai papiiiae, marginal papiilae, and submarginal papiiiae. paragyrinid (G. F. Johnston and Aitig 1986). Describes a tadpole having a single spiracle on the left side but well below the body axis, sometimes nearly midventray (e.g., phyllomedusine hylids); see amphigyrinid, laevogyrinid, mediogyrinid, and sinistral. paraviviparous (Altig and. Johnston 1989). (1) Developmental mode of a frog whereby a froglet is hatched from a egg someplace other than the reproductive traa of a parent; female parent is usudy involved and no additional energy is supplied beyond the vitellogenic yolk. (2) An ecomorphologicai guiid that includes those species with some form of paraviviparous development; see table 2.2, figures in chap. 12, direct development, endotrophy, exotrophy, exoviviparous, nidicolous, ovoviviparous, and viviparous. peribranchial chamber (opercular chamber). That space anterior to the peribranchial wail (= diaphragm) that separates the buccopharynx and abdominal cavities and delimited by the o p e r d u m ; branchial arches, giiis, filter plates, and other structures occur w t i this chamber; see opercuiar fold ihn and spiracle. peribranchial wall (diaphragm). See peribranchial chamber. persistent epidermal gill (Viertel 1991). Until stage 24 external, from stage 25 internal giiis persisting up to metamorphic climax; develop along the ventral parts of branchial arches I-W, epidermis overlays the pharyngeal endoderm; see chap. 5 and transient epidermal giiis. phenology. Temporal sequence of events, often refers to seasonal changes in c a h g and breeding of adults and the presente of tadpoles.

phytotelm, phytotelmata. Water-holding cavity in some part of a plant (e.g., bromeliad cistern or tree hole) or plant product (e.g., a nut shei); often extended to refer to other sma, isolated bodies of water (e.g., snail shells). A subset of lentic habitats. plasticity. Variation in the phenotypic expression of a genotype over or influenced by a range of environmentai cqnditions. PO,. Partial pressure of O, in the environment; see VO, preferred body temperature (PBT). Mean (sometimes median or mode) body temperature selected by an animal in a laboratory thermal gradient or thermal shuttlebox. premetamorphosis (Etkin 1968). Describes the developmental attainment of a tadpole in stages 25-35, the primary period of body growth; see froglet, larva, metamorph, metamorphic, prometamorphic, and tadpole. prometamorphosis (Etkin 1968). Describes the developmental attainment of a tadpole in stages 3 6 4 1 , the primary period of rapid limb growth; see larva, metamorph, metamorphic, prernetamorphic, and tadpole. psammonic. An ecomorphologicai guiid that indudes those lotic tadpoles that inhabit loose sand in slow-flowing streams; perhaps feed passively while buried; see table 2.2, chap. 3, figures in chap. 12, adherent, benthic, clasping, fossorial, gastromyzophorous, nektonic, neustonic, rheophilous, semiterrestrial, and suctorial. punctate. (1) A pattern of s m d , often dark spots or dots, usuay with distinct margins. (2) An individual melanophore that is more or less circular in outline; see blotch, mottled, and steiiate. Q,,. Rate of a process at temperature t ("C) divided by rate at t - 10; generay for any temperature interval Q , = rate at T,/rate at T, raised to the power of (10/q - T,). quantitative genetics. Science ". . . concerned with the inheritance of those differences between individuals that are of degree rather than of kind, quantitative rather than qualitative." (Falconer -. .. 1989). rasp. A general feeding mode whereby a tadpole uses keratinized mouthparts to harvest particulate or sma pieces of various materials from submerged surfaces or pick up materials from sedimented accumulations; see suspension feeding and suspension rasper. receptor unit (Zakon 1984) [plaque, stitch]. Receptors of mechanical stimuli that form the lateral line system; located in tight clusters and developmentay derived from and aligned in the same direction as the single primary organ; see neuromast. replacement tooth (Aitig and Johnston 1989). One of several f d y formed labial teeth interdigitated sequentiay beneath the erupted tooth; see labial teeth, tooth row, and tooth series. reticuioendothelial system. Connective and endothelial tissues related to the lymphatic system, produces immune responses, and functions in myelopoiesis. rheophiious. Generic term that describes tadpoles of several ecomorphological guiids that are modified in various ways for inhabiting microhabitats in the flowing parts of lotic systems; see adherent, benthic, dasping, fossorial, gastromyzophorous, nektonic, neustonic, psammonic, semiterrestrial, and suctorial. Riesenzeiien. "Aiarm cells" (Fox 1988); see chap. 5. rim, rimmed. (1) A raised margin of an opening, usuall!. the nares, with or without papiiiae. (2) With a rim. Rohon-Beard ceiis. Cells that mediate mechanoreception from free nerve endings in the epidermis, graduay replaced b?the dorsal root ganglion system; among the h s t senso?
-

GLOSSARY

347

neurons to appear in the spinal cord of anamniote vertebrates and persist in amphibians until metamorphosis. routine metabolic rate. Metabolic rate of an ectotherm that is acciimated to a specified temperature and exhibits normal activity. Schieimkopfchenzeilen. "Mucous-headed cell" (chap. 5). school [aggregation]. A specific form of congregated tadpoles that involve some sort of social interaaion and exhibit at least indications of polarized orientation and perhaps coordinated movements; see aggregation, biosocial aggregation, and taxic aggregation. scoptanura (Starrett 1973). Word synonym of Orton's (1953) Type 2; microhyloid of O. M. Sokol (1975). semiterrestrial [subaerial]. An ecomorphological guiid that includes those lotic tadpoles that have marginal papiiiae with an anterior gap; L T m usudy 213, jaw sheaths usudy with a high and narrow arch; inhabit rock faces, leaves, and the forest floor that provide damp or wet surfaces with little free water; see table 2.2, figures in chap. 12, adherent, belly flap, belly sucker, benthic, clasping, fossorial, gastromyzophorous, nektonic, neustonic, psarnmonic, rheophilous, and suctorial. serration. Sawlike projections of various densities, orientations, shapes, sizes, and densities along the cutting edge of the jaw sheaths; should not be used in reference to marginal papillae or labial teeth; see cusp. sheath. Hollow, basal part of a labial tooth (Gosner 1959) embedded in the tissue of the tooth ridge; see head, body, taii sheath, tooth row, and tooth series. sinistral. To the left. (1) Describes a single spiracle on left side (laevogyrinid); see amphigyrinid, laevogyrinid, mediogyrinid, and paragyrinid. (2) Describes a vent tube with the aperture opening to the left of the plane of the ventral h; dextral and medial. see size refuge. Avoidance of predation by the attainment of a body size which reduces or eliminates predation by gape-limited predators; see gape limited. spinal nerve. One of 20-29 nerves that arises from the spinal cord; typicdy formed by the confluence of a dorsal sensory root and a ventral motor root; the number reduces to about 11 after metamorphosis; see cranial nerve. spiracle [atrial opening; spout]. One or m o opening(s) of different shapes and positions for the exit of the water pumped into through the buccopharynx for respiration and feeding; not homologous with the elasmobranch spiracle; see amphigyrinid, laevogyrinid, mediogyrinid, paragyrinid, and sinistral. spot. Dark, pigmented mark with relatively even margins but larger than a dot and more rounded and discrete than a bltch; see mottled, punctate, and steilate. standard metabolic rate (SMR). Metabolic rate of an ectothenn under resting conditions and acclimated to a specified temperame; equivalent to basal metabolic ratc of mammals. stellate. (1) Describes a pigmented mark that is roughly starshaped or at least has multiple projections from a central region. (2) An individual chromatophore that is simiiarly star-shaped; see blotch, dot, spot, mottled, and punctate. Sternzellen. See Kupffer cell (chap. 5). Stiftchenzellen. "pin-cells" (chap. 5). submarginal papiila(e) (Altig 1970) [extramarginal, infralabial, inner or retromarginal papiae]. Fleshy projection(s) anywhere on the face of the oral disc except the margin and perhaps some other places in tadpoles that lack jaw sheaths (e.g., at the sites where jaw sheaths would be expected to occur); see barbel, marginal papilla, and papilla.

suctorial. (1) Describes the usudy large, ventral-facing oral apparatus of tadpoles that maintain position and feed in fast water by adhering to rocks via the large oral disc (chap. 4); should not be used in reference to the beily sucker of gastromyzophorous tadpoles; see beily flap. (2) An ecomorphologicai guiid that includes those lotic and rheophilous tadpoles with s m d , closely spaced marginal papiiiae in a complete series around the oral disc; LTRF > 213 to a maximum of 17/21; inhabits faster water than adherent and clasping tadpoles, position maintenance via oral disc continuous; see table 2.2, figures in chap. 12, beily flap, benthic, fossorial, gastromyzophorous, nektonic, neustonic, psammonic, and semiterrestrial. suspension feeder, suspension feediig (= fiiter feeding). (1) Harvesting of naturdy suspended particles by pumping water in through the mouth, over the buccopharyngeal filtering system and out the spiracle(s); tadpoles that specialize in this mode of feeding usudy lack a keratinized and most soft mouthparts. (2) An ecomorphologicai guiid that includes those lentic tadpoles that are behaviorally and morphologicdy (e.g., lateral eyes and depressed bodies) modified to feed in this manner; see table 2.2, figures in chap. 12, arboreal, benthic, carnivorous, macrophagous, nektonic, neustonic, and suspension rasper. suspension rasper. An ecomorphologicai guiid that includes those lentic tadpoles that seemingly feed by fiitering suspended particles from within the water column and rasping submerged surfaces; jaw sheaths and labial teeth (usudy 2/ 3) present, marginal papiae with anterior gap; eyes lateral; flagellum common t d ; see table 2.2, figures in chap. 12, arboreal, benthic, carnivorous, macrophagous, nektonic, neustonic, and suspension feeder. tadpole. Nomeproductive exotrophic or a nidicolous endotrophic larva of a frog between stages 25-41; does not refer to immatures of other amphibians; see froglet, hatchiing, juvenile, larva, mature, and metamorph. taii. That part of a tadpole minus the body; with a segmented musculaturc and dorsal and ventral fins; defined for measurement as the straight-line distance from the tail tip anteriorly to the body t e r k u s . ta axis (Van Dijk 1966). An imaginary line connecting the body terminus and the tip of the straightened tail; see body axis tail-body junction. Literay the junaion of the body and taii of a tadpole, but for different purposes, this junaion can be defined three ways. For most purposes of description and measurement, the site of the body terminus should be med. In some cases the ventral body tcrminus (where the ventral extent of the curvame of the posterior margin of the body contaas the plane of the ventral margin of the tail musde) and the dorsal body termius (where the dorsal extent of the curvature of the posterior margin of the body contacts the plane of the dorsal margin of the tail musde) can be used. taii flipping. A surprisingly effiuent and accurate means of locomotion by tadpoles and particularly hatchlings, espec i d y those that hatch from arboreal egg masses; move by frantic flips of the tail and not tail undulations. ta height [tail depth]. Greatest vertical distance (i.e., height) of the tail muscle plus both fins; this site may not be the maxirnum dimension of both fins. taii length. Straight line, longitudinal distance from the body terminus to the absolute tail tip, not just to the end of the musculame; see body length. tail ratio. Total length divided by tail length; see body ratio. tail sheath. Dense layer(s) of probable conneaive tissue in the

348

GLOSSARY

proximal third of the taii of some tadpoles, sometimes visible in living specimens but more commonly seen in preserved specimens. taxic aggregation (aggregation) [asocial aggregation]. A general term that describes a group of tadpoles that are congregated because of each individual responding to some soa (e.g., light or temperature) of taxis and not because of social interaction; qualities of a school (i.e., polarized orientation and coordinated movements) are not implied; see biosocial aggregation and school. temperature acclimation. Accommodation by an organism in a laboratory setting by physiologically adjusting to the new temperatures; under natural conditions, it is called temperature acclimatization. temperature acclimatization. See temperature acclimation. tooth ridge [dental plate, labial ridge]. Serial, transverse fleshy ridges on the face of the oral disc surmounted by a row(s) of labial teeth generated in a mitotic wnes in the base of the ridges; present in some tadpoles that never have labial teeth see tooth row and tooth seria. tooth row [fringed fold, tooth series, supralabial row on upper or anterior labium, infralabial row on lower or posterior labium]. Transverse, u u d y close-spaced row of labial teeth each embedded within a tooth ridge. Rows on either labium face the mouth; see A-2 gap, gap, labial tooth row formula, and tooth series. tooth seria. Coiiective term for the erupted labial tooth at a given site in a tooth row and each replacement tooth interdigitated sequentially below it; see A-2 gap, gap, tooth ridge, tooth series, labial tooth row formula. total length [standard length]. Straight line distance measured from the tip of the snout to the tip of the t d ; see body length, body terminus, and tail length. transient epidermal gill (Vieael 1991). External giiis of other authors, develop along the ventrolateral paas of branchial arches I-IV and atrophy by stage 25; see chap. 5 and persistent epidermal gill. typhlosolis. Muscular invagination of the foregut. umbelliform [funnel mouth]. Describes an upward-facing oral disc, a convergent trait in several families, used for harvesting particulate matter from the surface film regardless of actual mouth position or orientation; oral disc may be formed from parts of either labium and keratinized mouthparts often lacking see neustonic and suctorial. umbraculum (Van Dijk 1966). Fleshy projection of the iris into or over part of the pupil in some ranids; assumed to protea eye from excessive light; see elygium. uniserial. Describes structures occurring in one series or row
-

per site, as one row of labial teeth per tooth ridge (e.g., most tadpoles); see biserial and multiserial. Urwirbelfortsatz. Primordial vertebral process; see chap. 4. velum (Wassersug 1976a, b; chap. 5). Glandular organ between the buccal cavity and the pharynx that regulates water flow oharvnx. into the I vent [mal opening, anus]. Posterior intestinal opening of a tadpole. Because arnphibians have a doaca, this opening is not an anus, and ali referentes as such should be omitted; see vent flap and vent tube. vent flap [mal flap]. Fleshy flap of various shapes that extends from the body w d posteriorly and ventral to the vent tube. Sometimes encloses the hind limb buds and occurs most commonly in rheophiious tadpoles. vent tube [anal or cloacal tube, mal tube piece, cloacal taii piece, proctodaeal tube]. A tube for voiding of feces with many variations in morphology; projeas from the body wall distally to an opening that may lie parallel with the plane of the ventral h (medial) or to the left (sinistrai) or right (dextral) of that plane. In a few cases, the tube does not exit the body on the sagittai lhe; see vent and vent flap. ventral body terminus. See body terminus and taii-body junction. visceral arch. Cartilaginous or bony bars of the visceral skeleton, including the mandibular arch (= mandibulare and palatoquadratum), hyoid arch, and three branchial arches. visceral pouch. Pharyngeal (endodermal) evaginations between the visceral arches from which the Eustachian,tubeis derived; pouches 2-4 open as gill slits; the ilter plate anlagen and the branchiogene endocrines are derived from this endoderm. viviparous (Altig and Johnston 1989). (1) Developmental mode of a frog which results in a larva (= fetus) that develops by consurning maternally derived, oviducal materiais after the exhaustion of viteiiogenic yolk. (2) An ecomorphological guild that includes species that have viviparou development; see table 2.2, figures in chap. 12, direct development, endotrophy, exotrophy, exoviviparous, nidicolous, ovoviviparous, and paraviviparous. VO,. A measure of the volume of 0, consumed per unit time; see PO,. whole-body 0, conductance. Change in the rate at which 0, can be consumed by a metabolic 0, regulator per unit +ange in ambient PO,; equivalent to the slope of a plot of VOZagainst ambient PO, in the range of metabolic 0, conformity. xenoanura (Starrett 1973). Word synonym of Oaon's (1953) Type 1tadpole; pipoid of O. M. Sokol(1975).
i

LITERATURE CITED

Abe, A. S., and H. P. Godinho. 1991. Tolerante to high temperatures in tadpoles of L~ptoductylus~scus Hyh@cwaand ria in temporary ponds (Arnphibia, Leptodactylidae, Hylidae). 2001. Anz. 226:280-284. Abelson, A., T. Miloh, and Y. Loya. 1993. Flow pattems induced by substrata and body morphologies of benthic organisms, and their roles in deterrnining availability of food particles. Limnul. Oceanograph. 38: 1116-1 124. Abraham, M. H., and C. Rafols. 1995. Factors that influence tadpole narcosis. An LFER anaiysis.J. Chem. Soc. 10:18431851. Abravaya, J. P., and J. F. Jackson. 1978. Reproduction inMacrogenwjlottus alipwi Carvalho (Anura: Leptodactyiidae). Nat. Hist. Mus. LosAngeles Co. Contrib. Sci. (298):1-9. Accordi, F., and P. Cianfoni. 1981. Histology and ultrastructure of the adrenai gland of Rhaaphomts ieucomystax (Arnphibia, Anura). Boll. 2001. 48:277-284. Accordi, F., and E. Grassi-Milano. 1977. Catecholaminesecreting ceils in the adrenai gland of Buf0 buj during metamorphosis and in the aduit. Gen. Comp. Endom'ml. 33: 187-195. . 1990. Evolution and developmentof the adrenai gland in amphibians. pp. 257-268. In Biology andphyswlogy of amphibians. Progress in wology 38. Proc. FirstIntl. Symp. Bwl. Physwl.Amphibians, Karlsruhe, Federai Republic of Germany, edited by W. Hanke. Stuttgart: Gustav Fischer-Verlag. Adams, M. J., K. L. Richter, and W. P. Leonard. 1997. Surveying and monitoring amphibians using aquatic funnel traps. pp. 47-54. In Sampling amphibiam in lentic habitats, edited by D. H. Olson, W. P. Leonard, and R B. Bury. Northwest Fauna Number 4 Society for Northwestem Vertebrate Biology, Olyrnpia, WA. Adamson, L., R. G. Harrison, and I. Bayley. 1960. The development of the whistiing frog E i e u t h e r o ~ l u s rm~~nicensis of Barbados. Proc. 2001. Lundun 133:453469. Soc. Adamson, M. L. 1981a. Gjnniwh batrachiensis (Waiton, 1929) n. comb. (Oxyuroidea; Nematoda) frorn tadpoles in eastern and central Canada. CanuianJ.2001. 1344-1 350. 59: . 1981b. Seasonai changes in populations of Gjnnzwh batrachiensis (Waiton 1929) in wild tadpoles. Canadian J. 2001. 1377-1386. 59: Adler, K. 1970. The role of extraoptic photoreceptors in amphibian rhythrns and orientation: A review. J. Herpetol. 4:99-112. . 1976. Extraocular photoreception in amphibians. Photachem. Photobwl. 23:275-298. Adolph, E. F. 1931. The size of the body and the size of the environment in the growth of tadpoles. Bwl. Bull. 61:350-375. Ma'a, F. M. 1979. Nyctositum ameita, n. gen., n. sp., an endocomensai ciliate of tadpoles ofAcanthixalusspzmsus. Protistologica 15:333-336. . 1983. Neonyctothms (Ciiiophora, Heterotrichida), a new genus comensai in amphibian tadpoles (Anura) from Cameroons. Protistologica 19:141-147. Af'ia'a, F. M., and J.-L. Arniet. 1994. Progrs rcents dan Ia connaissance des Nyctothres (Protozoaires, Ciiis, Htrotriches) associs aux Anoures. Alytes 12:75-92. Agarwal, S. K., and I. A. Niazi. 1980. Development of the mouth-parts in the tadpoles of Rana t@na (Daud.). Proc. IndianaAcad. Sci. 89: 127-131. Aggarwai, S. J., and A. Riggs. 1969. The hemoglobins of the bullfrog, Rana catesbeiana I. Purification, m i n o acid composition, and oxygen equi1ibria.J. Bwl. Chem. 244:2372-2383.

350

LITERATURE CITED

Ahl, E. 1931. AnuraIII (Polypedatidae).DasTieweich 55: 1 4 7 5 . Ahlgren, M. O., and S. H. Bowen. 1991. Growth and survival of tadpoles (Bufo amricanus) fed amorphous detritus derived from natural waters. Hydrobwlogia 218:49-5 1. Aho, J. M. 1990. Helminth communities of amphibians and reptiles: Comparative approaches to understanding patterns and processes. pp. 157-1 95. In Parasite communities:Patterns and processes, edited by G. W. Esch, A. O. Bush, and J. M. Aho. New York: Chapman and Haii. Aichhorn, H., and A. Lametschwanduler. 1996. Vascuiar regression during amphibian metamorphosis - a scanning electron microscope study of vascular corrosion casts of the ventral velum in tadpoles of &nopus iaais Daudin. Scanning 18:447455. Aichinger, M. 1987. Annual activity patterns of anurans in a seasonal neotropical environment. Oecologa 71:583-592. Akin, G. C. 1966. Self-inhibitionof growth in Ranapipiens tadpoles. Physwl. 2001. 39:341-356. Aiberch, P. 1982. The generative and regulatory roles of development in evolution. pp. 19-36. I Enaironmental aduptan tion and evolution, edited by D. Mossakowski and G. Roth. Stuttgart: Gustav Fischer-Verlag. . 1985. Problems with the interpretation of developmental sequences. Syst. 2001. 34:46-58. . 1987. Evolution of a developmental process -irreversibility and redundancy in amphibian metamorphosis. pp. 2 3 4 6 . In Develupment as an evolutwnary process, edited by R. A. RaE and E. C. RaE. New York: Aian R. Liss, Inc. Aiberch, l?, and E. A. Gaie. 1983. Size dependence during the development of the amphibian foot. Colchicine-induced digitalL loss and reduckon. J Embryol. Exp. Morphol. 76: 177-197. Aiberch, P., S. J. Gouid, G. Oster, and D. B. Wake. 1979. Size and shape in ontogeny and phylogeny. Paleobiology 5 :296-3 17. Aicala, A. C. 1962. Breeding behavior and early development of frogs of Negros, Philippine Islands. Copeia 1962:679-726. Aicaia, A. C., and W. C. Brown. 1957. Discovery of the frog Cornufrguentheri on Negros Island, Philippines, with observations on its life history. Hqetologica 13:182-184. . 1982. Reproductive biology of some species of Philautus (Rhacophoridae) and other Philippine anurans. Philippine J Biol. 11:203-226. Aicocer, I., X. Santacruz, H. Steinbeisser, K. H. Thierauch, and E. M. De1 Pino. 1992. Ureotelism as the prevailing mode of nitrogen excretion in larvae of the marsupial frog Gnstrotheca riabambae (Fowler)(Anua,Hylidae). Comp.Bwchem. Physiol. 101k229-23 1. Aiexander, R. D. 1974. The evolution of social behavior. Ann. Ra7 Ecol. Syst. 51325-383. Alford, R. A. 1986a. Effects of parentage on competitive ability and vulnerability to predation in Hyla chrysoscelis tadpoles. Oecologa 68: 199-204. . 1986b. Habitat use and positional behavior of anuran larvae in a northern Florida temporary pond. Copia 1986:408423. . 1989a. Effects of parentage and competitor phenology on the growth of larvai Hyla chrysoscelis. Oikos 54:325-330. . 1989b. Competition between larval Ranapalustvis and Bufo americanus is not dected by variation in reproductive phenology. Copeia 1989:993-1000. . 1989c. Variation in predator phenology dects predator performance and prey community composition. Ecology 70:206-219.
-

. 1994. Interferente and exploitation competition in larval Bufo marinus. pp. 297-306. In Advances in ecology and environmental sciences, edited by P. C. Mishra, N. Behera, B. K. Senapati, and B. C. Gum. New Deihi: Ashish Publ. House, New Deihi. Aiford, R. A., and M. L. Cmmp. 1982. Habitat partitioning among size classes of larval southern leopard frogs, Rana utn'cularia. Copeia 1982:367-373. Alford, R. A,, and R. N. Harris. 1988. Effeas of larval growth history on anuran metamorphosis. American Nat. 131:91-106. A h r d , R. A., and G. D. Jackson. 1993. Do cephalopods and the larvae of other taxa grow asymptotically?American Nat. 141:717-728. Aiford, R. A., and H. M. Wilbur. 1985. Priority effects in experimental pond communities: Competition between Bufo and Rana. Ecology 66: 1097-1105. Aiiey, K. E., and M. D. Barnes. 1983. Birth dates of trigeminal motoneurons and metamor~hicreorganization of the iaw . Neurol. 218:395myoneuronal system in frigs. J comp. 405. Aison, J. D. 1992. Acoustic modulation of neural activity in the preoptic area and ventral hypothalamus of the green treefrog (Hyla cinerea).J Comp.Physwl. 171A:387-395. Ai-Mukthar, K . A. K., and A. C. Webb. 1971. An ultrastructural study of primordial germ celis, oogonia, and early oocytes in Xevwpus laevi5.J. Embtyol. Exp. Morphol. 26: 195-21 7. Altig, R. 1970. A key to the tadpoles of the continental United States and Canada. Herpetologca 26:180-207. . 1972. Notes on the larvae and premetaniorphic tadpoles of four Hyla and three Rana with notes on tadpole color patterns.J ElzshaMitchell Sci. Soc. 88:113-119. . 1973. Preliminary scanning electron observations of keratinized stmctures in amphibian larvae. HISS News-J. 1:129-131. . 1975. Freeze-drying anuran tadpoles for taxonomic exaniinations. Hqetol. Ra7 6: 13. . 1987. Key to the anuran tadpoles of Mexico. Southwest. Nat. 32:75-84. Aitig, R., and R. A. Brandon. 1971. Generic key and synopsis for free-living larvae and tadpoles of Mexican amphibians. Tulane Stud. 2001. 17: 10-1 5. Bot. Altig, R., and E. D. Brodie, Jr. 1972. Laboratory behavior of Ascaphw truei tadpoles. J Hqetol. 6:21-24. Aitig, R., and A. Channing. 1993. Hypothesis: Functional significance of colour and pattern of anuran tadpoles. Hqetol. J. 3:73-75. Aitig, R., and M. T. Christensen. 1981. Behavioral characteristics of the tadpoles of Rana heckscheri. J Herpetol. 15:151-154. Aitig, R., and G. F. Johnston. 1986. Major characteristics of free-living anuran tadpoles. Smithsonian Herpetol. Info. Sem. (67):1-75. . 1989. Guilds of anuran larvae: Relationships arnong developmental modes, morphologies, and habitats. Hqetul. Monogr: (3):81-109. Aitig, R., and J. l? Keliy. 1974. Indices of feeding in anuran tadpoles as indicated by gut characteristics. Herpetulogica 30:200-203. Altig, R., and W. ~McDearman. 1975. Clearance times and per cent assimilation of five species of anuran tadpoles. Hqetuhgica 3 1:67-69. Aitig, R., and W. L. Pace. 1974. Scanning electron photomicrographs of tadpole labial teeth. J Herpetol. 8:247-25 1.
-

LITERATURE CITED Altman, J. S., and E. A. Dawes. 1983. A cobalt study of medullary sensory projections from lateral iine nerves associated with cutaneous nerves, and the VIIIth nerve in adult =nopus.J. Comp.Neurol. 213:310-326. Aivarado, R. H., and T. C. Cox. 1985. Action of polyvalent cations on sodium transport across skin of larval and adult Rana 236: catesbeiana.J. Exp. 2001. 127-136. Aivarado, R. H., and S. R. Johnson. 1966. The effects of neurohypophysial hormones on water and sodium balance in larvai and adult bullfiogs (Rana cutesbeiana). Comp. Biochem. Physwl. 18A:549-561. Alvarado, R. H., and A. Moody. 1970. Sodium and chloride transport in tadpoles of the bullfiog Rana catesbeiana.AmericanJ.Physwl. 218:1510-1516. Amboroso, E. C. 1968. The evolution of viviparity. Pwc. Roy. Soc.Med. 61:1188-1200. Amer, F. I. 1968-1969. On the occurrence and nature of Bidder's organ in the Egyptian toad, Bufo regztlaris Reuss. 2001. Soc. Egypt Bull. 21:39-62. . 1972. Observations on the urinogenid ducts of the Egyptian toad, BufO r g u l a k Reuss. 501. Ejypt BuU. Soc. 24:15-19. -. Amiet, J.-L. 1970. Morphologie et dveloppement de Ia larve de Leptodactylodonventrimamwratus (Boulenger)(Amphibien Anoure).Ann. Fac. Sci. Cameroun 1970:53-71. . 1971. Le ttard d'rlstylosternus c m ~ a t u s Boulenger (Amphibien Anoure).Ann. Fac. Sci. Cameroun 1971:85-98. 1974a. La ponte de Ia larve d'Opistbotbylux immaculatus Boulenger (Amphibien Anoure). Ann. Fac. Sci. Camwmn 1974:121-130. . 1974b. Le ttard d'Hyperolius obstetricansAhl (Amphibien Anoure). BuU. Inst. France Aj-zca Noire 36:973-981. . 1976. Les formes larvaires d'Hypwolius hrebsi Mertens et H. koehkri Mertens (Amphibien Anoures). Ann. Fac. Sci. Cameroun 1976:159-169. . 1979. Description de I'adulte et de Ia larve d'Hypemlzus bopeeti n. sp. (AmphibiaAnura, Hyperoliidae).Ann. Fac. Sci. Cameroun 1979:113-124. Anderson, J. D. 1972. Phototactic behavior of larvae and adults of two subspecies of Ambystoma mamodactylum. Herpetologica 28:222-226. Andrn, C., L. Hendrickson, M. Olsson, and G. Nilson. 1988. Effects of pH and aluminum on embryonic and early larval stages of Swedish brown frogs Rana arvalis, R. temporaria, and R. dalmatina.Holarctic Ecol. 11:127-1 35. Andrn, C., and G. Nilson. 1985. Breeding pool characteristics and reproduction in an island population of nattejack toads, Bufo calamita Laur., at the Swedish west coast. AmphibiaReptilia 6: 137-142. Andres, G. 1963. Eine experimenteiie analyse der Entwicklung der larvalen Pigmenunuster von fiinf Anurenarten. Zoo&iEa 40: 1-112. Andres, K. H. 1970. Anatomy and ultrastructure of the olfactory bulb in fish, Amphibia, reptiles, birds, and mammals. pp. 177-194. In T&e and smell in vevtebrates. CIBA Found. Symp., edited by G. E. W. Wolstenholme and J. Knight. London: J. and A. Churchiil. Andrews, R. M., and B. R. Rose. 1994. Evolution ofviviparity: Constraints on egg retention. Physwl. 2001. 1006-1024. 67: Angel, F. 1946. Faune de France. Rtptiks et amphibiens. Paris: Paul LeChevaiier. Anholt, B., S. Negovetic, and C. Som. 1998a. Methods for anaesthetizing and marking larval anurans. Herpetol. Rev. 29:153-154.
- - -

351

Anholt, B., S. Negovetic, C. Som, and R. Mulheim. 1998b. Marking tadpoles with VIE. <http://www.mplpwrc.usgs/ gov/marking/anholt.html> . Anholt, B. R., D. K. SkeUy, and E. E. Werner. 1996. Factors modLfying antipredator behavior in larval toads. Herpetologica 52:301-313. Anholt, B. R., and E. E. Werner. 1995. Interaction between food avadability and predation mortality mediated by adaptive behavior. Ecology 76:2230-2234. Annandale, N. 1912. Biological results of the Abor Expedition, 1911-12. I. Batrdia. Rec. IndianMus. 8:7-36, plates 11-IV. . 1917. The occurrence ofRanapkshei Gnther, in Kashmir. Rec. IndianMus. 13:417-418. . 1918. Some undescribed tadpoles from the hiils of southem India. Rec. IndianMus. 15:17-23, plate I. . 1919. The tadpoles ofNyctibatrachuspy~maeus Ixaand lus variabilis: A correction. Rec. Indian Mas. 16:303. Annandale, N., and S. L. Hora. 1922. Paraiiel evolution in the fish and tadpoles of mountains torrents. Rec. IndianMus. 24:505-510. Annandale, N., and C. R. N. Rao. 1918. The tadpoles of the families Ranidae and Bufonidae found in the plains of India. Rec. IndianMus. 15:2540, plate IV. Antal, M., R. Kraftsik, G. Szkely, and H. Van Der Loos. 1986. D i s d dendrites of frog motor neurons: A computer aided electron microscopic study of cobalt-Med cells. J. Neurocytol. 15:303-310. Antal, M., I. Tornai, and G. Szkely. 1980. Longitudinal extent of dorsal root fibres in the spinal cord and brain stem of the frog. Neuroscience 5: 1311-1322. Archey, G. 1922. The habitat and life history of Liopelmu hochstetteri. Rec. CanterburyMus. 2:59-71. Ardila-Robayo, M. C. 1979. Status sistematico de1 genero Geobatrachus Ruthven 1915 (Amphibia: Anura). Caldasia 12:383495. Armstrong, J. B., and G. M. Malacinski, eds. 1989. Deve&mental bwlogy of the axolotl. New York: Oxford University Press. Arnauld, E., and R. Cambar. 1970. Nouvelles observations sur les modalits d'induction de Ia morphogense du msonphros chez Ia larve de Ia grenouille agile (Rana dalmatina). C. R. Hebd. SeancesAcad. Sci. Paris 270D:2563-2565. Arnold, E. N. 1994. Do ecological analogues assemble their cornmon features in the same order?An investigation of regularities in evolution, using sand-dweUing lizards as examples. Philosoph. Trans.Roy. Soc. London 344:277-290. Arnold, E. N., and J. A. Burton. 1985. A f i l d ~ u i d e the reptiles to and amphibians ofBritain and Europe. Glasgow: Wiiam Coliins Sons & Co. Arnold, S. J., and R. J. Wassersug. 1978. Differential predation on metamorphic anurans by garter snakes (ThamnUphis): Social behavior as a possible defense. Ecology 59:1014-1022. Arnoult, J. 1965. Contnbution a I'tude des batraciens de Madagascar. Ecologie et dveloppement de Mantella aurantiaca Mocquard, 1900. BuU. Mus. Natl. dJHist. Nat. 37:93 1-940. Arnouit, J., and M. Lamotte. 1968. Les Pipidae de l'Ouest africain et du Cameroun. Bull. Inst. France Aj-zque Noire 30A: 270-306. Arnouit, J., and M. Razarihelisoa. 1967. Contribution a I'tude des batraciens de Madagascar le genre&ntidactyIus aduites et formes larvaires de M . betsikanus (Blgr.),M . curtzts (Blgr.) et M . alutus (Peracca).Bull. Mus. Natl. dJHist. Nat. 39:471487. Arrayago, M.-J., A. Bea, and B. Heulin. 1996. Hybridization

352

LITERATURE CITED

experiments between oviparous and viviparous strains of Lmerta vivipara: A new insight into the evolution of viviparity in reptiles. Herpetologica 52:33-342. Arthur, W. 1982. A developmental approach to the problem of variation in evolutionary rates. Bwl. J. Linnean Soc. 18:243-261. Aschoff, L. 1924. Das ReticuloendothelialeSystem. Ergebn. Innere Med. Ktnderheilkde. 26: 1-1 18. Ashby, K. R. 1969. The popuiation ecology of a selfmaintaining colony of the common frog (Rana tempuraria). J Zool. 158:453474. Ashley, H., P. Katti, and E. Frieden. 1968. Urea excretion in the buiifrog tadpole: Effect of temperature, metamorphosis, and thyroid hormones. Develop. Bwl. 17:293-307. Atkinson, B. G., and J. J. Just. 1975. Biochemical and histological changes in the respiratory system of Rana cambeiana larvae during normal and induced metamorphosis. Develop. Bwl. 45:151-165. Atkinson, B. G., and G. H. Little. 1972. Growth and regression in tadpole pancreas during spontaneous and thyroid hormone-induced metamorphosis. 1. Rates of macromolecd a r synthesis and degradaaon. Mech. & Develop. 1~299e 312. Atoda, K. 1950. Metamorphosis of the "nonaquatic frog" of the Palau Islands, Western Carolines. PmzfiE Sci. 4:202-207. Atwell, W. J. 1918-1919. The development of the hypophysis of the Anura. Anat. Rec. 15:73-91. Auburn, J. S., and D. H. Taylor. 1979. Polatized light perception and orientation in larval buiifrogs Rana catesbeiana. Anim. Behav. 27:658-668. Audo, M. C., T M. Mann, T. L. Polk, C. M. Loudenslager, W. J. Diehl, and R. Aitig. 1995. Food deprivation during different periods of tadpole (Hyla chysoscelis) ontogeny dects metarnorphic performance difirently. Oecologa 103:518-522. Augert, D., and I? Joly. 1993. Plasticity of age at maturity between two neighbouring popuiations of the common frog (Ranatemporaria L. ). CanadianJ Zool. 71:26-33. . 1994. Dispersal of Rana tempuraria tadpoles in large fishponds. Alpes 12:31-40. Ax, I? 1984. Dasphylognetische *em: Systematisierung der iebenden Natur au&rund ihrer Phylogenese. Stuttgart: Gustav Fischer-Verlag. Axelsson., E.., E Nustrom. and C. Bronmark. 1997. Cravfish ! predation on amphibik eggs and larvae. ~m~hibia-&tilia 18:217-228. Azevedo-Ramos, C., M. Van Sluys, J.-M. Hero, and W. E. Magnusson. 1992. Influence of tadpole movements on predation by odonate naiad5.J Herpetol. 26: 335-338. Babalian, A. L., and A. I. Shapovalov. 1984. Synaptic actions produced by individual ventrolateral tract fibres in frog lumbar motoneurons. Exp. Brain Res. 54:551-563. Babbitt, K. J., and G. W. Tanner. 1997. Effects of cover and predator identity on predation of Hyla sguirella tadpoles. J. Herpetol. 31: 128-130. Babbitt, K. J., and F. Jordan. 1996. Predation on Bufo t m e k tadpoles: Effects of cover and predator density. Cupeia 1996:485488. Baccari, G. C., S. Minucci, and L. Dimatteo. 1990. Organogenesis of the Harderian gland in Rana escuienta and BuJvindZr. Boll. Zool. 57:221-224. Bachrnann, K. 1969. Temperature adapmtions of amphibian embryos. AmericanNat. 103:115-130. Baculi, B. S., and E. L. Cooper. 1968. Lymphomyeloid organs

V of Amphibia. I Normal histology in larval and adult Ranu cate5beiana.J Morphol. 126:463476. Bagenal, T. B. 1967. A method of m a r h g fish eggs and larvae. Nature 214: 113. Bagnara, J. T. 1965. Pineal regulation of body blanching in amphibian larvae. Progr Brain Res. 10:489-506. . 1972. Interrelationships of melanophores, iridophores, and xanthophores. pp. 171-180. In P&mentatwn:Itsgenesis and bwlogic control, edited by V. Riley. New York: AppletonCentury-Crofts. . 1976. Color change. pp. 1-52. In Physwhgy oftheAmphibia. Vol. 3, edited by B. Lofts. New York: Academic Press. Bagnara, J. T., S. K. Frost, and J. Matsumoto. 1978. On the development of pigment patterns in amphibians. Ammicun Zool. 18:301-312. Bagnara, J. T., J. Matsumoto, W. Ferris, S. K. Frost, W A. Turner, Jr., T. T. Tchen, and J. D. Taylor. 1979. Common origin of pigment cells. Science 203 :410415. Bailey, C. H., and I? Gouras. 1985. The retina and phototransduction. pp. 344-355. In Principies of neural science. 2d ed., edited by E. R. Kandel and J. H. Schwartz. New York: Elsevier. Baker, C. L. 1949. The comparative anatomy of the aortic arches of the urodeles and their relation to respiration and degree of metamorphosis.J TennesseeAcad. Sci. 24: 1 2 4 0 . Baker, G. C., and T. J. C. Beebee. 1997. Microenvironmental effects on competition between Rana and BuJ larvae, and on the abundante of Prototheca richardsi, in s m d fish-ponds. HcQMo~.J7: 149-1 54. Balinsky, B. I. 1972. The ine structure of the amphibian limb bud. Acta Embryol. Ikp.Suppl. 1972:455-470. . 1981. An intmductwn to embyology. 5th ed. New York: W. B. Saunders. Balinsb J. B. 1970. Nitrogen metabolism in amphibians. pp. 519-624. In Comparative bwchemisty ofnitrgen metabolism, edited by J. W. Campbell. New York: Academic Press. Balinsky. J. B., E. L. Choritz, C. G. L. Coe, and G. S. Van Der Schans. 1967. Urea cycle enzyrnes and urea excretion d u ~ g the development and metamorphosis ofXempuslamk. Gmp. Bwchem. Physwl. 22A:53-57. Balinsky, J. B., T. L. Coetzer, and F. J. Mattheyse. 1972. The effect of thyroxine and hypertonic environment on the enzymes of the urea cycle in Xenopus ihmi.Comp. Bwchem. Phys~1.43B:83-95. Balinsky, J. B., and R. J. Devis. 1963. Origin and differentiation of cytoplasrnic structures in the oocytes of Xempzrs Luvis. Acta Embryol. Morphol. Ikp.655-108. Banks, B., and T. J. C. Beebee. 1987. Spawn predation and larvai growth inhibition as mechanisms for niche separation in anurans. Oecologia 72:569-573. Bantle, J. A., J. N. Dumont, R. A. Finch, and G. Linder. 1990. Atlas of abnormalities: A guide for the performance of FETAX. United States Amzy Med. Res. Develop. Command P~bl. DAMD17-88-C-803 1. Barch, S. H., J. R. Shaver, and G. B. Wilson. 1966. An electron microscopic study of the nephric'unit in the frog. Trans. American Micros. Soc. 85: 350-359. Bargmann, W. 1937. Uber sezernierende Zellelemente im Nephron von Xenoptrs l&. Zeit. ZelIfmsch. Mikros. Anat. 25 :764-768. . 1978. Niere und ableitende Harnwege. pp. 14-44.In Handbuch der mikroskoptichenAnatomie de5 Menschen. Vol. 7, edited by W. Bargmann. New York: Springer-Verlag. Bargmann. W., A. Knoop, and T. H. Schiebler. 1955. Histolog-

LITERATURE CITED ische, cytochemische und elektronenmikroskopische Untersuchungen am Nephron (mit Bercksichtigung der Mitochondrien).Zeit. Zellfmsch.Mikros. Anat. 42:386-422. Barker, J., and G. C. Grigg. 1977.AfiIdgude toAush.alian@gs. Adelaide, Australia: Rigby Ltd. Biirlcher, F., R. J. Mackay, and G. B. Wiggins. 1978. Detritus processing in a temporaty vernal pool in southern Ontario. Arch. Hydrobwl. 81:269-295. Barnes, M. D., and K. E. Mey. 1983. Maturation and recycling of trigeminal motoneurons in anuran lamae. J. Comp. Nearol. 218:406414. Barreto, L., and G. Moriera. 1996. Seasonal variation in age structure and spatial distnbution of a savanna larvai anuran assemblage in central Brazi1.J. Helpetol. 30:87-92. Barrington, E. J. W. 1942. Gastric digestion in the lower vertebrates. Biol. R v 17:1-27. e. . 1946. The delayed development of the stomach in the frog (&nu temporaria) and the toad (Bu. bufo). Proc. Zool. Soc.Londan 116:l-21. . 1957. The alimentary canal and digestion. In Thephysiology $f"hes. Vol. 2, Metabolim. Edited by M. Brown. New York: Academic Press. Barrio, A. 1953. Sistemtica, morfologia y reproduccin de Physahmus henselii (Peters) y Pseudopaludicola falczipes (Hensel)(Anura, Leptodactylidae).Physzs 20: 379-389. . 1976. Obsemaciones etoecologicas sobre Hylorina sylvatica Beii (Anura, Leptodactylidae).PhysZr 27: 153-157. Bany, T. H. 1956. The ontogenesis of the sound-conducting apparatus of Bafo angusticeps Smith. Molphol.Jahrb. 97:477544. Barandun, J., and H.-U. Reyer. 1997. Reproductive ecology of Bombina variegata: Development of eggs and larvae. J. Herpetol. 31:107-110. Bardsley, L., and T. J. C. Beebee. 1998. Interspecific competition between Bufo larvae under conditions of cornrnunity transition. Eco& ,79:1751. Bqoglu, M., and N. Ozeti. 1973. Turkiye amphibileri. Ege Univ m i t e Fen Fakiiltesi KZtaplar Seriri (50):1-155. Bateson, P. 1983. Optimal outbreeding. pp. 257-277. In Mate choice, edited by P. Bateson. Cambridge: Cambridge University Press. Battaglini, P., and P. Boni. 1967. Indigenous microbial flora and the large intestine in tadpoles. Experientia 23:950-951. Batten, T. F. C., and P. M. Ingleton. 1987. The hypothalamus and pituitaty gland. pp. 285-412. In Fundamentais o wmf parative vertebrate endocrinology, edited by I. Chester-Jones, P. M. Ingleton, and J. G. Phillips. New York: Plenum. Bavay, A. 1873. Notes sur I'Hylodes martinicensis Tschudi et ses metamorphosis. Rev.Sci. Nat. 1:281-290. Beach, D. H., and M. Jacobson. 1979a. Patterns of ceii proliferation in the retina of the clawed frog during deve1opment.J Comp. Neurol. 183:603-614. . 1979b. Influentes of thyroxine on ceii proliferation in the retina of the clawed frog at different ages. J. Comp. Nearol. 183:615-624. Beachy, C. K. 1995. Eficts of larvai growth history on metamorphosis in a stream-dwelling salamander (Desmgnathw ochmphaeus).J. Helpetol. 29: 375-382. Bearns, H. W, and R. G. Kessel. 1974. The problern of germ ceii determinants.Intl. R v Cytol. 39:413479. e. Beaumont, A. 1970. Apparition des ceiiules insulaires a dans le pancras lamaire d'un Amphibien Anoure: Discoglosaspictus Otth. C. R. Hebd. SeancesAcad. Sci. Paris 271D:1104-1106. Beaumont, A., and J. Deu&. 1958. La kratinisation des dents

353

et du bec larvaires chez Alytes o b s t h n s Laurenti. Arch. Anat. Mimos. Molphol. Exp. 48:307-324. Beck, C. W 1997. Effea of changes in resource levei on age and size at metamorphosis in Hyla squzrella. Oewlogia 112:187-192. . 1998. Mode of fertilization and parental care in anurans. Anim. Behav. 55 :439-449. Beckenbach, A. T. 1975. Infiuence of body size and temperature on the critical oxygen tension of some plethodontid salamanders. Physwl. Zool. 48:338-347. Beddard, F. E. 1896. On the diaphragm and on the muscular anatomy ofXen+sJ with remarks on its f i t i e s . Proc. Zool. Soc. London 53:841-850. Beebee, T. J. C. 1991. Purification of an agent causing growth inhibition in anuran larvae and its identilication as a uniceiiular unpigmented alga. CanadianJ Zool. 69:2146-2153. . 1995. Tadpole growth: 1s there an interference effea in nature? Herpet01.J. 5:204-205. Beebee, T. J. C., and J. R. Griffin. 1977. A preliminaty investigation into natterjack toad ( B u . calamita) breeding site characteristics in Britain.J. Zool. 181:341-350. Beebee, T. J. C., and A. L.-C. Wong. 1992a. Leucine uptake by enterobacterial and algal members of larval anuran gut flora. Comp. Bwchem. Physwl. 101B:527-530. . 1992b. Prototheca-mediated interference competition between anuran larvae operates by resource diversion. Physiol. 2501. 65:815-831. Beigl, I. 1989. Untersuchung der Urogenitalverbindungen anurer Entwicklungsgeschichte des Amphibienherzens. Morphol.Jahrb. 51:355412. Beiswenger, R. E. 1972. Aggregative behavior of tadpoles of the American toad, Bufo americanas, in Michigan. Dissertation, University of Michigan. . 1975. Structure and function in aggregations of tadpoles of the American toad, Bufo americanas. Herpetolagica 31:222-233. . 1977. Diel patterns of aggregative behavior in tadpoles of B a . americanas, in relation to light and temperature.Emlogy 58:98-108. . 1978. Responses of B u . tadpoles (Amphibia, Anura, Bufonidae) to laboratory gradients of temperature. J. Herpetol. 12:499-504. Beiswenger, R. E., and F. H. Test. 1966. Effects of environmental temperature on rnovements of tadpoles of the American toad, Bu) twrestris americunzw. Pap. MiEhigan Acad. Sci., Arts Letten 51:127-141. Bell, B. D. 1978. Obsemations on the ecoiogy and reproduction of the New Zeaiand leiopelmid frogs. Helpetologiea 34: 340-354. . 1985. Development and parental-care in the endemic New Zealand frogs. pp. 269-278. In Bwlogy ofAastralian @gs and repties, edited by G. Grigg, R. Shine, and H. Ehmann. Chipping Norton, New South Wales: Surrey Beatty. Beiio y Espinosa, D. 1871. Zoologiche Notizen aus Puerto Rico, nach dem Spanischen frei bearbeitel von Herrn E. von Martens in Berlin. Zool. Garten (Frankfurt a. M.) 12:348-351. Belova, Z. A. 1964. On the feeding habits of the tadpoles of &na dibunda P d . in the Volga Delta. Zool. Zhar: 453:1188-1192. Benbassat, J. 1970. Erythroid cell development during natural amphibian metamorphosis. Deve@. Bwl. 21 :557-583. Bennett, A. F., and W. R. Dawson. 1976. Metabolism. pp. 127223. In Bwlogy o the Reptilia. Vol. 5. PhyswlogyA, edited by f C. Gans and W R. Dawson. New York: Academic Press.

354

LITERATURE CITED notypes in the housefly at varying densities and proportions (the green strain). Evolutwn 18:312-330. Binckiey, C. A., B. Plesky, K. Werner, and S. Droege. 1998. Using the Visible Implant Flourescent Elastomer (VIE) tagging system to mark amphibians. <http://www.mplpwrc. usgs.gov/marking/vie. html> . Binkley, S., K. Mosher, F. Rubin, and B. White. 1988.Xen~pus tadpole melanophores are controlled by dark and light melatonin without influente of time of day.J.PinealRes. 5:87-98. Birks, R., H. E. Huxley, and B. Katz. 1960. The fine structure of the neuromuscular junction of the frog. J. Physiol. 150:134-144. Bishop, J. E. 1973. Limnulogy o a smaU Mahyan river Sungai f Gombak. The Hague: Dr. W. Junk. Bixby, J. L., and N. C. Spimr. 1982. The appearance and development of chemosensitivity in Rohon-Beard neurons of the Xempus spinal c0rd.J. Physwl. 330:513-536. Black, D. 1917. The motor nuclei of the cerebral nerves in phylogeny. A study of the phenomena of neurotaxis. J. Cmp. Neurol. 28:379-427. Black, J. H. 1971. The formation of tadpoles holes. Herpetol. Rev. 3:7. . 1974. Tadpoles nests in Okiahoma. Okiuboma Geol. Notes (3):105. Blackburn, D. G. 1994. Discrepant usage of the term "ovoviviparity " in the herpetologicai literature. Herpetol.J. 4:65-72. Blackler, A. W. 1970. The integrity of the reproductive cell iine in the Amphibia. Cun: Top. Deveiup. Biol. 5:71-88. Blair, W. F. 1961. Calling and spawning seasons in a mixed population of anurans. Ewlogy 42:99-110. Blaustein, A. R. 1988. Ecologicai correlates and potentiai h c tions of kin recognition and kin association in anuran larvae. Behav. Genet. 18:449-464. Blaustein, A. R., M. Bekoff, J. A. Byers, and T. J. Daniels. 1991. Kin recognition in vertebrates: What do we reay know about adaptive vaiue?Anim. Behav. 41 :1079-1083. Blaustein, A. R., M. Bekoff, and T. J. Daniels. 1987a. Kin recognition in vertebrates (excluding primates): Empirical evidente. pp. 287-331. In Kin rewgnition in animals, edited by D. J. C. Fletcher and C. D. Michenei: New York: Wdey. . 1987b. Kin recognition in vertebrates (excluding primates): Mechanisms, fnctions, and f u m e research. pp. 333-357. In Kin recognitwn in animals, edited by D. J. C. Fletcher and C. D. Michener New York: Wiley. Blaustein, A. R., K. S. Chang, H. G. Lefcort, and R. K. O'Hara. 1990. Toad tadpole kin recognition: Recognition of haif siblings and the role of maternal cues. Ethol. Ecol. Eml. 2:215-226. Blaustein, A. R., and R. K. O'Hara. 1981. Genetic control for sibiing recognition. Nature 290:246-248. . 1982a. Kin recognition in Rana cas& tadpoles: Maternal and paternal effects.Anim. Behav. 30: 1151-1 157. . 1982b. Kin recognition cues in Rana m& tadpoles. Behav. Neural Bwl. 36:77-87. . 1983. Kin recognition in Rana m& tadpoles: Effects of rearing with nonsiblings and varying the strength of the stimulus cues. Behav. Neural Bwl. 39:259-267. . 1986a. An investigation of kin recognition in redlegged frog (Rana aurora) tadpoles.J. Zool. 209: 347-3 53. . 1986b. Kin recognition in tadpoles. Sci. Americati 254: 108-1 16. . 1987. Aggregation behaviour in Rana m u u tadcde poles: Association preferences among wiid aggregations and responses to non-kin. Anim. Behav. 35: 1549-1555.

Bennett, A. F., and P. Licht. 1974. Anaerobic metabolism during activity in amphibians. Comp. Biochem. Physiol. 48A:319-327. Benninghoff, A. 1921. Beitrage zur vergleichenden Anatomie und Entwickiungsgeschichte des Amphibienherzens. Morphol. Jabrb. 51:355412. Benson, W W. 1971. Evidence for the evolution of unpalatability through kin selection in the Heliconiiae (Lepidoptera). AmericanNat. 105:213-226. Bentiey, P. J. 1971. Enhm'nes and osmoregzrlatwn.A wmparative acwunt ofthe reguhtwn ofwatm and salt in vevtebrates. Vol. 1. Zoophysiology and Ecology. New York: Springer-Verlag. Bentley, P. J., and G. P. Baldwin. 1980. Comparison of transcutaneous permeability in skins of larvai and adult saiamanders (Ambystomatigrinum).AmericanJ. Physwl. 239R: 5-9. Bentiey, P. J., and D. J. Greenwaid. 1970. Neurohypophysiai fnction in builfrog (Ranacatesbeiana) tadpoles. Gen. Comp. Enhcrinol. 14:412-415. Bernardini, G., P. Podini, R. Maci, and M. Camatini. 1990. Spermiogenesis in Xen~pus s : From late spermatids to W spermatozoa. Mol. Reprod. Deveiup. 26: 347-355. Bernasconi, A. F. 1951.Uber den Ossifikationsmodusbei Xenupus lua& Daud. Denkrchl: Scbweiz. Natu$ Ges. 2: 196-252. Berninghausen, F. 1994. Fe'eld- und Bestimmungsschhsel &r Kaulquappen. 2d ed. Natursditzbund Deutschland. . 1997. WelcheKulgua@e i s t h ? Hanover: NABU. Berns, M. W 1965. Mortality caused by kidney stones in spina&-fed frogs (Ranapiaiens).BwScience 15:297-298. Berry, P. Y 1972. Undescribed and little known tadpoles from . west Malaysia. Herpetologica 28:338-346. Berthold, G. 1980. Microtubules in the epidermal cells of Carausius morosus (Br.). Their pattem and relation to pigment migrati0n.J. Inseet. Physwl. 26:421425. Berton, J.-P. 1964. Anatomie vasculaire du rein de quelques amphibiens urodles et anoures.Ann. Sci. Nat. Zool. 6:229-280. Bertram, B. C. R. 1978. Living in groups: Predators and prey. pp. 64-96. In Behavioural ewlogy: An mlutwnay a p a c h , edited by J. R. Krebs and N. B. Davies. Oxford: Blackwell Sci. Publ. Berven, K. A. 1982. The genetic basis of aititudinal variation in the wood frog Rana ylvatica. 11.An experimental anaiysis of larvai development. Oecologia 52: 360-369. . 1987. The heritable basis of variation in larvai developmental patterns within populations of the wood frog (Rana syivatica).Evolution 41 :1088-1097. - 9 9 0 . Factors affecting population fluctuations in larval 1 and adult stages of the wood frog (Rana ylvatica). E w k y 71: 1599-1608. Berven, K. A., and B. G. Chadra. 1988. The relationship among egg size, density, and food leve1 on larval development in the wood frog (Ranayivatiea). Oecologia 75 :67-72. Berven, K. A., and D. E. Giii. 1983. Interpreting geographic variation in life-history traits. American Zool. 23:85-97. Berven, K. A., D. E. Gi, and S. J. Smith-Giii. 1979. Countergradient selection in the green frog, Rana chmitans. Evolutwn 33:609-623. Bhaduri, J. L. 1953. A study of the urinogenital system of Salientia. Proc. Zool. Soc. Bengal6:1-11l. Bhaduri, J. L., and S. L. Basu. 1957. A study of the urogenital system of Saiientia, Part I. Ranidae and Hyperoliidae of Africa. Ann. Mus. Roy. Congo Bebe, Sci. Zool. 55: 1-56. Bhaduri, J. L., and J. C. Daniel. 1956. The tadpoles of Upmoam globulosum.J. Bombay Nat. Hist. Soc. 53 :713-716. Bhda, S. C., and R. R. Sokai. 1964. Competition among ge-

LITERATURE CITED Bhunein, A. R., R. K. O'Hara, and D. H. Olson. 1984. Kin pdrence behaviour is present after metamorphosis in Rana cauudue fiogs. Anim. Behav. 32:445450. Bhmein, A. R., and B. Waldman. 1992. Kin recognition in anuran amphibian larvae. Anim. Behav. 44:207-221. Btiunein, A. R, and S. C. Waiis. 1995. Aggregation behaviour and kin recognition. pp. 568-602. InAmphibian bwlogy, Vol. 2. Chipping Norton, New South Wales: Surrey Beatty. BLuistein, A. R., T. Yoshikawa, K. Asoh, and S. C. Waiis. 1993. Ontogenetic shifts in tadpole kin recognition- loss of signal and perception.Anim. Behav. 46: 525-538. Bbmein, L., and B. P. Kotler. 1993. Oviposition habitat selection by the mosquito, Culiseta longiareolata: Effects of conspadcs, food, and green toad tadpoles. Ecol. Entowwl. 18:104-108. BLauuein, L., and J. Margaiit. 1994. Mosquito larvae (Culiseta m m i u t a ) prey upon and compete with toad tadpoles :B@ piridis).J. Anim. Eml. 63: 84 1-850. . 1995. Spatial distributions of Culiseta longiareoiuta iculicidae: Diptera) and BufO viridis (Amphibia: Bufoni&e) arnong and within desert pools. J. Arid EnPimn. 2!2:199-211. . 1996. Priority effects in temporacy pools: Nature and ourcome of mosquito larva-toad tadpole interactions depend on order of entrance. J. Anim. Ewl. 65: 77-84. E l u b e y , S. 1958. Cannibaiism in Rana ylvatica tadpoles a weii h a n phenomenon. HmpetologKa 14:34. W E. J. 1905. The life histocy o f X e m p s lamis Daud. Trans. h. Edinburgh 41 :789-821. I E g k - R 1976. The muscular control of vertebrate swim1 ming movements. Bwl. Ra7 52:181-218. . 1978. Golgi-staining of "primary " and "secondacy " motoneurons in the developing spinal cord of an amphibian. J. Cmnp.Neurol. 180:679-690. &amas-Schiosser, R. M. A. 1975. Observations on the iarvd k d o p m e n t of some Malagasy frogs, with notes on their c a h g v and biology (Anura: Dyscophinae, Scaphiophrynime and Cophyiiiae).BeaufOrtia 24:7-26. . 1979a. Biosystematics of the Malagasy fiogs. I. Mand h a e (Ranidae).Beaufaia (352):1-76. . 1979b. Biosystematics of the Malagasy frogs 11. The -U =S Boaphic (Rhacophoridae). Bijd. tot de Dzirkunde 49:261-3 12. . 1981. On endemic Malagasy frogs (Ranidae, Rhaco$midae and Hyperoliidae). Monit. Zool. Italiano (N.S.) 15:217-224. 1 9 8 2 . Observations on the Malagasy frog genus Heterim k Laurent, 1944 (Hyperoliidae).Beaufovria 32: 1-11. Ehmmm-Schiosser, R. M. A., and C. P. Blanc. 1991. Amphib3aa.Farrm- de Madqasuw 75: 1-379, 12 plates. lbmmers-Schlosser, R. M. A., and L. H. M. Blomrners. 1984. Tbc amphibians. pp. 89-104. In ICIy envimnmentsWwas~ p : edited by A. Jolly, P. Oberle and R. Aibinac. New York: Rrpmon. E b i n . Ai. S. 1992a. Comparing bivariate reaction norms m m q three species: Time and size at metamorphosis n three species of Hyiu (Anura: Hylidae). Oecologia 90-2-293. . 1992b. Genetic correlations among morphometric and cates of growth and differentiation in the green aac h g , H+ cinerea. Evolution 46735-744. 1 4J. E. V 1882. Uber den Conus arteriosus und die Arter0 Pcnbogen der Amphibien. Morphol.Jahrb. 7:488-572. kmxight-Horowitz, S. S., and A. M. Simmons. 1997. Tran-

355

w-

sient "deaness" accompanies auditocy development during metamorphosis form tadpole to frog. Pmc. Natl. Acad. Scz. United States ofAmerica 94: 14877-14882. Bogart, J. P. 1981. How many times has terrestrial breedmg evolved in anuran amphibians?Monit. Zool. Italiano (N.S.) Suppl. 15:29-40. Bogert, C. M. 1969. The eggs and hatchiings of the Mexican leptodactylid frog Eleutherodactylus dedeuwatus Taylor. American Mas. Novitates (2376):1-9. Bokermann, W. C. A. 1963. Girinos de an6bios brasileiros-2 (Amphibia, Saiientia). Rev. Brasil. Bwl. 23:349-353. . 1965. Notas sobre as espcies de Thmopa Fitzinger (Amphibia, Leptodactylidae). Anais Acad. Brasil. Cienc. 37:525-537. . 1966. O gnero Phyllodpes Wagler, 1830 (Anura, Hylidae). Anais Acad. Brasil. Cienc. 38 :335-344. . 1974. Observaoes sobre desenvolvimento precoce em Sphaevwrhynchus bmmelicola Bok. 1966 (Anura, Hylidae). Ren Brasil. Biol. 34:35-41. Bolt, J. R. 1977. Dissorophoid relationships and ontogeny, and the origin of the Lissamphibia.J. Paleontol. 51:235-249. . 1979.Amphibamuspandiceps as a juvenile dissorophid: Evidence and implications. pp. 529-563. In Mawn CreekfOssils, edited by M. H. Nitecki. New York: Academic Press. Bond, A. N. 1960. An analysis of the response of salamander gills to changes in the oxygen concentration of the medium. Develop. Bwl. 2: 1-20. Boonkoom, V., and R. H. Aivarado. 1971. Adenosinetriphosphatase activity in gis of larval Rana catesbeiana.Amrican J. Physwl. 220: 1820-1824. Boord, R. L., and L. M. Eisworth. 1972. The central terminal fields of posterior lateral line and eighth nerves ofxenupus. American Zool. 12:727. (Abstr.) Boord, R. L., L. B. Grochow, and L. S. Frishkopf. 1970. Organization of the posterior ramus and ganglion of the eighth cranial nerve of the bulifrog Rana catesbeianu.American Zool. 10:155. (Abstr.) Booth, D. T. 1990. The significance of the boundary layer in cutaneous breathing, aquatic lungless salamanders. Physwlo8b-t33:35 (Abstr.). Boothby, K. M., and A. Roberts. 1995. Effects of site of taaile stimuiation on the escape swimming responses of hatchling Xenupushmis embcy0s.J. Zool. 235:113-125. Bork, T., E. Schabtach, and P Grant. 1987. Factors guiding op. tic fibers in developing Xenupus retina. J. Comp. Neurol. 264: 147-158. Borkhvardt, V. G., and Y. B. Malashichev. 1997. Position of the epicoracoids in arciferal pectoral girdles of the fire-beiiies Bombina (Amphibia: Discoglossidae). Russian J. Herpetol. 4:28-30. Borland, J. R. 1943. The production of experimental goiter in Ranu pipiens tadpoles by cabbage feeding and methyl cyanide.J. Exp. Zool. 94: 115-137, 3 plates. Bossard, H. J. 1971. Experimenteiie Analyse nir Entwicklung der chimien Vorniere bei Triturus und Bombina. Wihelm R o d Arch. Entwicklunpnech. Org. 168:282-303. Boulenger, G. A. 1886. On the reptiles and batrachians of the Solomon Islands. Trans Zool. Soc. London 12:35-62. . 1892 (1891). A synopsis of the tadpoles of the European batrachians. Proc. Zool. Soc. London 1891:593-627, plates 45-47. . 1897-1898. The tailless batrachians of Europe. Parts I and 11. London: Ray Society. . 1919. On Aelurophyrne mammata Gthr., an addition
-

356

LITERATURE CITED gations in tadpoles of SutphiUpus hurteri observed in 1954. WmmannJ.Biol. 15161-68. . 1965. Gnomes of the n%ht: The s p a w o t toads. Philadelphia: University of Pennsylvania Press. . 1968. The formation of feeding schools in tadpoles of spadefoots. WasmannJ.Bwl. 26: 11-16. Bragg, A. N., and W. N. Bragg. 1958. Variations in the mouth parts in tadpoles of Scaphiopus (Spea) bombzJ%ons Cope (Amphibia: Salientia).Southwest. Nat. 3:55-69. Bragg, A. N., and S. Hayes. 1963. A study of labial teeth rows in tadpoles of Couch's spadefoot. Wasmann J. Bwl. 21:149-154. Branch, L. C. 1983. Social behavior of the tadpoles of Phyllomedusa vailianti. Copeia 1983:420428. Brattstrom, E. H. 1962. Thermal control of aggregation behavior in tadpoles. Hevetolofiica 18:38-46. . 1963. A preliminary review of the thermal requirements of arnphibians. Ecology 44:238-255. . 1968. Thermal acclimation in anuran amphibians as a fwiction of latitude and altitude. Comp. Bwchem. Physwl. 24:93-111. . 1970. Amphibia. pp. 135-166. In Comparativephyswlogy of theimoyguiatwn. Vol. I. Invertebrates and nonmammalian vertebrates, edited by G. G. Whittow. New York: Acadeniic Press. . 1979. Amphibian temperature regulation studies in the field and laboratory. American Zool. 19:345-356. Brauer, A. 1898. Ein neuer Fall von Brutpflege bei Froschen. 2001. Jahrb.Abt. Syst., Geogx Bwl. Thiere 12:89-94. Braus, H. 1906. Vordere extremitat und Operculum bie Bombinator-larven. Ein Beitrag zur Kenntnis morphogener correlation und regulation. Sonderabd.Movhol. Jahrb. 35: 139-220. Breden, E, and C. H. Kelly. 1982. The effect of conspecific interactions on metamorphosis in Bufo amnwicunus. Ecology 63: 1682-1689. Breden, E, A. Lum, and R. Wassersug. 1982. Body size and orientation in aggregates of toad tadpolcs Bu$ woodhousei. Copeia 1982:672-680. Bregman, B. S., and W. L. R. Cruce. 1980. Normal dendritic morphology of frog spinal motoneurons: A Golgi study.J. Comp. Neurol. 193:1035-1045. Bresler, J. 1954. The development of labial teeth of salientian larvae in relation to temperature. Copeia 1954:207-211. Bresler, J., and A. N. Bragg. 1954. Variations in the rows of labial teeth in tadpoles. Copeia 1954:255-257. Bridges, C. M., and W. H. N. Gutzke. 1997. Effects of environmental history, sibship, and age on predator-avoidance responses of tadpoles. CanadianJ. Zool, 75 $7-93. Briggs, R., and M. Davidson. 1942. Some effects of spinach feeding on Ranaptpiens tadpoles. J. Exp. 2001. 90:401411. Brink, H. E. 1939. A histological and cytological investigation of the thyroids ofArthroIeptellabicolmvillie~ji Bufo angusand ticeps during normal and accelerated metamorphosis. Proc. Linnean Soc. Lonhn 151: 120-125. Britson, C. A., and R. E. Kissell, Jr. 1996. Effeas of food q.pe on developmental characteristics of an ephemeral pondbreeding anuran, Pseudacris trism'ataferiarztm. Herpetologica 52:374-382. Brock, G. T. 1929. The formation and fate of the operculum and gili chambers in the tadpole of Rana temporaria. Qwrt. J Micros. Sci. 73:335-343. Brockelman, W. Y. 1969. An analysis of density effectsand predation in Bufo americanus tadpoles. Ecology 50:632-644.

to the batrachian fauna of Kashmir. Rec. Indian Mus. 16:469470. Bourges, M., and C. Bachelerie. 1974. Role des ceiiules pithliales adjacentes aux colonnes dentaires dans Ia dynamique des odontoides com& chez le ttard d'Al~esobstetricans Laurenti. C . R. Soc. Bwl. Parir 168:1335-1339. Bourne, G. R. 1997. Reproductive behavior of terrestrial breeding &ogs ~hthero&lus johmtonei in Guyana. J. Herpetol. 31:221-229. . 1998. Am~hisexud~arentalbehavior of a terrestnal breeding frog ~ i u t h w o d b ~ $johnstonei in Guyana. Behav. l~s Ecol. 9:l-7. Bournoure, L. 1937. Le sort de Ia lignc germinale chez Ia grenouilie rousse aprs i'action des rayons ultra-violets sur le ple infrieur de I'oeuf. C. R. Hebd. Seances Acad. Sci. Paris 104:1837-1839. Bournoure, L., R. Aubry, and M.-L. Huck. 1954. Nouveiles recherches exprirnentales sur les origines de Ia ligne reproductrice chez la grenouille rousse. J. Embyol. Exp. Mmphol. 2:245-263. Bourquin, 0. 1985. A note on Hemisus mamwratum, conlpiled from the narrative of D. J. Bourquin. Soztth AJCncanJ. Sci. 81:210. Bourret, R. 1942. Les batraciens de I'Indochine. Mem. I&. Oceanagraph.Indochine 6: 1-547. Bowen, S. H . 1980. Detrital nonprotein arnino acids are the in key to rapid g r o ~ n h %pia in Lake Valencia, Venezuela. Science 207:1216-1218. Bradfield, K . S. 1995. Do the ecological requirements of tadpoles always determine their distributions among habitats? Honours thesis, James Cook Universi$ Townsville, Australia. Bradford, D. F. 1983. Wmterkill, oxygen relations, and energy metabolism of a submerged dormant amphibian, Rana muscosa. Ecology 64:1171-1183. . 1984a. Water and osmotic balance in ovenvintering tadpoles and frogs, Rana muscosa. Physiol. Zool. 57:474480. . 1984b. Temperature modulation in a high-elevation amphibian, Rana muscosa. Copeia 1984:966-976. . 1989. Aliotopic distribution of native frogs and introduced fishes in high Sierra Nevada lakes of California: Implications of the negative effect of fish introductions. Copeia 1989:775-778. Bradford, D. F., and R. S. Seymour. 1985. Energy conservation during the delayed-hatchingperiod in the frog Pseuabphye bibroni. Physiol. 2001. :491-496. 58 . 1988a. Iduence of water potential on growth and survival of the embryo, and gas conductance of the egg, in a terrestrial breeding frog, Pseuabphye bibroni. Physwl. Zool. 61 :470474. . 1988b. Iduence of environmental PO, on embryonic oxygen consurnption, rate of development, and hatching in the frog Pseuabphryne bibroni. Physiol. Zool. 61 :475-482. Bradford, D. F., C. Swanson, and M. S. Gordon. 1994. Effects of low pH and aluminum on arnphibians at high elevations in the Sierra Nevada, California. C a n d a n J . Zool. 72: 12721279. Bragg, A. N. 1949. Field preservation of amphibian eggs and larvae. Turtox News 27:262-263. . 1956. Dimorphism and caiinibalism in tadpoles of Scaphiopus bombzfrons (Amphibia, Salientia). Southwest. Nat. 1:105-108. . 1957. Aggregational feeding and metamorphic aggre-

LITERATURE CITED Brodie, E. D., Jr., and D. R. Formanowicz, Jr. 1983. Prey size preference of predators: Differential vulnerability of larvai anurans. Herpetologia 39:67-75. . 1987. Antipredator mechanisms of larvai anurans: Protection of palatable individuais. Herpetologica 43 :369-3 73. Brodie, E. D., Jc, and E. D. Brodie, 111. 1978. The development of noxiousness of Bujb americanus tadpoles to aquatic insect predators. Herpetolo@a 34:302-306. Bronmark, C., and P. Edenharnm. 1994. Does the presence of fish d e c t the distribution of treefrogs (Hyh arborea). Consem. Bwl. 8:841-845. Bronmark, C., S. D. Rundle, and A. Erlandsson. 1991. Interactions between freshwater snails and tadpoles: Competition and facilitation. Oecologia 87: 8-18. Brook, A. J. 1980. The breeding seasons of frogs in Victoria and Tasmania. VictorianNat. 97: 6-1 1. Browder, L. W., C. A. Erickson, and W. R. Jeffrey. 1991. Develc~pmental Bwlogy. 3d ed. New York: W. B. Saunders. Brown, C. A., R. J. Wassersug, and T. Naitoh. 1992. Metamorphic changes in the vagai stimulation of the aiimentary tract in the green frog, Rana clamitans (Ranidae). Acta Anat. 145:340-344. Brown, D., A. Grosso, and R. C. De Sousa. 1981. The amphibian epidermis: Distribution of mitochondria-rich ceiis and the effect of oxytocin.J Cell Sci. 52: 197-213. Brown, D. R., A. W. Everett, and M. R. Bemett. 1989. Compartrnental and topographicai distributions of axons in nerves to the amphibian (Bujb marinus) gluteus muscle. J. Comp.Neurol. 284:231-241. Brown, H. A. 1967. Embryonic temperature adaptations and genetic compatibility in two ailopatric populations of the spadefoot toad, Scapbwpus hammondi. Evolutwn 2 1:742761. . 1969. The heat resistance of some anuran tadpoles (Hylidae and Pelobatidae). Copeia 1969:138-147. . 1990a. Temperature, thyroxine, and induced metamorphosis in tadpoles of the primitive fiog,Ascaphus. Gen. Comp. Endocviml. 79: 136-146. . 1990b. Morphologicai variation and age-class deterrnination in ovenvintering tadpoles of the taiied frog, kcaphus truei.J. Zool. 220:171-184. Brown, L. E., and R. R. Rosati. 1997. Effects of three different diets on survival and growth of larvae of the African clawed frog Xenopus W.Prgl: Fish-Cult. 59:54-58. Brown, R. M., and D. H. Taylor. 1995. Compensatory escape mode trade-offs between swimming performance and maneuvering behavior through larvai ontogeny of the wood frog, Rana sylvatica. Copeia 1995:1-7. Brown, S. C., E. A. Horgan, L. M. Savage, and P. S. Brown. 1986. Changes in body water and plasma constituents during buirog development: Effects of temperature and horm0nes.J. Exp. Zool. 237:25-33. Brown, T. E., A. W. Morley, N. T. Sanderson, and R. D. Tait. 1983. Report of a large fishkiii resulting from natural acid water conditions in Austraiia.J. Fish Bwl. 22: 335-350. Brown, W. C., and A. C. Alcala. 1983. Modes of reproduction of Philippine anurans. pp. 416428. In Advances zn herpetology and evolutiona~ biolo~y, edited by G. G. Whittow. Cambridge, MA: Mus. Comp. Zool. Broyles, R. H. 1981. Changes in the blood during amphibian metamorphosis. pp. 461490. In Metamorphosis: A problem in developmental bwlogy. 2d ed., edited by G. G. Whittow. New York: Plenum.

357

Broyles, R. M., and M. J. Deutsch. 1975. Differentiation of red blood ceiis in vitro. Science 190:471-473. Broyles, R. H., and E. Frieden. 1973. Sites of haemoglobin synthesis in amphibian tadpoles. NatureNew Bwl. 241:207209. Broyles, R. H., G. M. Johnson, P. B. Maples, and G. R. Kindeli. 1981. Two erythropoietic microenvironments and two larvai red ceii lines in bullfrog tadpoles. Develop. Bwl. 81:299-314. Bmt, D. G. 1993. Matemai brood care by Dendrobatespumilw: A frog that feeds its y0ung.J. Herpetol. 27:96-98. Buchan, A. M. J. 1985. Regulatory peptides in the amphibian pancreas. CanadianJ.Zool. 63:2121-2124. Buehr, M. L., and A. W. Blackier. 1970. Sterility and partiai sterility in the South African clawed toad foiiowing the pricking of the eggs.J Embryol. Exp. Morphol. 23:375-384. Buiiock, T. H. 1955. Compensation for temperature in the metabolism and activity of poikilotherms. Bwl. Rev. 30:311342. Bdock, T. H., D. A. Bodznick, and R. G. Northcutt. 1983. The phylogenetic distribution of electroreception: Evidence for convergent evolution of a primitive sense modality. Brain Res. 6:2546. Burd, G. D. 1992. Development of the olfactory nerve in the clawed frog, Xenopus l e : 11. Effects of hypothyroidism.J. Comp.Neur01.315:255-263. Burggren, W W 1982. "Air-gulping" improves blood oxygen transport during aquatic hypoxia in the goldish Carassius auratus. Physwl. Zool. 55 :327-334. . 1984. Transition of respiratory processes during amphibian metamorphosis: From egg to adult. pp. 31-53. In Respiratwn and metabolism ofembyonic vertebrates, edited by G. G. Whittow. Dordrecht: Dr. W Junk. . 1985. Gas exchange, metabolism, and '6entilation" in gelatinous frog egg masses. Physwl. Zool. 58:503-5 14. Burggren, W W, and M. Doyle. 1986. Ontogeny of regulation of giii and lung ventilation in the bufrog, Rana catesbeiana. Respil: Physwl. 66:279-291. Burggren, W. W, and M. E. Feder. 1986. Effect of experimental ventilation of the skin on cutaneous gas exchange in the b d fr0g.J. Exp. Bwl. 121:445-449. Burggren, W. W., M. E. Feder, and A. W. Pinder. 1983. Temperature and the balance between aeriai and aquatic respiration in larvae of Rana bevlandieri and Rana catesbeiana. Physiol. Zool. 56:263-273. Burggren, W. W, and R. Fritsche. 1997. Amphibian cardiovascular development. pp. 166-182. In Developmentof cardiovascularsystems.Moiecules to wpnisms, edited by W. W Burggren and B. B. Keiier. Cambridge: Cambridge University Press. Burggren, W. W, R. L. Infantino, Jc, and D. S. Townsend. 1990. Developmentai changes in cardiac and metabolic physiology of the direct-developing tropical frog Eieutheroa'mtyluscogui.J. Exp. Bwl. 152:129-147. Burggren, W. W, and J. J. Just. 1992. Developmental changes in physiological systems. pp. 467-530. In Environmental physiology of the amphibians, edited by M. E. Feder and W. W Burggren. Chicago: University of Chicago Press. Burggren, W, and R. Moalli. 1984. "Active" regulation of cutaneous gas exchange by capiary recruitment in amphibians: Experimental evidence and a revived model for skin respiration. Respil: Physwl. 55:379-392. Burggren, W W, and A. Mwalukoma. 1983. Respiration during chronic hypoxia and hyperoxia in larvai and adult b d -

358

LITERATURE CITED

frogs (Rana catesbeiana) I. Morphological responses of lungs, skin, and gd1s.J. Exp. Biol. 105:191-203. Burggren, W. W., and N. H. West. 1982. Changing respiratory irnportance of gills, lungs, and skin during rnetamorphosis in the bullfrog Rana catesbeiana.Resp. Physwl. 47: 151-164. Bums, R. K. 1955. Urinogenital systern. pp. 462491. InAnalysis of deveupment, edited by B. H. Wiiiier, l? A. Weiss, and V. Hamburge~ Philadelphia: W. B. Saunders. Burton, l? R. 1985. Ultrastructure of the olfactory neuron of the bulifrog: The dendrite and its rnicrotubules. J. Comp. Neurol. 242: 147-160. Busk, M., E. H. Larsen, and F. B. Jensen. 1997. Acid-base regulation in tadpoles of Rana catesbeiana exposed to environmental hypercapnia.J. Exp. Bwl. 200:2507-25 12. Byrd, C. A., and G. D. Burd. 1991. Developrnent of the olfactory bulb in the clawed frog,Xenopus W s : A rnorphological and quantitative analysis.J. Comp.Neurol. 3 14:79-90. Bytinski-Salz, H., and H . Elias. 1938. Studi sui crornatofori dei Discolossidae. I. Melanofori paraepiderernici di Diswglossus pictus (Amphibia, Anura). Arch. Italiano Anat. EmbrMl. 40: 1-36. Cadle, J. E., and R. Altig. 1991. Two lotic tadpoles frorn the Andes of southern Peru: Hyh amzata and Bufo veraguensis, with notes on the cali of H y h armuta (Amphibia: Anura: Hylidae and Bufonidae). Neotmp. FaunaEnviron. 26:45-53. Caldwell, J. l? 1974. A re-evaluationof the Hyla bistincta species group, with descriptions of three new species (Anura: Hylidae). Occm.Pap. Mus. Nat. H&.., Uniu.Kansm (28):1-37. . 1982. Disruptive seleaion: A tail color polymorphisrn in Acris tadpoles in response to diierential predation. CanadianJ.Zool. 60:2818-2827. . 1986. A description of the tadpole of Hylasmithii with comments on tail coloration. Copeia 1986:1004-1006. . 1989. Structure and behavior of Hyhgeographica tadpole schools with comments on classification of group behavior in tadpoles. Copeia 1989:938-948. . 1993. Brazil nut fmit capsules as phytotelmata: Interactions among anuran and insea lmae. Canadian J. Zool. 71: 1193-1201. . 1994. Natural history and survival of eggs and early larval stages of Agalychnis calcarzjier (Anura: Hylidae). Herpetol. Nat. Hist. 2:57-66. Caldwell, J. P., and M. C. de Arajo. 1998. Cannibalistic interactions resulting frorn indiscriminate predatory behavior in tadpoles of poison frogs (Anura: Denbrodatidae).Bwtropica 30:92-103. Caldwell, J. E, and E T. Lopez. 1989. Foam-generating behavior in tadpoles of Leptodactylus mystaceus. Copeia 1989:498502. Caldwell, J. E, J. H . Thorp, and T. O. Jervey. 1980. Predatorprey relationships among l m a l dragonflies, salamanders, and frogs. Oewloga 46:285-289. Calef, G. W. 1973. Natural rnortality of tadpoles in a population of Rana aurora. Ewlogy 54:741-758. Caliery, E. M., and R. E Elinson. 1996. Developrnental regulation of the urea-cycle enzyrne arginase in the direct developing fiog Eleuthevodactylus ~0qui.J. Zool. 275:61-66. Exp. Cambar, R. 1947a. Dbut du fonaionnernent du pronphros chez l'ernbryon de grenouille. C. R. Seances Soc. Biol. FilosOf: 141:752-754. . 194%. Destine6 du canal de WoM aprs ablation prcoce du pronphros chez les lmes d'amphibiens anoures. C. R. Hebd. Seanm A d . Sci. Paris 225 :644-646.

. 1948. Recherches exprirnentales sur les faaeurs de Ia rnorphognse du rnsonphros chez les amphibiens anoures. Bull. Bwl. France-Belgique 823216285. . 1952a. L'uretre primaire de la grenouiile se dveloppe, en direaion postrieure, partir d'un blastrne voisin du pronphros; drnonstration exprimentale. C. R. Seanm Soc. Bwl. Filosof: 146:77-80. . 1952b. Essai exprimental de dveloppernent de I'uretre primaire des amphibiens anoures en milieu atypique. C. R. Seances Soc. Bwl. Filosof: 146:453-455. . 1954. Influente de fonaionnernent du pronphros sur la rnorphognse de I'uretre primaire chez l'ernbryon de grenouille. C. R. Hebd. Seances Acad. Sci. Paris 238:21912192. Cambar, R., and E. Arnauld. 1970. Les rnodalits d'acroissernent du rnsonphros chez Ia l m e de grenouiile agile (Rana dalmatina). C. R. Hebd. SeancesAcad. Sci. Paris 270D: 17051707. Cambar, R., and J.-D. Gipouloux. 1956a. Table cronologique du dveloppernent ernbryonaire et larvaire du crapaud cornrnun, Bufo bufo. Bull. Bwl. France-Belgique 90: 198-217. . 1956b. Mesure exprimentale de Ia distance laquelle I'uretre primaire peut induire la rnorphognse du rnsonphros (amphibiens anoures). C.R. Hebd. SeancesAd. Sci. Paris 242:2862-2865. . 1956c. Les blastrnes du rnsonphros possdent une raaivit durable l'aaion inductrice exerce par l'uretre prirnaire; drnonstration exprimentale chez les amphibiens anoures. C.R Hebd. SeancesAcad. Sci. Paris 242:2992-2994. . 1970a. Les faaeurs dterrninant la rnigration des cellules de I'uretre primaire sont spcifiques; drnonstration exprimentale chez les Amphibiens Anoures. C.R Hebd. Seances A d . Sci. Paris 270D: 1359-1361. . 1970b. L'aaion inductrice exerce ar l'uretre p rima ire sur la rnorphognse du rnsonphroi est iniersp2cifique; drnonstration exprimentale chez les Amphibiens Anoures. C.R. Hebd. Seances A d . Sci. Paris 270D: 1607-1609. . 1973. Recherches exprimentales sur Ia localisation et Ia nature des facteurs dterrninant la rnigration des cellules de l'uretre prirnaire chez Ia grenouiile agile, Rana dulmatina Bon. (Amphibien Anoure). Ann. Embryol. Molphol. 6:309-316. Cambar, R., and J. Garcin. 1962. Rgenration de l'uretre chez Ia grenouiile agile (Ranadalmatina Bon.). C.R Hebd. Seanm Acad. Sci. Paris 254:3570-3572. Cambar, R., and B. Marrot. 1954. Table chronologique du dveloppernent de la grenouiile agrle (Rana dalmatina Bon.). Bull. Biol. France-Belgique 88:168-177. Cambar, R., and M. S. Martin. 1959. Table chronologique du dveloppernent ernbryonnaire et larvaire du crapaud accoucheur (Alyts obstetricans Laur.). Actes Soc. Linneenne Bordeaux 98: 3-79. Camerano, L. 1890. Richerche intorno alio sviluppo ed alie cause de1polimorisrno dei girini degli Anfibi Anuri.AttiAcd .T&no 26:3-14. . 1892. Nuove richerche intorno alio sviluppo ed alie cause de1 polirnorisrno dei girini Anfibi Anuri. 1 . Azione 1 deiia luce. Am'Accad. Torino 28:2-17. Campbell, F. R. 1970. Ultrastructure of the bone rnarrow of the frog. AmericanJ.Anat. 129:329-356. Campbell, J. A., and B. T. Clarke. 1998. A review of frogs of the genus Otuphryne (Microhylidae) with a description of a new species. Herpetologca 54:30 1-3 17.

LITERATURE CITED

359

Campbell, J. A. and T M. Kubin. 1990. A key to the larvae of Xenops W: Morphological events and evidence for a hatching enzyme. Develop. BBi. 38:1-13. PItmoh>.la (Hylidae), with a description of the tadpole prem e d to be Piectrohyla avia. Southest. Nat. 35:91-94. Carroll, E. J., Jr., A. M. Senevirame, and R. Ruibal. 1991. GasCampbell, J. A., and E. N. Srnith. 1992. A new frog of the getric pepsin in an anuran larvae. Develop. Gowth Dzfe"er. nus Rychohyla (Hylidae) frorn the Sierra de Santa Cruz, Gua33:499-507. ternala, and description of a new genus of Middle Arnerican Carver, V . H., and E. Frieden. 1977. Gut regression during stream-breedingtreefrogs. Herpetologica 48:153-167. spontaneous and triiodothyronine induced rnetamorphosis Campeny, R., and A. Casinos. 1989. Densities and buoyancy in in Rana catesbeiana tadpoles. Gen. Comp.Endomcnnol. 31:202tadpoles of rnidwife toad, Alytes obstetrians. Zool. Anz. 207. 223:6-12. Castanet, J., and M. H. Caetano. 1995. Iduence of habitat Candelas, G. C., and M. Gornez. 1963. Nitrogen excretion in condition on skeletal rnass and bone smiaure of anuran amtadpoles of .ptoakcty1us albilabris and Rana catesbeiana. phibians. CanadianJ.Zool. 73:234-242. AmerKan Zwl. 3:521-522. (Abstr.) Casteil, L. L. and R. Mann. 1994. Optimal staining of lipids in Cannateila, D.C. 1985. A phylogeny of prirnitive frogs (archaebivalve larvae with Nile Red. Aqumulture 119:89-100. obatrachians). Dissertation, University of Kansas. Casterlin, M. E., and W W Reynolds. 1977. Behavioral fever Cannateila, D.C., and L. Trueb. 1988a. Evolution of pipoid in anuran amphibian larvae. Lije Sci. 20:593-596. fiogs: Intergeneric relationships of the aquatic frog farnily . 1979. Behavioural thermoregulation in Rana ppiens Pipidae (Anura). Zoo1.J.Linnean Soc. 94:1-38. tadpoles. J. Thenn.Bwl. 3:143-145. . 1988b. Evolution of pipoid frogs: Morphology and Caston, J., and A. Bricout-Berthout. 1985. Influente of stimulaphylogenetic relationships of Pseudbymenochims. J. Herpetol. tion of the visual system on the activity of vestibular nuclear 22:439456. neurons in the frog. Brain Behav. Evol. 26:49-57. Capranica, R. R. 1976. Morphology and physiology of the au- Catton, W.T. 1976. Cutaneous rnechanoreceptors. pp. 629ditory systern. pp. 55 1-575. In Frg neurobiolo~y, handbook, a 642. In F r g neurobiolo~y, handbook, edited by R. Llins and a edited by R. Llins and W Precht. Beriin: Springer-Verlag. W. Precht. Beriin: Springer-Verlag. Capranica, R R., and A. J. M. Moffat. 1974. Frequency sensi- Cecil, S. C., and J. J. Just. 1979. Survival rate, population dentivity of auditory fibers in the eighth nerve of the spadefoot sity, and developrnent of a naturally occurring anuran larvae toad, Scaphwpus cottchi.J. Acoust. Soc. America Suppl. 55-85. (Rana catesbeiana).Copeia 1979:447-45 3. Caramaschi, U. 1979. O girino de Oantophlynw carnalboi Sav- Cei, J. M. 1962. B a ~ m w de Chih. Santiago: Edic. Univ. Chile. s age & Cei, 1965 (Amphibia, Anura, Ceratophrydidae). h . . 1964. The tadpole of Ba@mhuphrynuspatagonicusGalB d . BMl. 39:169-171. lardo. Herpetolo~ica 20:242-245. Caramaschi, U., J. Jim, and C. M. de C a ~ a l h o 1980. Obser. . 1980. Arnphibians of Argentina. Monit. Zool. Italiano n e s sobre Aplastodiscuspmviridk A. L u a (Arnphibia, An(N.S.) Monogl: (2):1-609. ura, Hylidae).ReP.BrAFZ1. Bwl. 40:405408. . 1987. Additional notes to "Amphibians of Argentina": Caramaschi, U., and G. Kisteumacher. 1989. O girino de CrossoAn update, 1980-1986. Monit. Zool. Italiano (N.S.)21 :209~ 1 u @mhystomus (Reinhardt e Luetken, 1862)(Anura, s 272. Leptodactylidae).Rev. Brasil. Bwl. 49:237-239. Cei, J. M., and L. Capurro. 1958. Biologia y desarroilo de EupCaramaschi, U., and I. Sazirna. 1984. Uma nova espcies de sophus taeniatus Girard. Invest. Zool. Chiienus (4):159-182. % m p a da Serra do Cip, Minas Gerais, Brasil (Arnphibia, Cei, J. M., and E. Crespo. 1982. Differences in larval morphology of dopatric isolated populations of the Odontuphynus Leptodactylidae).Rev. Brasil. Zool. 2:139-146. . 1985. Uma nova espcie de Crossodaqlus da Serra do ouidentalis group frorn western Argentina. Arq. Mus. Boqe Cip, Minas Gerais, Brasil (Arnphibia, Leptodactylidae). 1A:335-340. RCP. Brasil. 2001. 3:4349. Cei, J. M., and V. G. Roig. 1961. Batracios recolectados por la Cardoso, A. J., and C. F. B. Haddad. 1990. Redescripao e biexpedicin biolgica Erspamer en Comentes y selva oriental ologia de Paratelmatobius~aigeae(Anura, Leptodactylidae). de Misiones. Notas Bwl. Fm. Cien. Ex. Fk. Nat. Univ. Nm. Pup. Aml. Zool. 37:125-132. Nordeste, Cwrintes 1:1 4 0 . Culson, B. M. 1988. Patten'sfiundations of embryology. 5th ed. Chacko, T. 1965. The development and rnetamorphosis of the S m York: McGraw-Hiil. e , hyobranchial skeleton in Rana tigrina, the Indian b d frog. Carmona-Osalde, C., M. A. Olvera-Novoa, A. Flores-Nava, Acta Zool. 46:3 1 1-328. and M. Rodriguez-Serna. 1996. Estirnation of the protein . 1976. Development of the chondrocranium in Rana requirernent for buMrog (Ranacatesbeiana) tadpoles, and its tkrina Daud.J. Zool. Soc. India 28: 103-139. Chang, L.-T., and C.-Y. Hsu. 1987. The relationship benveen dect on rnetamorphosis ratio. A q ~ u l t u r e 141:223-231. Carpenter, C. C. 1953. Aggregation behavior of tadpoles Rana the age and rnetamorphic progress and the development of tadpole ovaries. Proc. Natl. Sci. Counc. Rpub. China p-pretiosa. Herpetolo~ica 77-78. 9: Clrpenter, K. L. 1978. Hernopoiesis in the pronephros ofriana 11B:211-217. ppims. Thesis, Pennsylvania State University. Channing, A. 1978. A new bufonid genus (Arnphibia: Anua) frorn Rhodesia. Herpetologica 34:394-397. Carpenter, K. L., and J. B. Turpen. 1979. Experimental studies on hernopoiesis in the pronephros of Ranapipiens. Dzffwen. 1984. Identification of ghost frog tadpoles. Proc. Ehct7: riation 14:167-174. Mimos. Soc. South A f i m 14:89-90. Cur. K M., and R. Altig. 1991. Oral disc muscles of anuran . 1986. A new species of the genus Stron~ylopus Tschudi tadpo1es.J. Morphol. 208:271-277. from Namaqualand, Cape Province, South Africa (Anura: . 1992. Configurations of the r e m abdominis of anRanidae). Ann. Cape Prw. Mus. 16:127-135. uran tadpo1es.J. Morphol. 214:351-356. . 1993. Observations on the natural history of StephoCurou, E. J., Jr., and J. L. Hedrick. 1974. Hatching in the toad paedes anotis (Bufonidae).J. Herpetol. 27:213-214.

360

LITERATURE CITED

Channing, A., and R. C. Boycott. 1989. A new frog genus and species from the mountains of the southwestern Cape, South Africa (Anura: Ranidae). Copeia 1989:467-471. Channing, A., and M.-D. C. De Capona. 1987. The tadpole of Hperolius mitchelli (Anura: Hyperoliidae).SouthAJncanJ Zool. 22:235-237. Channing, A., R. C. Boycott, and H. J. Hensbergen. 1988. Morphological variation of Heleophyne tadpoles from the Cape Province, South Africa (Anura: Heleophrynidae).J. Zool. 215:205-216. Cherdantseva, E. M., and V. G. Cherdantsev. 1995. The organization of variability in seriaiiy homologous structures and its correlation with the growth rates (on examples of oral fields in the tadpoles of brown frogs). Zool. Zhur. 74:92-107. Chesler, M., and C. Nicholson. 1985. Organizationof the Xum terminale in the fr0g.J. Comp.Neurol. 239:43 1444. Chester-Jones, I. 1987. Structure of the adrenal and interrenal glands. pp. 95-1 24. In Fundumentah of wmparative ve~ebrate enducrinology, edited by I. Chester-Jones, E M. Ingleton, and J. G. Philiips. New York: Plenum. Chibon, E 1960. Dveloppement au laboratoire d'Eleutherodmtylus martinicensis Tschudi, batracien anoure dveloppement direct. Bull. Zool. Frana 85 :412418. . 1962. Diffrenciation sexueiie de Eleutherodaciylus martinicensis Tschudi batracien anoure a dveloppement direa. Bull Zool. Frana 87:509-5 15. Chieffi, G., M. Milone, and L. Iela. 1980. Amphibian reproduction: Reproductive physiology in the male Rana escubnta L. Boll. Zool. 47:63-70. Chieffi, G., R. K. Rastogi, L. Iela, and M. Milone. 1975. The function of fat bodies in relation to the hypothalamohypophyseai-gonadai axis in the frog, Rana esculenta. Cell TZrs.Res. 161:157-165. Chopra, D. P., and J. D. Simnett. 1969a. Changes in mitotic rate during compensatory renal growth in Xempus laevis tadpoles after unilateral pronephrectomy.J. Embyol. Exp. Movphol. 21 :539-548. . 1969b. Demonstration of an organ-specific mitotic inhibitor in amphibian kidney. The effects of adultXenUpus tissue extracts on the mitotic rate of embryonic tissue (in vitro). Exp. CellRes. 58:319-322. . 1970a. Stimulation of mitosis in amphibian kidney by organ specific antiserum. Nature 225:657-658. . 1970b. Stimulation of ceil division in pronephros of embryonic grats foiiowing partial nephrectomy in the host (XempusW s ) . J Embyol. Exp. Morphol. 24: 525-533. . 1971a. Stimulation of ceii division in laNd kidney (Xempus W s ) by rat kidney antiserum. Exp. Cell Res. 64396402. . 1971b. Tissue-specificmitotic inhibition in the kidneys of embryonic grats and partiaiiy nephrectomized host &nopus hmis.J. Embyol. Exp. Morphol. 25: 321-329. Chou, W-H., and J.-Y. Lin. 1997. Tadpoles of Taiwan. Spec. Publ. Nat. Mus. Taiwan (7):1-98. Chovanec, A. 1992. The influente of tadpole swimming behaviour on predation by dragody nymphs. Amphibia-Reptilia 13:341-349. Christman, S. P. 1974. Geographic variation for sait water tolerante in the frog Rana sphenocephala. Copeia 1974:773-778. Ciantar, D. 1983. Models of morphogenesis. A study of the development of the nephric system in Xenopw. Dissertation, University of Southampton. Cioni, C., F. de Palma, and A. Stefaneiii. 1989. Morphology of

afFerent synapses in the Mauthner ceii of IarvalXenopus [aevis. J. Comp.Neurol. 284:205-2 14. Claas, B., B. Fritzsch, and H. Mnz. 1981. Common efferents to lateral line and labyrinthine hair ceiis in aquatic vertebrates. Neurosci. Lett. 27:23 1-235. Clarke, B. T. 1983. A morphological re-examination of the frog genus Nannupbys (Anura: Ranidae) with comments on its biology, distribution and relationships. Zoo1.J Linnean Soc. 79:377-398. Clarke, F. M., A. I. Ward, and J. R. Downie. 1995. Factors afFecting the distribution and status of the golden tree frog, Phyliodytes auratus, in Trinidad. British Herpetol. Soc. Bull. (54):3-9. Claussen, D. L. 1973. The thermal relations of the tailed-frog, Ascaphus truei, and the Pacific treefrog, Hyla +lia. Comp. Bwchem. Physwl. 44A:137-153. Clemens, D. T., and M. E. Feder. 1990. Effects of perfusion and medium-blood PO, difference on oxygen uptake across isolated perfused frog skin. Physwhflist 33 :70 (Abstc). Cline, M. J., and D. W. Golden. 1979. Ceiiular interactions in hematopoiesis. Nature 277: 177-18 1. Cocroft, R. B., and W R. Heyer. 1988. Notes on the frog genus Thoropa (Amphibia: Leptodactylidae) with a description of a new species (Thoropa saxatilis).Proc. Bwl. Soc. Washington 101:209-220. Coggins, L. W. 1973. An ultrastmctural and radioautographic study of early oogenesis in the toad Xenopus laeois.J. Cell Sci. 12:71-93. Cohen, E E 1966. Biochemical aspects of metamorphosis: Transition from ammonotelism to ureotelism. Hamey Lect. 60: 119-154. . 1970. Biochemical differentiation during amphibian metamorphosis. Science 168:533-543. Cohen, P. P., R. F. Bruckner, and S. M. Morris. 1978. Cellular and molecular aspects of thyroid hormone action during amphibian metamorphosis. pp. 273-381. In Homnalproteins andpeptides: Thyroid hmmones. Vol. 6. New York: Academic Press. Collins, J. E 1979. Intrapopulation variation in the body siix at metamorphosis and timing of metamorphosis in the b d frog, Rana catesbeiana.Ewlogy 60: 738-749. Collins, J. P., and M. A. Lewis. 1979. Overwintering tadpoles and breeding season variation in the Ranapipiens complex in Arizona. Southwest. Nat. 24: 371-373. Constantine-Paton,M. 1987. A neural pattern unfolding: Properties of retinotectal differentiation in fiog tadpoles. Neuml. Neurobwl. 44231-253. Constantine-Paton, M., and R. R. Capranica. 1976. Axonal guidance of developing optic nerves in the frog. I. Anatomy of the projection fiom transplanted eye primordia. /. Comp. Neurol. 170:17-3 1. Constantine-Paton, M., E. C. Pitts, and T. A. Reh. 1983. The relationship between retinal axon ingrowth, terminal morphology, and terminal patterning i the optic tectum of the n fr0g.J. Comp.Neurol. 2 18:297-3 13. Cooke, A. S. 1974. DXerential predation by newts on anuran tadpoles. BritishJ.Herpetol. 5 :386-390. Cooper, E. L. 1966. The lymphomyeloid organs of Rana catesbeiana.Anat. Rec. 154:456. (Abstr.) . 1967. Lyrnphomyeloid organs of Amphibia. I. Appearance during larvai and adult stages.J Morphol. 122:38 1-398. . 1976. Immunity mechanisms. pp. 163-272. In Physwlo ~ of the Amphibia. Vol. 3, edited by B. Lofts. New York: v Academic Press.

LITERATURE CITED

361

vai Rana areolata (Anura: Ranidae) by two species of backCooper, E. L., and R. K. Wright. 1976. The anuran amphibian swirnmer (Insecta: Hemiptera: Notonectidae). Herpeto[ogicu spleen. An evolutionary model for terrestrial vertebrates. pp. 42: 171-174. 47-60. In Immum-aspem ofthespleen, edited by J. R. Battisto and J. W. Streilein. Amsterdam: Elsevier/North-Hoiiand Crossland, M. R 1997. Impact of eggs, hatchlings, and tadBiomedical Press. poles of the introduced cane toad B . marinus (Anura: Buu fonidae) on native aquatic fama in northern Queensland, Cooper, K. W. 1936. Demonstration of a hatching secretion in Rana p t p s Schreber. Proc. Natl. Acad. Sci. Unzted States of Australia. Dissertation, James Cook University, Townsville, 22:433-434. Austraiia. Cooper, R S. 1943. An experimental study of the development Crossland, M. R.,, and R. A. Alford. 1998. Evaluation of the of the larval oifactory organ of Ranapipiens Schreber.J. Exp. toxicity of eggs, hatchlings, and tadpoles of the introduced toad B u . marinus (Anura: Bufonidae) to native Australian Zool. 93:415452. Copenhaver, W. M. 1955. Heart, blood vessels, blood, and enaquatic predators.AustralianJ Ewl. 23: 129-137. todermal derivatives. pp. 440461. In Analyzis of helopmnt, Crowder, W. C., M. Nie, and G. R. Ultsch. 1998. Oxygen upedited by B. H. Willier, P. A. Weiss, and V. Hamburger. Philtake in buiifrog tadpoles (Rana catesbeiana). J. Exp. Zml. adelphia: W. B. Saunders. 280:121-134. Corkran, C. C., and C. Thoms. 1996. Amphibians of Orgon, C m p , M . L. 1974. Reproductive strategies in a tropical anuran community. Misc. Publ. Mus. Nat. H&., Univ. Kansas nkbington and British Columbia. Edmonton: Lane Pine (61):l-68. Press. . 1981. Energy accumulation and anlphibian metamorCorless, J. M., and R. D. Fetter. 1987. Structural features of the terminal loop region of frog retinal rod outer segment disk phosis. Oecobia 49: 167-169. . 1983. Oppomuiistic cannibalism by amphibian larvae membranes: 111. Implications of the terminal loop complex for disk morphogenesis, membrane fusion, and celi surface in temporary aquatic environments. Amrican Nat. 121: 28 1-287. interactions.J Comp. Neurol. 257:24-38. . 1984. Ontogenetic changes in vulnerability to preCorless, J. M., R. D. Fetter, and M. J. Costeiio. 1987a. Structurai features of the terminal loop region of frog retinal rod dation in tadpoles of Hyla pseudqpuma. Herpetokgica outer segrnent disk membranes: I. Organization of lipid 40:265-271. components.J. Comp.Neurol. 257: 1-8. . 1986. Cannibalism by younger tadpoles: Another hazard of metamorphosis. Copeia 1986:1007-1009. Coriess, J. M., R D. Fetter, O. B. Zampighi, M. J. Costeiio, and D. L. Wd-Buford. 1987%. Structural features of the termi. 1989a. Effect of habitat drying on developmental time and sue at metamorphosis in Hyla pseuapuma. Copeia nal loop region of frog retinal rod outer segment disk membrana: 11. Organization of the terminal loop complex. J. 1989:794-797. Gmp.Neurol. 257:9-23. . 1989b. Life-history consequences of feeding versus nonfeeding in a facultatively nonfeeding toad larva. OewComeii, T. J., K. A. Berven, and G. J. Gamboa. 1989. Kin recopition by tadpoles and froglets of the wood frog Rana logia 78:486-489. +aticu. Oecokgia 78: 3 12-3 16. . 1990. Possible enhancement of growth in tadpoles through cannibalism. Copeia 1990:560-564. Gmtm-right, S. A. 1988. Intragurld predation and competition: -inanalysis of net growth shifis in larval anuran prey. Cana. 1991. Choice of oviposition site and egg load assessdianJ. Zool. 66: 1813-1821. ment by a treefrog. Herpetologica 47:308-315. . 1992. Cannabilism in amphibians. pp. 256-276. In Com\-right, S. A., and C. E. Nelson. 1990. An examination of rnultiple factors aff'ecting community structure in an aquatic Cannibalkm. Ewlogy and evolutirm amng diverse taxa, edited by M . A. Elgar and B. J. Crapi. Oxford: Oxford University amphibian cornmunity. Oeco[oga 83: 123-131. Cor\1ja, N., and P. d'Ascanio. 1981. Spinal projections from the Press. mesencephalon in the toad. Brain Behav. Evol. 19:205-2 13. C m p , M. L., and M. Vaira. 1991.Vulnerability of Pkwodema Com. H. H. 1967. Avoidance of anoxic water by tadpoles of borelli tadpoles to an avim predator: Effea of body size and Rama temporaria. Hydrobwlogia 30:374-384. density. Herpetologica 47:3 16-32 1. Costa. H. H., and S. Balasubramanian. 1965. The food of the Cuiien, M. J., and H. D. Webster. 1979. Remodelling of optic nerve myelin sheaths an axons during metamorphosis in ndpola of Rhmophorus mctger cmtctger (Blyth). CeylonJ. Sci. 5:105-109, 1plate. Xenupus luatis. J. Comp. Neurol. 184:353-362. Coa-den, R R., B. M. Gebhardt, and E. P. Volpe. 1968. The Culley, D. D., Jr. 1991. Bulifrog culture. pp. 185-205. In Pmhistophysiology of antibody-forming sites in the marine d d n in aquatic anzmals, edited by C. E. Nask Amsterdam: mad. Zeit. Zellfonch.Mikros. Anat. 85 :196-205. Elsevier Sci. Publ. Coa-ies, R. B., and C. M. Bogert. 1944. A preliminary study of Cuiiey, D. D., Jr., S. P. Meyers, and A. J. Doucette, Jr. 1977. A high density rearing system for larval anurans. Lab Anim. the thermal requirements of desert reptiles. Bull. Amrican ,hNat. Hia. 83:261-296. I . 6:34-41. Cor. T. C., and R H . Alvarado. 1979. Electrical and transport Cuiiey, D. D., Jr., and P. K. Sotairidis. 1983. Progress andprobhms associated with bullfiog tadpoh dkts and nutrctrctwn. An3d b c t e r i s t i c s of skin of larval Rana catesbeiana. American]. P+1.237:1274-1279. nual Dr. Schoii Conferente on the hTutritionof Captive Wdd . 1983. Nystatin studies of skin of larval Rana catesbeiAnimais, Chicago. una.A&an J. Physwl. 244R: 58-65. Cummins, C. P. 1989. Interaction between the effects of pH Cta=dmv, I., R. N. Rausch, L. P. W o h u t h , and E. J. Bauer. L. and density on growth and development in Rana temporaria tadpoles. Funct. Ecol. 3:45-52. 1992. Seasonal rhythms of development and temperature selettion in larval bulifrogs, Rana catesbeiana Shaw. Physwl. Cupp, P. V., Jr. 1980. Thermal tolerante of five salientian amZad.65:346-359. phibians during development and metamorphosis. HerpetoCmma J. T., and J. Travis. 1986. Sue-limited predation on larb i c a 36:234-244.

362

LITERATURE CITED Hourdry. 1982. Developrnent of peroxisornes in amphibians. 11. Cytochemicai and biochernicai studies on the liver, kidney and pancreas.J Evp. 2001. :57-65. 223 . 1983. Developrnent of peroxisornes in amphibians. 111. Study on liver, kidney, and intestine during thyroxineinduced rnetamorphosis.J Evp. 2001. 227:413422. Daua M., and J. Hourdry. 1983. Modi6cations du rgime alirnentaire chez les amphibiens anoures en rntamorphose. SOC.2001. Paris 108:409415. Daua M., J. Hourdry, J. S. Hugon, and D. Mnard. 1980. Amphibian intestinal b m h border enzyrnes during thyroxineinduced rnetamorphosis. A biochemicai and cytochemicai study. Histochemisty 70:3 3 4 2 . Davenport, C. B., and W. E. Castle. 1895. Studies on rnorphogenesis. 111. On the acclirnation of organisrns to high ternperature. Arch. EntwiEklungsmech. Mech. Org. 2:227-249. Davidson, E . H. 1986. Gene activiq in early dme&ment. 3d ed. New York: Academic Press. Davies, M. 1989. Ontogeny of bone and the role of heterochrony in the rnyobatrachine genera Upmoieza, Crinia, and Pseudophryne (Anura: Leptodactylidae: Myobatrachinae).J. Mmphol. 200:269-300. . 1991. Descriptions of the tadpoles of some Australian limnodynastine leptodactylid frogs. Trans. Roy. Soc. South Australia 115:67-76. Davies, M., and S. J. Richards. 1990. Developrnental biology of the Australian hylid frog Nyccimystesdayi (Giinther).Trans. Roy. Soc. South Australia 114:207-2 11. Davis, M. R., and M. Constantine-Paton. 1983. Hyperplasia in the spinai sensory system of the fiog. I. Plasticity in the rnost caudal dorsai root gangh0n.J Comp.Neurol. 221 :444-452. Davis, W. M., and R. F. Nunnernacher. 1974. Nerve fibers in the syrnpathetic tnink of Ranapip'ens: A quantitative electron microscope study. Anat. Rec. 179:331-342. Dawson, J. T. 1982. Kin recognition and schooling in the American toad (BuJO a&anus). Dissertation, State University of New York, Albany. De Albuja, C. M., M. Campos, and E. M. De1 Pino. 1983. Role of progesterone in oocyte rnaturation in the egg-brooding hylid frog Gastrotheca Gbambae (Fowler). J Evp. 2001. 227:271-276. De Bavay, J. M. 1993. The developrnental stages of the sphagnum fiog, Kyarranus sphagnicolous Moore (Anura, Myobatrachidae).AustralianJ 2001. 1:151-20 1. 4 De Beer. G. R. 1926. Studies on the vertebrate head. 11. The orbit&ternporal region of the skull. Qmrt. J Micros. SSn. 70:263-370. . 1951. Embryos and ancestm. Oxford: Clarendon Press. . 1985. The dmeiupment of the vertebrate kull. Reprint, Chicago: University of Chicago Press. DeBenedictis, P. A. 1974. Interspecific cornpetition between tadpoles of Ranapipiens and Rama sylaatica: An experimental field study. Ecol. Monogr 44: 129-15 1. Debski, E. A., and M. Constantine-Paton. 1990. Evoked preand postsynaptic activity in the optic tectum of the cannulated tadpo1e.J Comp. Physwl. 167A:377-390. Decker, R. S. 1976. Influente of thyroid horrnones on neuronai death and dzerentiation in larvai Ramapipiens. Deve&. Bwl. 49:lOl-118. Degani, G. 1982. Amphibian tadpole interaction in a winter pond. Hydrobwlogia 96:3-7. . 1986. Growth and behaviour of six species of amphibian larvae in a winter pond in Israel. Hydrobwlogia 1405-10.

Curtis, S. K., R. R. Cowden, and J. W. Nagel. 1979. Ultrastructure of the bone rnarrow of the saiamander Pfethodunglutimsus (Caudata: Plethodontidae).J Morphol. 159:151-160. Cusirnano-Carollo,T. 1963. Investigation on the ability of the neurai folds to induce a rnouth in the Diswglossuspictzts ernbryo. Acta Embryol. Morphol. Evp. 6: 158-168. . 1969. Phenornena of induction by the transverse neural fold during the forrnation of the rnouth in ~coglosuspictus. ActaEmbryol.Evp. 1969:97-110. . 1972. The mechanisrn of the forrnation of the larvai rnouth in DZrco&ssus pictus. Acta Embryol. Evp. 1972:289332. Cusimano-Carollo, T., A. Fagone, and G. Reverbei. 1962. On the origin of the larvai rnouth in the anurans. Acta Embryol. Morphol. Evp. 5:82-103. Czechura, G. V, G. J. Ingram, and D. S. Liern. 1987. The genus Nyctimystes (Anura: Hylidae) in Australia. Rec. Australian Mus. 39:333-338. Cdowska, R. 1972. The fine structure of the "germinai cytoplasrn" in the egg of Xenupus W.WiUielmRodArch. Entwicklungsmech. Org. 169:335-344. Da Cruz, C. A. G. 1980. Descrio de uma nova espcie de Phyllomedusinae do Estado do Esprito Santo, Brasil (Amphibia, Anura, Hylidae). Rev. Brasil. Bwl. 40:683-687. . 1982. Descrio de grupos espcies de Phyllornedusinae Brasileiras com base em caracteres larvrios (Amphibia, i Anura, Hylidae). Arq. Unia. Fed. Rural Ru de Janeiro 5: 147-171. Da Cruz, C. A. G., and 0 . L. Peixoto. 1978. Notas sobre o girino de Darypops schirchi Miranda-Ribeiro (Amphibia,Anura, Microhylidae).Rev.Brasil. Bwl. 38:297-299. . 1980a. Notas sobre o girino de Sphaenmhynchus orophilus (Lutz & Lutz, 1938)(Amphibia, Anura, Hylidae). Rev. Brasil. Bwl. 40:383-386. . 1980b. Observaes sobre a larva de Procmatophrys appendicuiuta (Giinther 1873)(Amphibia, Anura, Leptodactylidae). Rev.Brasil. Bwl. 40:491493. Dainton, B. H. 1988. Cold tolerance and thyroid activity in developing tadpoles of Xevwpus iuemk.J Herpetol. 22: 301-306. . 1991. Heat tolerance and thyroid activity in developing tadpoles and juvenile adults of Xevwpus hmis (Daudin).J Therm.Bzol. 16:273-276. Daniel, J. E, and A. B. Burch. 1933. A rotary disc for the observation of objects in proiie. Unia. Calqomia Publ. 2001. 39:201-204. Danos, S. O., J. S. Maki, and C. C. Rernsen. 1983. Stratiication of microorganisrns and nutrients in the surface microlayer of srnali freshwater ponds. Hydrobwloga 98: 193-202. Dameli, R. M., Jr. 1949. The aortic arches and associated arteries of caudate Amphibia. Copeia 1949:18-3 1. Das, I. 1994. The internai oral rnorphology of some anuran larvae fiorn south India: A scanning electron rnicroscopic study. Amphibia-Reptilia 15:249-256. Dash, M. C., and A. K. Hota. 1980. Density effects on the survivai, growth rate, and metamorphosis of Rana tgrina tadpoles. Ewlogy 61:1025-1028. Dash, M. C., and B. K. Mahapatro. 1990. Effect of food quality and quantity on growth and rnetamorphosis of BuJOstomatic 1arvae.J Ecobwl. 2:67-76. w Dash, M. C., and P. K. Mishra. 1989. Energy cost of rnetamorphosis in a tropical tree frog Polypdutes muiutus (Gray).J Ecobwl. 1:131-136. Daua, M., R. Calvert, D. Mnard, J. S. Hugon, and J.

LITERATURE CITED

363

Degani, G., and E. Nevo. 1986. Osmotic stress and osmoregulation of tadpoles and juveniles of Pelobates yriacus. Comp. Bwchem. Physwl. 83A:365-370. DeHaan, R. L. 1965. Morphogenesis of the vertebrate heart. pp. 377400. In Orgamgenesis, edited by R. L. DeHaan and H . Urspning. New York: Holt, Rinehart, and Winston. De Jongh, H. J. 1967. Relative growth of the eye in larval and metamorphosingRana tempuraria. Growth 3 1:93-103. . 1968. Functional morphology of the jaw apparatus of larval and metamorphosing Rana temporaria L. Netherlands J. 2001. 18:l-103. De Ia Riva, I. 1995. A new reproductive mode for the genus A d e m m a (Amphibia: Anura: Leptodactylidae): Taxonomic implications for certain Boiivian and Paraguayan popuiations. Stud. Neotrop. Fauna Environ. 30: 15-29. Delbos, M. 1975. Observations sur I'ultrastructure de I'uretre primaire et du blastme msonphrtique avant sa differenciation chez Rana hlmatina (Amphibien Anoure). C. R. Hebd. SeancesAcad.Sci. Paris 280D:893-895. Delbos, M., J.-D. Gipouloux, and R. Cambar. 1971. Observations sur I'infrastructure des diffrents types cellulaires constituant la gonade larvaire chez Ia grenouiile agile Rana dalmatina Bon. (Amphibien Anoure). C. R. Hebd. SeancesAcad. Sci. Paris 272D:2372-2374. Delbos, M., J.-D. Gipouioux, and S. Guennoun. 1981. Caractres uitrastructuraux et cytochimiques des cellules germinales primordiales des amphibiens anoures. Arch. Anat. Mimos. Morphol. Exp. 70: 117-128. . 1982. Intraendodermal and intramesenteric migration of anuran amphibian germ cells: Transmission and scanning electron microscopy.J. Morphol. 171:355-360. Delbos, M., J. Lestage, and J.-D. Gipouloux. 1980. Localisation cytochimique d'une importante activit adnylate cyclasique dans le cordo-msodermeembryonnaire des amphibiens anoures. Arch. Anat. Micms. Morphol. Exp. 69:47-56. Delbos, M., K. R. Miller, and J.-D. Gipouiow. 1984. Freezefracture of Ranapipiens gonad anlage: Study of primordial germ cells and other cellular types. Arch. Anat. Mimos. Morphol. Exp. 73:57-67. D e h o , G. 1977. I differenziamento deiie ghiandole granulose l cutanee in larve di Bombina variegata pachypus (Bonaparte)(Anfibio, Anuro, Discoglosside). Ricerca al microscopio ottico e al microscopio elettrinico.Arch. Italiano Anat. Embriol. 82:337-363. D e h o , G., R. Brizzi, and C. Cailoni. 1982. Development of cutaneous glands in Salamandrina terdigitata (Lacpde, 1788)(Amphibia: Urodela); findings by light and electron microscopy. Zeit. Mikms. Anat. Forseb. 96:948-971. Deiidow, B. C. 1989. Reevaluation of the effects of growth hormone and prolactin on anuran tadpole growth and development. J Exp. Zool. 249:279-283. De1 Pino, E. M. 1975. Adaptaciones reproductivas para Ia vida terestre de1 sapo marsupial Gmotheca riobambae (Fowler), Anura, Hylidae. Reu. Uniu. CatolicaEcuadm 3: 119-140. . 1980a. Morphology of the pouch and incubatory integurnent in marsupial fkogs (Hyiidae). Copea 1980:10-17. . 1980b. E1 mantenimiento y aspectos de1 comportaniento encautiverio de1 sapo marsupial Gmotheca riobambae (Hyiidae).Rev. Univ. Catolica Ecuadm 8 : 4 1 4 9 . . 1989a. Marsupial frogs. Sci.American May:llO-118. . 1989b. Modifications of oogenesis and development in marsupial frogs. Devebment 107:169-187. . 1996. The expression of Brachyury (T) during gastrula-

tion in the marsupial frog Gmotheca ribobambae. Deveiup. BwL. 177:64-68. De1 Pino, E. M., I. Aicocer, and H . Grunz. 1994. Urea is necessary for the cuiture of embryos of the marsupial frog G m o theca riobambae, and is tolerated by embryos of the aquatic frog Xenopus k s . Deveiup. Growth Dzffe1: 36: 73-80. De1 Pino, E. M., and R. P. Elinson. 1983. A nove1 development pattern for frogs: Gastruiation produces an embryonic disk. Nature 306:589-591. De1 Pino, E. M., and B. Escobar. 1981. Embryonic stages of Gastrotheca riobambae (Fowler) during maternal incubation and comparison of development with that of other eggbrooding hyiid fr0gs.J. Morphol. 167:277-295. De1 Pino, E. M., M. L. Galarza, C. M. de Albuja, and A. A. Humphries, Jr. 1975. The maternal pouch and development in the marsupial frog Gmotheca riobambae (Fowler). Bwl. Bull. 149:480491. De1 Pino, E. M., and A. A. Humphries, Jr. 1978. Multiple nuclei during early oogenesis in Fiectonotuspygmaeus and other marsupial frogs. Bwl. Bull. 154:198-212. De1 Pino, E. M., and S. Loor-Vela. 1990. The pattern of early cleavage of the marsupial frog Gastrothecariobambae. Deveiupment 110:781-787. De1 Pino, E. M., C. Murphy, and P. H. Masson. 1992. 5 s rRNA-encoding genes of the marsupial fkog Gmotheca riobamba. Gene 111:235-242. De1 Pino, E. M., H. Steinbeisser, A. Hofmann, C. Dreyer, M. Campos, and M. F. Trendelenburg. 1986. Oogenesis in the egg-brooding frog Gastrothecariobambae produces large oocytes with fewer nucleoli and low RNA content in comparison to Xenopus l m i . Dzfferentiation 32:24-33. Delsol, M., and J. Flatin. 1972. Anatmie du wme vasculaire d s ttards de batraciens. Paris: Lib. Facult. Sci. e Dempster, W; T. 1930. The morphology of the amphibian endolymphatic organ.J. MovphoL. Physwl. 50:71-126. Denis, H . 1959. Sur I'induction des capsules nasales chez les Amphibiens.Areh. BwL. 70:851-873. Denton, J., and T. J. C. Beebee. 1991. Palatability of anuran eggs and embryos.Amphzbia-Reptilia 12:111-114. Denver, R. J. 1997. Environmental stress as a developmentai cue: Corticotropin-releasinghormone is a proximate mediator of adaptive phenotypic plasticity in amphibian metamorphosis. Homz. Behav. 31:169-179. De Prez, G. R., P. M. Ruiz-Carranza, and M. P. RamirezPinilla. 1992a. Modificaciones tegumentarias de larvas y adultos durante e1 cuidado parend en Minyobates virolinensis (Amphibia: Anura: Dendrobatidae). Calasia 17:75-86. . 1992b. Especializaciones de1 tegurnento de incubacion de Ia hembra de Cryptobatrachw bouiengeri (Amphibia: Anura: Hylidae). C h i a 17:87-94. De Quiroga, B. G., M. Lopez-Torres, and P. Gil. 1989. Hyperoxia decreases lung size of amphibian tadpoles without changing GSH-peroxidase or tissue peroxidation. Comp.Bwchem. Physwl. 92A:581-588. De S, R. 0 . 1983. Descripcion de la larva deAgenteohylaszemersi (Mertens 1937), (Anura: Hylidae). Res. J. Cknc. Nat. Montm'deo 3 :4 0 4 1. . 1988. Chondrocranium and ossification sequence of Hyla lanc$omzis. J Morphol. 195:345-355. De S, R. O., and S. Hill. 1998. Chondrocranial anatomy and skeletogenesis in Dendrobates auratus. J. Herpetol. 32:205-210. De S, R. O., and E. O. Lavilla. 1997. The tadpoles of Pseudis

364

LITERATURE CITED

minuta (Anura: Pseudidae), an apparent case of heterochrony. Amphibia-%tilia 18:229-240. De S, R. O., and L. Trueb. 1991. Osteology, skeletal developrnent, and chondrocranial structure of Hamptophlyne boliviana.J Molphol. 209:311-330. De Saint-Aubain, M. L. 1981. Shunts in the gi iiament in tadpoles of Rana tempovaria and Bufo bufo (Arnphibia, Anura). J Exp. 2001. 143-145. 217: . 1982. The rnorphology of amphibian skin vascularization before and after rnetamorphosis. Zoomol;~hology 100:55-63. . 1985. Blood flow patterns of the respiratory systerns in larval and adult amphibians: Functional rnorphology and phylogenetic significance. Zeit. 2001. Syst. Evol.-forsch. 23:229-240. Desser, S. S., J. Lorn, and I. Dykova. 1986. Developrnental stages of Sphmrospora ohlmacheri (Whinery 1893) n. cornb. (Myxoma: Myxosporea) in the renal tubules of bulifrog tadpoles, Rana catesbeiana, frorn Lake of Two Rivers, Aigonquin Park, Ontario. CanadianJ 2001. 642231-2217. Deuchar, E. M. 1975. Xenupus: The South Afizcan clawed f7.g. New York: Wiley. Deunff, J., and A. Beaurnont. 1959. Histognese des dents et du bec corns chez les larves de Discglossuspictus 0 t h . C.R Soc. Bwl. Paris 1959:1162-1 164. De Villiers, C. G. S. 1929a. Some observations on the breeding habits of the Anura of the Steiienbosch Flats, in particular of Cmostemum capense and Bufo anp-ticeps. Ann. Transvad Mas. 13:123-141. . 1929b. Some features of the early developrnent ofBreviceps. Ann. Transvad Mus. 13:142-1 5 1. . 1929c. The developrnent of a species of Arthroleptelh frorn Jonkershoek, Steiienbosch. South Afizcan J Sci. 26:481-5 10. De Vlaming, V. L., and R. B. Bury. 1970. Therrnal selection in tadpoles of the tailed-frog, Rrcaphus h e i . J Heqetol. 4: 179-189. De Waal, S. W. P. 1973. Die rnesonefros van Xenupw l m i (Daudin) 'n histologiese oorsig. Navm. .Nas. Mus. (Bloernfontein) 2:356-385. De Wltte, G. F. 1948. Faune de Belgique. Arnphibiens et Reptiles. Bruxelles. Deyrup, I. J. 1964. Water balance and kidney. pp. 251-328. In PhysiologyoftheAmphibia.Vol. 1, edited by J. A. Moore. New York: Acadernic Press. D i z , N., M. Saiiaberry, and H. Nufez. 1983. The tadpole of Telmatobufovenustus (Anura: Leptodactylidae) with a consideration of generic relationships. Heqetologica 39: 111-113. D i z , N. F., and J. Valencia. 1985. Larval rnorphology and phenetic relationships of the Chilean Alsodes, Telmatobius, Caudiverbera and Insuetophlynus (Anura: Leptodactylidae). Copeia 1985:175-181. Diz-Paniagua, C. 1985. Larval diets related to rnorphological charaaers of five anuran species in the biological reserve of Donaia (Huelva, Spain).Amphibia-Reptilia 6:307-322. . 1986. Reproductive penod of amphibians in the biological reserve of Donaia (SW Spain). pp. 429-432. In Studies in heqetoio~y,edited by Z . RoTek Prague: Charles University Press. . 1987. Tadpole distribution in relation to vegetal heterogeneity in ternporary ponds. Heqeto1.J 1:167-169. . 1988. Temporal segregation in larval amphibian cornrnunities in ternporary ponds at a locality in SW Spain.Amphibia-Reptilia 9: 15-26.

. 1989. Larval diets of two anuran species, Peiodytespunctatus and Bufo bufo, in SW Spain. Amphibia-Reptilia 10:7175. Di Bernardino, M. A. 1961. Investigations of the gerrn-plasrn in relation to nuclear transplantation. J Embyol. Exp. Movphol. 9:507-513. DiBona, D. R., M. M. Civan, and A. Led. 1969. The anatornic site of the transepithelial perrneability barriers of toad bladder./: CeU Bwl. 40: 1-7. Dickman, M. 1968. The effea of grzing by tadpoles on the structure of a periphyton cornmunity. Ecology 49:11881190. Diener, E., and G. J. V. Nossal. 1966. Phylogenetic studies on the immune response. I. Localization of antigens and irnmune response in the toad, Bufo mavinus. Immumlo~y 10:535-542. Diesel, R., G. Baurle, and E Vogel. 1995. Cave breeding and froglet transport: A nove1 pattern of anuran brood care in the Jamaican frog, Eleutherodac~luscundalli. Copeia 1995:354-360. Dietz, T. H., and R. H. Aivarado. 1974. Na and C1 transport across gi chamber epitheliurn of Rana catesbeiana tadpoles. AmeriuGnJ Physwl. 226: 764-770. Di Grande, F. 1968. Sviluppo postmetamorfico deiie gonadi e degli organi di Bidder in esernplari di Bufo bufo precocernente panirradiati con raggi X. Rend. Accad. Naz. Lincei. Ser. 8,45:437-442. . 1987. Aspetti deiia rnorfogenesi deiie gonadi e degli organi di Bidder in giovani esernplari di Bufo bufo sperimentalrnente privati di gran parte deiie loro cellule germinaii. I. Stadi prernetamorfici. Arch. Italiano Anat. Embriol. 92: 159177. Di Grande, F., and 0 . Marescalchi. 1987. Aspetti deiia modogenesi deiie gonadi e degli organi di Bidder in giovani esernplari di Bufo bufo sperimentalmentepnvati di gran parte deiie loro cellule germinali. 11. Stadi prernetamorfici postmetarnorfici. Arch. Italiano Anat. Embriol. 92: 179-208. Dionne, J.-C. 1969. Tadpole holes: A true biogenic sedirnentary structure.J Sedivnent. Pehol. 39:358-360. DiStefano, R. J., M. J. Roell, B. A. Wagner, and J. J. Descoske. 1994. Relative perforrnances of four preservatives on fish and crayfish. Trans.A v n e k n Fhh. Soc. 1232317-823. Ditrich, H., and A. Lametschwandtner. 1992. Glorneruiar developrnent and growth of the renal blood vacular systern in Xenopus iamis (Arnphibia, Anura, Pipidae) during rnetamorphic c1irnax.J Moqhol. 213:335-340. Dixon, J. R, and W. R Heyer. 1968. Anuran succession in a ternporary pond in Colima, Mexico. BuU. Southem Calqomia Acad. Sci. 67: 129-137. Dixon, J. R, C. Mercolii, and A. A. Yanosky. 1995. Some aspeas of the ecology of Pseudisparaduxa frorn northeastern Argentina. Heqetol. Rev. 26: 183-184. Dixon, J. R., and C. Rivero-Blanco. 1985. A new dendrobatid frog (Colosrethus)frorn Venezuela with notes on its natural history and that of related species.J Hqetol. 19:177-184. Dixon, K. E. 1981. The origin of the primordial gerrn ceils in the Arnphibia. NetherhndcJ Zool. 315 3 7 . Dodd, J. M. 1950. Ciliary feeding rnechanisrns in anuran larvae. Nature 165:283. Dodd, M. H. I., and J. M. Dodd. 1976. The biology of rnetarnorphosis. pp. 467-599. In Physwlogy of the Amphibia. Vol. 3, edited by B. Lofts. New York: Academic Press. Donnelly, M. A., R. O. De S, and C. Guyer. 1990a. Descnption of the tadpoles of Gastwphrynepictiventris and Nelson-

LITERATURE CITED ophlyne aterrima (Anura, Microhylidae), with a review of morphologicai variation in free-swimming microhylid larvae. American Mus. Navitates (2976):1-19. Donnelly, M. A., and C. Guyer. 1994. Patterns of reproduction and habitat use in an assemblage o neotropical hyiid frogs. Oecologa 98:291-302. Donnelly, M. A., C. Guyer, and R O. De S. 1990b. The tadpole of a dm-poison frog Phyllobates lugubris (Anura: Dendrobatidae). Proc. Bwl. Soc. Washington 103:427431. Donnelly, M. A., C. Guyer, D. M. Krempels, and H. E. Braker. 1987. The tadpole of Agdychnis cakarz.r (Anura: Hyiidae). Copeia 1987:247-250. Donner, K. O., and T. Reuter. 1976. Visual pigments and photoreceptor function. pp. 251-277. In Frog neurobwlogy, a handbook, edited by R. Llins and W. Precht Berlin: Springer-Verlag. Doughty, P., and R. Shine. 1998. Reproductive energy docation and long term energy stores in a viviparous lizard (Euhmprus tympanum).Ecology 79:1073-1083. Dowling, J. E. 1976. Physiology and morphology of the retina. pp. 278-296. In Frog neurobwiugy, a handbook, edited by R. Liins and W. Precht Beriin: Springer-Verlag. Downes, H., and P. M. Courogen. 1996. Contrasting effects of anesthetics in tadpole bioassays. J. P h a m o l . Exp. 278:284-278. Downie, J. R. 1990a. Temporal changes in the behavior o foam-making Leptodagylus jscus tadpoles. Herpetol. J. 1:498-500. . 1990b. Functions o the foam in foam-nesting leptodactyiids: Antipredator effects of Physalmmus pustulosus foam. Hqet01.J. 1:501-503. . 1994a. Developmental arrest in Leptodagylusjscus tadpoles (Anura: Leptodactylidae). 1: Descriptive analysis. Herpet01.J. 4:29-38. . 1994b. Developmentai arrest in Leptodagylusfiscus tadpoles (Anura: Leptodactyiidae). 2: Does a foam-borne factor block development?Hqeto1.J 4: 3 9 4 5 . . 1996. A new example ofemale parentai behaviour in Leptodactylus validus, a frog of the leptodactylid "melanonotu? species group. Herpeto1.J. 6:32-34. Downie, J. R., and A. Weir. 1997. Developmental arrest in Leptodagy1usjscus tadpoles (Anura: Leptodactylidae) 111: Effect o length of arrest period on growth potentiai. Herpeto1.J. 7:85-92.17. Dressler, G. R., and P. Grusse. 1988. Do multigene families regdate vertebrate development?Trends Genet. 4:214-219. Drewes, R. C. 1984. A phylogenetic analysis of the Hyperoiiidae (Anura): Treefrogs of Mrica, Madagascar, and the Seychelle Islands. Occm. Pap. CdifoorniaAcad. Sci. (139):170. Drewes, R. C., R. Altig, and K. M. Howell. 1989. Tadpoles of three frog species endemic to the forest o the Eastern Arc Mountains, Tanzania. Amphibia-Reptilia 10:435-443. Dreyer, T. F. 1915. The morphology othe tadpole of Xenopzs laevis. Trans.Roy. Soc. South A j k a 4:241-258. Drysdale, T. A., and R. P. Eiinson. 1993. Head ectodermal patterning and axial development in frogs. American Zool. 33:417423. Dubois, A. 1985. Diagnose prliminaire d'un nouveau genre de Ranoidea (Amphibien, Anoures) du sud de I'Inde. Alytes 4:113-118. . 1987. Miscellanea taxinomica batrachologia (I).Alytes 5:7-95. . 1992. Notes sur Ia clasification des Ranidae (Amphib-

365

iens Anoures). Bull. Mensuel. Soc. Linneenne h Lyon 61:305-352. . 1995. Keratodont formulae in anuran tadpoles: Proposals for a standardizati0n.J. Zool. Syst. Evol. Res. 33: 1-15. Dubois, A., and A. Ohler. 1994. Formules dentaires des ttard d'hphibiens Anoures: Une nouvelle mthode d'expression et de comparaison.Ann. Sci. Nat. Zool., Paris 15:163-166. Dudley, R., V. A. King, and R. J. Wassersug. 1991. The implications of shape and metamorphosis for drag forces on a generalized pnd tadpole ( ~ a n catesbeiana).-@eis 1991: a 252-257. Dueiiman, W. E. 1970. The hyiid frogs of Middle America. Mus. Nat. Hist., Univ.Kansm Monogr (1): 1-75 3, 72 plates. . 1972. A review o the neotropical frogs of the Hyh bogotenszs group. Occas. Pap. Mus. gat. H& Univ. Kahas (l1):l-31. . 1978. The biology o an equatorial herpetofauna in Amazonian Ecuador. Misc. Publ. Mus. Nat. H&., Univ.Kansas (65):1-352,4 plates. 1 9 8 5 . Reproductive modes in anuran amphibians: Phylogenetic sig&cance of adaptive strategies. South ~ f i c a n ] Sci. 81:174-178. . 1989. Aiternative life-history styles in anuran amphibians: Evolutionary and ecological implications. pp. 101-126. InAltemative lzf-histolys~ylesofanimals, edited by M. N. Bruton. Dordrecht: Kluwer Acad. h b l . Dueiimann, W. E., and R. Altig. 1978. New species of tree frogs (Hyiidae) from the Andes of Colombia and Ecuador. Hqetologua 34: 177-185. Duellrnan, W. E., and 1. A. Campbell. 1992. Hyiid frogs othe genus Plectrohyla: Systematics and phylogenetic relationships. Mkc. Publ. Mus. Zool., Univ. Kansm (1 :1-32. 81) Dueiiman, W E., and R. O. De S. 1988. A new genus and species of South American hyiid frog with a highly modked tadpole. Trop.Zool. 1:117-1 36. Dueiiman, W E., and M. J. Fouquette, Jr. 1968. Middle American frogs of the Hyh microcephah group. Univ. Kansas Publ. Mus. Nat. H&. 17:5 17-5 57. Duciiman, W E., and P. Gray. 1983. Developmental biology and systematics of the egg-brooding hyiid frogs, genera Flectonotus and Fritzzana. Herpetoiugica 39:333-3 59. Dueiiman, W E., and D. M. Hillis. 1987. Marsupial frogs (Anura: Hylidae: Gastrotheca) of the Ecuadorian Andes: Resolution otaxonomic problems and phylogenetic relationships. Herpetologica 43: 141-173. Dueiiman, W. E., and M. S. Hoogrnoed. 1984. The taxonomy and phylogenetic relationships of the hyiid frog genus St$ania. Misc. Publ. Mus. Nat. Hist., Univ. Kansas (75): 1-39. Dueiiman, W. E., and J. Lescure. 1973. Life history and ecology of the hyiid frog Osteocephalustaurinus, with observations on larval behavior. Occm. Pap. Mus. Nat. Hzst., Univ. Kansas (l3):l-12. Dueiiman, W. E., and J. D. Lynch. 1969. Descriptions ofAtelopus tadpoles and their relevante to atelopodid dassification. Hevpetohgua 25 :23 1-240. Dueiiman, W E., and S. J. Maness. 1980. The reproductive behavior of some hylid marsupial frogs. J Herpetol. 14:213-222. D u e h a n , W. E., and A. Schwartz. 1958. Amphibians and reptiles osouthern Florida. Bull. Florida St. Mus. 3:181-324. Dueiiman, W. E., and L. Trueb. 1983. Frogs othe Hyh columbiana group: Taxonomy and phylogenetic relationships. pp. 33-5 1. Ln Advances in herpetology and molutiona? bwiugy, ed\ ,

366

LITERATURE CITED

ited by A. G. J. Rhodin and K. Miyata. Carnbridge, MA: Mus. Comp. Zool. . 1986. Bwlogy of amphibians. New York: McGraw-Hili. . 1989. Two new treefrogs of the Hylapamaps group from the A m w n Basin in southern Peru. Hepologca 45:l-10. Dugs, A. 1834. Recherches sur Postologie et la myologie des batraciens a leurs difFrens ages. Mm. 154ca. Roy. I'Inst. Eiance, Sci. Math. Phys. 6: 1-216. Duke, J. T., and G. R. Ultsch. 1990. Metabolic oxygen regulation and conformity during submergence in the salamanders Siren iaartina,Amphiuma means, andAmphiuma tntnd&ylum, and a comparison with other giant salamanders. Oecologa 84:16-23. Duke, K. L. 1978. Nonfollicular ovarian components. pp. 563582. In The vertebrate mary comparative biology and evolution, edited by R. E. Jones. New York: Plenum. Dumas, P. C. 1964. Species-pair aiiopatry in the genera Rana and Phrynosoma. Ecology 45 :178-18 1. D d a ~ D. G. 1967. The develo~ment the muscuiature of . of th:hindlimb of the frog, ~ana'~ipiens.J. p h o l119:241~m . 258. Dunlap, D. G., and C. K. Satterfield. 1985. Habitat selection in larvai anurans: Early experiente and substrate pattern selection in Rana pipienr. Develop. Psychobwl. 18:37-58. Dunlop, S. A., and L. D. Beazley. 1981. Changing retinal ganglion cell distribution in the frog Helewpurus eyrei. J. Comp. Neurol. 202:221-236. Dunn, E. R. 1924. Some Panamanian frogs. O u . Pap. Mus. Zool., Univ.Michigan (15 1):1-16. . 1926. The frogs of Jamaica. Proc. Boston Soc. Nat. H h . 38:lll-130. Dunn, R. F. 1978. Nerve fibers of the eighth nerve and their distribution to the sensory nerves of the inner ear in the bulfrog. J. Comp.Neurol. 182:62 1-636. Dunson, W. A. 1977. Tolerante to high ternperature and salinity by tadpoles of the Philippine frog, Rana cancrivora. Copeia 1977:375-378. Dunson. W. A.. and 1. Connell. 1982. S~ecific inhibition of hatc&ng inZamphibian embryos by l;>w pH. J. Herpetol. 16:314-316. Dunson, W. A., and J. Travis. 1991. The role of abiotic factors in comrnunity organization.A k a n N a t . 138:1067-1091. Du Pasquier, L. 1968. Les protins sriques et le complexe lymphomyloide chez le ttard d'Alytes obsteti-icans normal and thymectomis.Ann. I&. Pasteur 114:490-502. . 1970. Ontogeny of the immune response in animais having less than one miion lymphocytes: The larvae of the toad Alpes obstetn'cans. Immuvwlogy 19:353-362. Dupr, R K., J. J. Just, E. C. Crawford, Jr., and T. L. Powell. 1986. Temperature preference and responses of cutaneous temperature-sensitive neurons during buMrog development. Physwl. Zool. 59:254-262. Dupr, R. K., and J. W. Petranka. 1985. Ontogeny of temperature selection in larval amphibians. Copeia 1985:462467. Dupr, R. K., and S. C. Wood. 1988. Behavioral temperature regulation by aquatic ectotherms during hypoxia. Canuian J. Zool. 66:2649-2652. Dutta, S. K., and P. Mohanty-Hejmadi. 1984. Ontogeny of teeth row stnicture in Rana tigerina tadpoles.J. Bombay Nat. H&. SOL. 80:517-528. Dziadek, M., and K. E. Dixon. 1975. Mitosis in presumptive primordial germ ceils in post-blastula embryos of Xenopzs laevis.J. J Zool. 192:285-291. .

. 1977. An autoradiographicanalysis of nucleic acid synthesis in the presumptive primordial germ cells of Xenopus M s . J . Embryol. E&p. Movphol. 37:13-31. Eakin, R. M. 1963. Ultrastructural differentiation of the oral sucker in the treefrog, Hyh regilu. Develop. Bwl. 7: 169-179. Eakin, R. M., and F. E. Bush. 1957. Development of the amphibian pituitary with speciai referente to the neural lobe. Anat. Rec. 129:279-296. Eaton, T. H. 1959. The ancestry of modern Amphibia: A review of the evidente. Univ. finsas Publ. Mus. Nat. H&. 12:157-180. Ebbesson, S. O. E. 1970. On the organization of central visual pathways in vertebrates. Brain Behav. Evol. 3: 178-194. . 1976. Morphology of the spinal cord. pp. 679-706. In Frog neurobwlogy, a handbook, edited by R. Llins and W. Precht Berlin: Springer-Verlag. . 1980. Comparative neurology of the telencepham. New York: Plenum. Ebenman, B. 1988. Competition benveen age classes andpopulation dynamics.J. Theol:Bwl. 131:389400. Eberhard, M. L., and F. H. Brandt. 1995. The role of tadpoles and frogs as paratenic hosts in the life cycle of Dracunculus insigntc (Nematoda: Dracunculoidea).J. Parasitol. 81:792793. Echeverra, D. D. 1988. Oogenesis en las hembras juveniles de Bufo arenarum (Anura, Bufonidae). Rev. Mus. Argentina Cienc. Nat. 'Bemardino Rivaduvia' Inst. Nac. Invest. Cienc. Nat., Zool. 15:57-75. . 1994 (1991). Metamorphosis en Bufo arenamm (Anura, (Bufonidae): Consideraciones acercade Ia salida de 10s miembros anteriores de las larvas. Physis 49B: 1 4 . . 1995. Microscopia electrnica de barrido de1 aparato bucal y de la cavidad oral de Ia larva de Leptod&yLus ocellatus ( L i a e u s 1758)(Anura, Leptodactylidae). Alytes 12:159168. . 1996. Microscopa electrnica de barrido de1 aparato bucal y de Ia cavidad oral de la larva de Hyhstmigilata&ngwphila Gaiiardo, 1961 (Anura, Hylidae) con comentarios que facilitan su identificin durante Ia metamorphosis y postmetamorphosis. Phystc 54B:1-6. . 1997. Microanatomy of the buccal apparatus and oral cavity of Hyh minuta Peters, 1872 larvae (Anura, Hylidae), with data on feeding habits. Alpes 15:26-36. Echeverra, D. D., and A. M. Filipello. 1994 (1991). Consideraciones acerca de Ia formula dentaria y de1 aparato bucal de Ias larvas de Oduntilphrynus occihntalis (Berg, 1896)(Anura, Leptodactylidae). Physis 49B:59-63. Echeverra, D. D., and L. E. Fiorito de Lopez. 1981. Estadios de lametamorfosisen Bufo arenarum (Anura).P h y 40B: 15-23. ~ Echeverra, D. D., L. E. Fiorito de Lopez, and S. B. Montanelii. 1992. Esteromorfologia de1 aparato bucal y de Ia cavidad oral de las larvas de Ololygonfuscmaria (Lutz 1925)(Anura, Hylidae). &. Mus. Argentina Cienc. Nat. '%emardinoRivadavia" I&. Nac. Invest. Cienc.Nat., Zool. 16:2-13. Echeverra, D. D., L. E. Fiorito de I ~ p e z S. B. Montaneili, , O. B. Vaccaro, and A. M. Filipello. 1987. Consideraciones acerca de la formulas dentarias de Ias larvas de Bufo arenamm Hensel (Anura: Bufonidae). C u d . Herpetol. 3:33-39. Eddy, E. M. 1975. Germ plasm and the difFerentiation of the germ ceil line. Intl. Rev.Cytol. 43:229-280. Eddy, E. M., and S. Ito. 1971. Fine structural and radioautographic observations on dense perinuclear cytoplasmic material in tadpole oocytes.J. Cell Bwl. 49:90-108. Edgeworth, F. H. 1920. On the development of the hypo-

LITERATURE CITED branchial and laryngeal muscles in Amphibia. J. Anat. 44: 125-162. . 1930. On the masticatory and hyoid muscles of larvae of Xenopus l&. J.Anat. 64: 184-188. . 1935. The cranial muscles of vefiebrates. Cambridge: Cambridge University Press. Edwards, M. L. O., and E. B. Pivorun. 1991. The effects of photoperiod and different dosages of melatonin on metamorphic rate and weight gain in Xenopus W tadpoles. Gen. Comp.Endocrinul. 8 1:28-38. Ehmann, H., and G. Swan. 1985. Reproduction and development in the marsupial frog, Assa darlingtoni (Leptodaaylidae, Anura). pp. 279-285. In The bw&y ojAustrahian~ogs and reptiles, edited by G. Grigg, R. Shine, and H. Ehmann. Chipping Norton, New South Wales: Surrey Beatty. Eichler, V. B., and R. A. Porter. 1981. Rohon-Beardcells in frog development: A study of temporal and spatial changes in a transient ceii population.J. Comp.Neurol. 203: 121-130. Eiselt, J. 1942. Der Musculus opercularis und die mittlere Ohrsph'ie der anuren Amphibien. Arch. Natuvgesch. (n. sel:) 10:179-219. Ekblom, P. 1989. Developmentally regulated conversion of mesenchyme to epithelium. FASEBJ. 3:2141-2150. Ekman, G. 1913. Experimentelle Untersuchungen ber die Entwicklung der Kiemenregion (Kiemenfaden und Kiemenspatten) einiger anuren Amphibien. Morphol. Jahrb. 47:419-592. . 1924. Experimentelle Beitrage zur Herzentwicklung der Amphibien. Roux'Arch. 106:320-352. . 1925. Neue experimentelle Beitrage zur frhesten Entwicklung des Amphibienherzens. Acta Soc. Sci. Fennicae, C m m .Bwl. 1:1-36. Eldred, W. D., T. E. Finger, and J. Nolte. 1980. Central projections of the frontal organ of Ranapipiens, as demonstrated by the anterograde transport of horseradish peroxidase. Cell Tiss.Res. 211:215-222. Elepfandt, A. 1988a. Central organization of wave localization in the clawed frog, Xenupus W .I. Involvement and bilateral organization of the rnidbrain. Brazn Behav. Evol. 31:349-357. . 1988b. Central organization of wave localization in the clawed frog, Xenops imis. 1 . Midbrain topology for wave 1 directions. Brain Behav. Evol. 3 1:358-368. Elias, H. 1937. Zur vergleichenden Histologie und Entwickiungsgeschichte der Haut der Anuren. Zeit. Mikros. Anat. Fonch. 41:359416. Elinson, R. E 1987a. Change in development patterns: Embryos of amphibians with large eggs. pp. 1-21. In Deveiupment as an evolutwnarypcess. MBL Lect. Bwl. Vol. 8, edited by R. A. RaEand E. C. RafE New York: Alan R. Liss. . 1987b. Fertilization and aqueous development of the Puerto Rican terrestrial-breeding frog, Eletttherodugylus w qui.J. Morphol. 193:217-224. . 1990. Direct development in frogs: Wiping the recapitulationist slate clean. Seminan Deveiup. Bwl. 1:263-270. . 1994. Leg development in a frog without a tadpole (Eleutherodugyluscoqui).J. Exp. Zool. 270:202-21 O. Elinson, R. P., and E. M. De1 Pino. 1985. Cleavage and gastrulation in the egg-brooding, marsupial frog, Gastrotheca nobambae.J.Embryol. Exp. Morphol. 90:223-232. Elinson, R. P., E. M. De1 Pino, D. S. Townsend, F. C. Cuesta, and P. Eichhorn. 1990. A practical guide to the developmental biology of terrestrial-breeding frogs. Bwl. Bull. 179:163-177.

367

Emerson, S. B. 1976. A prelirninq report on the superficial throat musculature of the Microhylidae and its possible role in tongue action. Copeia 1976:546-55 1. . 1986. Heterochrony and frogs: The relationship of a life history trait to morphological form. American Nat. 127:167-183. Emerson, S. B., J. Travis, and M. Blouin. 1988. Evaluating a hypothesis about heterochrony: Larval life-history traits and juvenile hind-limb morphology in Hyla cyucijil: Evolution 42: 68-78. Engel, W., and M. Schmid. 1981. H-Y antigen as a too1 for the determination of the heterogametic sex in Amphibia. Cytogenet. Cell Genet. 30: 130-136. Engels, H. G. 1935. ber Umbildungsvorgiinge im Kardinalvenensystem bei Bildung der Urniere. GegenbaunMmphol. Jahrb. 76: 345-374. Engermann, W.-E., J. Fritzsche, R. Gnther, and F. J. Obst. 1993. Lurche und Kriechtiere Europas. Radebeul: Neuman Verlag. Epple A., and T. L. Lewis. 1973. Comparative histophysiology of the pancreatic islets. American Zool. 13:567-590. Erasmus, B. de W, B. J. Howeii, and H. Rahn. 1970-1971. Ontogeny of acid-base balance in the buiifrog and chicken. Respil: Physwl. 11:46-53. Erdmann, B. 1921. ber die Entwicklung der Atrioventrikularklappen bei den Anuren. Morphol.Jahrb. 5 1:339-354. Escher, K. 1925. Das verhalten. der Seitenorgane der Wirbeltiere und ihrer nerven beim Ubergang zum Landleben. Acta Zool. 6:3074$4. Estes, R., Z. V. Spinar, and E. Nevo. 1978. Early Cretaceous pipid tadpoles Trom Israel (Amphibia: Anura). ~erpetohgua 34:374-393. Estrada, A. R. 1987. Los nidos terrestres de dos especies de anfibios cubanos de1 gnero Eleutherodugylus (Anura: Leptodactylidae).Poeyana (352):1-9. 1 Etkin, W. 1934. The phenomena of anuran metamorphosis. 1 Oxygen consumption during normal metamorphosis. Physwl. Zool. 7: 129148. . 1936. The phenomena of anuran metamorphosis. 111. The development of the thyroid g1and.J. Mmphol. 59:69-89. . 1964. Metamorphosis. pp. 427468. In Physwlogy of the Amphibia. Vol. 1, edited by J. A. Moore. London: Academic Press. . 1968. Hormonal control of amphibian metamorphosis. pp. 313-348. In Metammphosis, edited by W. Etkin and L. I. Giibert. New York: Meredith Corp. Exbrayat, J. M., and S. Hraouibloquet. 1994. An example of heterochrony - the metamorphosis in Gymnophiona. Bull. SOC.Zool. France 199:117-126. Fabrezi, M. 1985. Anatomic cranial de larvas de Gastrotheca grbcilis (Anura: Hylidae). Bol.Asoc. Herpetol.Avgentina 2:2. Fabrezi, M., and G. Garca. 1994. Metamorfosis de1 aparato hiobranquial en Pleurodema borellii y Ceratophrys cvanwelli (Anu a : Leptodaaylidae).Acta Zool. Libana 42: 189-196. Fabrezi, M., and E. O. Lavilla. 1992. Estructura de1 condrocrneo y esqueleto hiobranquial en larvas de algunos hlidos neotropicales (Anura: Hylidae). Acta Zool. Lilloana 41:155-164. . 1993. Anatomia de1 condrocraneo en larvas de tres especies de Telmatobius de1 grupo meridional (Anura: Leptodactylidae).Physis 48B: 3 9 4 6 . Faier, J. M. 1972. Larval anuran teeth and some ecologic implications. Thesis, University of Nebraska. Faivovich, J., and G. R. Carrizo. 1992. Descripcin de Ia larva

368

LITERATURE CITED Feder, M. E., and W. W. Burggren. 1985. Cutaneous gas exchange in vertebrates: Design, patterns, control and implications. Bwl. Kev. 60:1-45. Feder, M. E., R. J. Fuli, and J. Piiper. 1988. Eiimination kinetics of acetylene and Preon 22 in resting and active lungless salarnanders. Respil: Physwl. 72:229-240. Feder, M. E., A. G. Gibbs, G. A. Griflith, and J. Tsuji. 1984a. Therrnal acclimation of rnetabolism in salamanders: Fact or artefact?J.Therm. Bwl. 9:255-260. Feder, M. E., and A. W. Pinder. 1988. Ventilation and its effect on "infinite pooP' exchangers.American Zool. 28:973-983. Feder, M. E., D. B. Seale, M. E. Boraas, R. J. Wassersug, and A. G. Gibbs. 1984b. Funaional confias benveen feeding and gas exchange i suspension-feeding tadpoles, Xenupus n 1aevii.J. Exp. Biol. 110:91-98. Feder, M. E., and R. J. Wassersug. 1984. Aerial versus aquatic oxygen consumption in the larvae of the clawed frog, Xenupus l m i .J Exp. Bwl. 108:231-245. Feinsinger, P., E. E. Spears, and R. W. Poole. 1981. A simple measure of niche breadth. Ecologv 62:27-32. Felsenstein, J. 1985. Phylogenies and the comparative method. AmeriLan Nat. 125: 1-15. Feminella, J. W., and C. P. Hawkins. 1994. Tailed frog tadpoles differentially alter their feeding behavior in response to nonvisual cues from four predators. J. North Amerian Benthol. Soc. 13:310-320. Feng, A. S., and W. Lin. 1991. DifFerential innervation patterns of three divisions of frog auditory midbrain ( t o m semicircdaris). J. Comp. Neurol. 306:613-630. Fermin, P. 1765. Dveloppemntpa$it du mYstrede lagr&at~n dufameux mapaud de Surinam. Maestricht. Fernndez, K. 1926. Sober la biologa y reproduccin de batracios argentinos (segunda parte). Bol.Acad. A'ac. Cienc. Crdoba 29:271-320, 1plate. Fernindez, K., and M. Fernndez. 1921. Sobre la biologa y reproduccin de algunos batracios argentinos. I. Cystignathidae. Ann. Soc. Cient.Argentina 91:97-140, 3 plates. Ferns, M. J., and A. H. Lamb. 1987. Regulation of cell numbers in the developing neurornuscular system in Xenopus laevii. Neurosci. Lett. Su.1. 27S:72. Ferrari, L. 1998. Tolerance of high electrolytic and nonelectrolytic osrnolarities in Bufi arenarum prernetamorphic tadpoles under organism density stress. Alpes 15:171175. Figge, F. H. J. 1936. The differential reaction of the blood vessels of a branchiai arch of Amblystoma tgrinum (Colorado axolotl) I. The reaction to adrenalin, oxygen, and carbon dioxide. Physhl. Zool. 9: 79-1 01. Figiel, C. R., Jt, and R. D. Semlitsch. 1991. Effects of nonlethal injury and habitat cornplexity on predation in tadpole populations. CanadianJ. Zool. 695330-834. Filipski, G. T., and M. V; H. Wilson. 1985. Staining nerves in whole cleared amphibians and reptiles using Sudan black B. Cupeia 1985:500-502. Pink, W. L. 1982. The conceptual relationship between ontogeny and phylogeny. Pabobwhgy 8:254-264. Fiorito de Lopez, L. E., and D. D. Echeverra. 1984. Morfogenesis de 10s dientes larvales y pico corneo de Bufi arenarum (Anura, Bufonidae). ZooIodia 13:5 73-578. . 1989. Microanatomia e histogenesis de1 aparato bucal en las larvas de Rufi arenarum (Anura, Bufonidae). Cuad. He~petol. 4:4-10. Fisher, R. A. 1930. The~enetical theory of natural selection. Oxford: Clarendon Press.

de Chmophrys pierotti (Vellard, 1948) (Leptodactylidae, Ceratophryinae).Alpes 10:81-89. Falconer, D. S. 1989. Introductwn to qzlantitative~enetics. ed. 3d London: Longwood. Fales, D. E. 1935. Experiments on the developrnent of the pronephros of Amblystomapunctatum. J. Exp. Zool. 72: 147-1 73. Fang, H., and E. l? Elinson. 1996. Patterns of distalless gene exoression and inductive interaction in the head of the direit developing fkog E b u t h e r o ~ l u s coqui. Develop. Biol. 179:160-172. Fanning, S. A. 1966. A synopsis and key to the tadpoles of Florida. Thesis, Florida State University. Faragher, S. G., and R. G. Jaeger. 1998. Tadpole buiiies: Exarnining rnechanisms of cornpetition in a cornmunity of lawal anurans. CanadianJ. Zool. 76: 144-153. Farel, P. B., and S. E. Bernelmans. 1985. Specificity of rnotoneuron projection patterns during development of the bullfrog tadpole (Rana catesbeinna). J. Comp. Neurol. 238:128-134. . 1986. Restoration of neuromuscular specificity following ventral rhizotorny in the builfrog tadpole, Rana catesbeiana. J. Comp Neurol. 254: 125-132. Farlowe, V. 1928. Algae of ponds as deterrnined by an examination of intesthal contcnts of tadpoles. Bwl. Bull. 55:443448. Farrell, A. P. 1997. Development of cardiovasctiiar systerns. pp. 101-113. In Devebpnent ofcardMvmcularsystems.Mobcdes to o r g a n h , edited by W. W. Burggren and B. B. Keller. Cambridge: Cambridge University Press. Fauth, J. E. 1990. Interactive effeas of predators and early larvai dynamics of the treefrog Hyla chysoscelis. Ecology 71: 1609-1616. Fauth, J. E., and i%! Resetarits, Jr. 1991. Interactions benveen J. the salamander Siren intermedia and the keystone predator Notophthalmus viriascens. Ecology 72 :827-838. Fawcett, D. W., S. Ito, and D. B. Slautterback. 1959. The occurrence of intercelidar bridges in groups of cells exhibiting synchronous Werentiation. J. BU>phys.Bwchem. Qtol. 5:453460. Feder, M . E. 1981. Effect of body size, trophic state, time of day, and experimentai stress on oxygen consumption of anuran larvae: An experimental assessrnent and evaluation of the literature. Comp. Bwchem. Physwl. 7OA:497-508. - . 1982. Effect of developrnental stage and body size on oxygen consumption of anuran larvae: A reappraisal.J Exp. Zool. 220: 3 3 4 2 . . 1983a. Responses to acute aquatic hypoxia in larvae of the frog Rana berlandieri.J. Exp. Bwl. 104:79-95. . 1983b. Effect of hypoxia and body size on the energy nietabolisrn of lungless tadpoles, Bufi woodhousei, and airbreathing anuran 1awae.J. Exp. Zool. 228: 11-19. - -1983c. The relation of air breathing and locornotion to . predation on tadpoles, Rana berlandieri, by turtles. Physiol. Zoo1 56:522-531. . 1984. Consequences of aerial respiration for amphibian larvae. pp. 71-86. In Resp'ratwn and mtabolism ofembryonic vertebrates, edited by R. S. Seymou~: Dordrecht: Dr. W. Junk. . 1985. Thermal acclimation of oxygen consumption and cardiorespiratory frequencies in frog larvae. Physiol. Zool. 58:303-311. . 1992. A perspec~ive environmental ph~siology on of the amphibians. pp. 1-6. In Environmental physwlogy ofthe amphibians, edited by M. E. Feder and W. W. Burggren. Chicago: University of Chicago Press.

LITERATURE CITED Fishwild, T. G., R. A. Schemidt, K. M. Jankens, K. A. Berven, G. J. Gamboa, and C. M. Richards. 1990. Sibling recognition by larval frogs (Ranapipiens,R. sylvatica and Pseudmtis crucqer). Herpetol. 24:4044. J. Fite, K. V, ed. 1976. The amphibian vistal system:A multidisciplinay appvmh. New York: Academic Press. . 1985. Pretectai and accessory-opticvisual nuclei of fish, Amphibia, and reptiles: Theme and variations. Brain Behav. Evol. 26:71-90. Fite, K. V., R. G. Carey, and D. Vicario. 1977. Visual neurons in frog anterior thalamus. BrainRes. 127:283-290. Fletcher, K., and N. B. Myant. 1959. Oxygen consurnption of tadpoles during rnetamorphosis.J. Physiol. 145:353-368. Flock, A. 1965. Electron microscopic and electrophysiologic studies on the lateral line canal organ. Acta Otolayngal. Suppl. 199:1-90. Flock, A., and J. Wersall. 1962. A study of the orientation of the sensory hairs of the receptor celis in the lateral line organ of fish, with special reference to the function of the receptors. J. CeU. Bwl. 15:19-27. Flores-Nava, A., M. A. Olvera-Novoa, and E. Gasca-Leyva. 1994. A cornparison of the effeas of three water-circulation regimes on the aquaculture of buiifrog (Rana catesbeiana Shaw) tadpoles. Agumulture 128:105-1 14. Flower, S. S. 1899. Notes on a second coliection of reptiles and batrachians rnade in the Malay Peninsula and Siam, frorn November 1896 to Septernber 1898, with a list of the species recorded from those countries. Proc. Zool. Soc. London 1899:885-916, plates 59-60. Floyd, R. B. 1983. Ontogenetic change in the ternperature tolerance of larval Bufo marinus (Anura: Bufonidae). Comp.Bwchem. Physwl. 75A:267-271. . 1984. Variation in temperature preference with stage of development of Bufo marinus larvae.J. Herpetol. 18:153158. . 1985. Effeas of photoperiod and starvation on the ternperature tolerance of larvae of the giant toad, Bufo marinus. Cupeia 1985:625-631. Ford, L. S., and D.C. Cannatelia. 1993. The major clades of frogs. Herpetol. Momgr. (7):94-117. Ford, T. D., and W. J. Breed. 1970. Tadpole holes forrned during desciccation of overbank pool. J. Sediment. Petrol. 40: 1044-1045. Forehand, C. J., and P B. Farel. 1982a. Spinai cord develop. rnent in anuran larvae: I. Primary and secondary neurons. J. Comp.Neuol. 209:386-394. . 1982b. Spinai cord developrnent in anuran larvae: 11. Ascendiig and descending pathways. J. Comp. Neurol. 209:395-408. Forge, l?,and R. Barbault. 1977. Ecologie de la reproduction et due dveloppernent larvaire d'un amphibien dserticole, Bufo pentoni Anderson, 1893, au Senegal. l h e Vie 31: 117-125. Forman L. J., and J. J. Just. 1976. The life span of red blood celis in the amphibian larvae, Rana catesbeiana. Deve&. Biol. 50: 537-540. Forrnanowicz, D. R., Jr. 1986. Anuran tadpoleJaquatic insect predator-prey interactions: Tadpole size and predator capture success. Herpetologica 42: 367-373. Forrnanowicz, D. R., Jc, andM. S. Bobka. 1989. Predation risk and microhabitat preference: An experimental study of the behaviorai responses of prey and predator. American Midl. Nat. 121:379-386. Forrnanowicz, D. R., Jr., and E. D. Brodie, Jr. 1982. Relative

369

palatabilities of rnernbers of a larvai amphibian community. Cupeia 1982:91-97. Formas, J. R. 1972. A second species of Chilean frog genus Telmatobufo (Anura: Leptodactylidae).J. Herpetol. 6: 1-3. Formas, J. R., N. F. Diaz, and J. Valencia. 1980. The tadpole of the Chilean frog Insuetuphynus mrp~'cus. Herpetologica 36:316-318. Formas, J. R., and E. Pugn. 1971. Reproduccion y desarrolio de: Batrmhyka antartandica (Barrio)(Anura, Leptodactylidae). Bol. Mus. N u . H&. Nat. Chile 32:201-213. . 1978. Tadpoles of Hylorina sylvatica, Eupsupbw vittatus, and Bufo rubrupu&atus in southern Chile. Herpetologica 34:355-358. Formas, J. R., and M. A. Vera. 1980. Reproductive patterns of Eupsapbus rosew and E. vittatus.J. Herpetol. 14:11-14. Fortrnan, J. R., and R. Altig. 1973. Charaaers of F, hybrid tadpoles benveen six species of Hyh. Cupeia 1973:411416. Foster, M. S., and R. W. McDiarmid. 1982. Study of aggregative behavior of Rcn'nupbynus donalis tadpoles: Design and analysis. Herpetologica 38:395-404. . 1983. Rhinophlynw donali5 (Alma de Vaca, Sapo Borracho, Mexican Burrowing Toad). pp. 419-421. In CostaRica Natural Histoy, edited by D. H. Janzen. Chicago: University of Chicago Press. Foster, R. G., and A. Roberts. 1982. The pineai eye in Xenupus l m 5 embryos and larvae: A photoreceptor with a direa excitatory effect on behavi0ur.J. Comp.Physwl. 145A:413419. Fox, H. 1961. Quantitative anaiysis of normal growth of the pronephric system in larvai Amphibia. PYOG. SOC.LonZool. don 136:301-315. . 1962a. Growth and degeneration of the pronephric systern of Rana temporaria. J. Embyol. Exp. Morphol. 10:103-114. . 1962b. A study of the evolution of the amphibian and dipnoan pronephros by an analysis of its relationship with the anterior spinai nerves. Proc. Zool. SOG. London 138:225256. . 1963. The amphibian pronephros. Quart. h Bwl. . 38:l-25. - . 1965. Early developrnent of the head and pharynx of Neocevatodus with a consideration of its phylogeny. J. Zool. 146:470-554. . 1970. Tissue degeneration: An electron microscopic study of the pronephros of Rana temporaria.J. Embyol. Exp. Morphol. 24: 139-1 57. . 1971. Celi death, thyroxine, and the developrnent of Rana tempuraria larvae with special reference to the pronephros. Exp. Gmntol. 6: 173-1 77. . 1972. Tissue degeneration: An electron microscopic study of the tail skin of Rana temporaria during rnetamorphosis. Arch. Bwl. LGge 83:373-394. . 1977. The anuran tadpole skin: Changes occurring in it during rnetamorphosis and some comparisonswith that of the adult. pp. 269-289. In Comparative anatomy of the skin. PYOG. Symp. Zool. SOL.London (39), edited by R. I. C. Spearrnan. London: Academic Press. . 1981. Cytologicai and rnorphological changes during amphibian rnetamorphosis. pp. 327-362. In Metamorphosis: Aproblem in dme&mental bwlogy. 2d ed., edited by L. I. Gilbert and E. Frieden. New York: Plenum. . 1984. Amphibian morphogenesis. Clifton, NJ: Humana Press. . 1985. Changes in amphibian skin during larval developrnent and rnetamorphosis. pp. 59-87. In Metamorphosis,
- -

370

LITERATURE CITED of interspecific variation in acid tolerance in fish. J. Exp. Bwl. 136:243-258. Freeman, J. A. 1965. The cerebellurn as a timing device: An experimental study in the frog. pp. 397420. In Neurobiology of cerebellar evolutwn and deveiupment, edited by R. Liins. Chicago: American Med. Assoc. Friedman, R. T., N. S. LaPrade, R. M. Aiyawar, and E. G. Huf. 1967. Chemical basis for the [H+]gradient across frog skin. AmericanJ.Physwl. 212:962-972. Friet, S. C., and A. W. Pinder. 1990. Hypoxia during natural aquatic hibernation of the bulifrog. American Zool. 30:69A. Fritzsch, B. 1981a. The pattern of lateral-iine afTerents in urodeles. A horseradish-peroxidase study. Cell Tiss. Res. 218:581-594. . 1981b. Electroreceptors and direction specific arrangement in the lateral line system of salamanders. Zeit. Naturfonch. 36C:493495. . 1988. Phylogenetic and ontogenetic origins of the dorsolateral auditory nucleus of anurans. pp. 561-585. In The evolutwn of the amphibian auitmy ystem, edited by B. Fntzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak. New York: Wiley. Fritzsch, B., A. M. Nikundiwe, and U. Wi. 1984. Projection patterns of lateral-line afTerents in anurans: A comparative HRP study.J. Cmp. Neurol. 229:45 1 4 6 9 . Fritzsch, B., M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak, eds. 1988a. The evolutwn of the amphibian auitmy ystem. New York: Wiley. Fritzsch, B., U. WahnschafTe, and U. C. Bartsch. 1988b. Metamorphic changes in the octavolateraiis system of amphibians. pp: 359-376. In The evolutwnof the amphibianauitmy *em, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak. New York: Wiley. Frohlich, J., H. Aurin, and P. Kemnitz. 1977. Zur epidermalen Kugelzelle von Kraiientroschlarven. V i . Anat. Ges. 71:1171-1175. Frost, D. R. 1985.Amphibianspeciesofthe world.A taxonomu and geographiE r$erence. Lawrence, KS: Assoc. Syst. Coll. Frost, S. K., and S. J. Robinson. 1984. Pigment cell differentiation in the fire-beliied toad, Bombzna orientalis. I. Structural, chemical, and physical aspects of the adult pigment pattern. J.Morphol. 179:229-242. Frye, B. E. 1964. Metamorphic changes in the blood sugar and the pancreatic islets of the frog, Rana clamitans. J. Exp. Zool. 155:215-224. Fujisawa, H. 1987. Mode of growth of retinal axons within the tecturn of Xenopw tadpoles, and inlpiications in the ordered neuronal connection between the retina and the tecturn. J. Comp.Neuml. 260: 127-1 39. F d e r P. M., and S. O. E. Ebbesson. 1973. Central projections of the trigeminal nerve in the bulifrog (Rana catesbeiana). J. Comp.Neurol. 152: 193-200. Funkhouser, A. 1977. Plasma osmolarity of Rana catesbeiana and Scaphi<Ipw hammondi tadpoles. Herpetolo~ua33:272274. Funkhouser, A., and S. A. Foster. 1970. Oxygen uptake and thyroid activity in Hyh regilla tadpoles. Herpetoloflira 26:366-371. Funkhouser, A., and K. S. Mas. 1969. Oxygen consumption during spontaneous amphibian metamorphosis. Physwl. Zool. 42: 15-21. G d , J. G. 1968. DBerential synthesis of the genes for ribosomal RNA during amphibian otigenesis. Proc. Natl. Acad. Sci. United States ofAmerim 60:553-560.

edited by M. Baiis and M. Bownes. Oxford: Clarendon Press. . 1986. Early development of caecilian skin with special reference to the epiderrnis.J. Herpetol. 20: 154167. . 1988. Riesenzehn, goblet cells, Leydig cells, and the large dear ceUs of Xenopus, in the amphibian larval epiderrnis: Fine structure and a consideration of their homology. J. Submimos. Cytol.Pathol. 20:437451. Fox, H., E. Bailey, and R. Mahoney. 1972. Aspects of the ultrastructure of the aiimentary canal and respiratory duas in Xenopus iuevis 1arvae.J. Mmphol. 138:387406. Fox, H., and L. Hamilton. 1964. Origin of the pronephric duct in Xenopw uvis. Arch. Bwl. 75:245-25 1. . 1971. Ultrastructure of diploid and haploid cells of Xenqplls laevis 1arvae.J.Embyol. Exp. Mmphol. 26:8 1-98. Fox, H., E. B. Lane, and M. Whitear. 1978. Sensory nerve endings and receptors in fish and amphibians. pp. 271-281. In The skin ofvertebrates, edited by R. I. C. Spearman and P. A. Riley. I ~ n d o nAcademic Press. : Fox, H., R. Mahoney, and E. Bailey. 1970. Aspects of the ultrastructure of the alimentary canal and associated glands of the Xenopw uvis larva. Arch. Bwl. Li286 81:21-50. Fox, H., and S. C. Turner. 1967. A study of the relationship between the thyroid and larval growth in Rana temporaria andXenopusuvis. A&. Bwl. 78:61-90. Fox, H., and M. Whitear. 1978. Observations on Merkel cells in amphibians. Bwl. Cell. 32:223-232. Foxon, G. E. H. 1964. Blood and respiration. pp. 151-209. In PhyswlogyoftheAmphibia.Vol. 1, edited by J. A. Moore. New York: Academic Press. Franchi, L. L., A. M. Mandi, and S. Zuckermann. 1962. The development of the ovary and the process of oogenesis. pp. 1-88. In The oaay.Vol. 1, edited by S. Zuckermann, A. M. Mandi, and P. Eckstein. New York: Acadernic Press. Frank, B. D., and J. G. Hollyfield. 1987a. Retinal gangiion cell morphology in the frog, Rana ppiens. J. Comp. Neurol. 266:413434. . 1987b. Retina of the tadpole and frog: Delayed dendritic development in a subpopulation of gangiion c e h coincident with metarnorphosis.J. Comp.Neurol. 266:4354k4. Frank, G. 1988. Granulopoiesis in tadpoles of Rana esculenta. Survey of the organs invo1ved.J. Anat. 160:59-66. . 1989a. Granulopoiesis in tadpoles of Rana escubnta. Ultrastructural observations on the developing granulocytes and on the development of eosinophii granules. J. Anat. 163:97-105. . 1989b. Granulopoiesis in tadpoles of Rana esculenta. Ultrastructural observations on the morphology and development of heterophii and basophii granules. J. Anat. 163:107-116. Fraser, E. A. 1950. The development of the vertebrate excretory system. Bwl. Rev. 25: 159-187. Freda, J. 1986. The iduence of acidic pond water on amphibians: A review. Water,Air, Soil Pollut. 30:439450. Freda, J., and W. A. Dunson. 1984. Sodium balance of amphibian larvae exposed to low environmental pH. Physwl. Zool. 57:435-443. . 1985. Field and laboratory studies of ion balance and growth rates of ranid tadpoles chronicaiiy exposed to low pH. Cupeia 1985:415-423. . 1986. The effect of prior exposure on sodium uptake in tadpoles exposed to low pH water. J. Comp. Physwl. 156B:649-654. Freda, J., and D. G. McDonald. 1988. Physiological correlates

LITERATURE CITED Gdardo, J. M. 1961. Observaciones biolgicas sobre Hylaraddiana Fitz., de la Provincia de Buenos Aires. Cienc. Invest. 17:63-69. Gdien, L., and C. Houiiion. 1951. Table cronologique de dveloppement chez Discoglossus pictus. Bull. Bwl. FranceBebique 85:373-375. Gans, C., and G. De Gueldre. 1992. Striated muscle: Physiology and functional morphology. pp. 277-313. In Enuirunmentalphyswlogy ofthe amphibians, edited by M. E. Feder and W. W. Burggren. Chicago: University of Chicago Press. Gardener, L. W, and A. M. Peadon. 1955. The role of the thyroid gland in direa development in the anuran, Ekutherodmtylus martinicensis. Gmwth 19:263-286. Garstang, W 1922. The theory of recapitulation: A critical restatement of the biogenetic law. J. Linnean Soc. Zool. 35 :8 1101. Garth, S., ed. 1961. Ooid'sMetammphoses, injfieen books. Translated by John Dryden et ai. New York: Hentage Press. Gartz, R. 1970. Adaptationsmorphologie der Melanophoren von Krdenfrosch-Larven. Cytabwlogie 2:220-234. Gascon, C. 1989a. The tadpole of Atelupus puuer Boulenger (Anura, Bufonidae) from Manaus, Amazonas. Rev.Brasil. Zool. 6:235-239. . 1989b. Predator-preysize interaction in tropical ponds. Rev.Brasil. Zool. 6: 701-706. . 1991a. Breeding of leptodmtylus kndeni: Responses to rainfd vanation. Copeia 1991:248-252. . 1991b. Population- and community-leve1analyses of species occurrences of central Amazonian rainforest tadpoles. Ecology 72: 1731-1746. . 1992a. Spatial distribution of Osteocephaus taurinus and Pipa avabali in a central Amazonian forest. Copeia 19923894-897. . 1992b. The effects of reproductive phenology on larval performance traits in a three-species assemblage of central Amazonian tadpoles. Oikos 65 :307-3 13. . 1992c. Aquatic predators and tadpole prey in central Amazonia: Field data and experimental manipulations. Ecology 73:971-980. . 1995. Tropical larval anuran fimess in the absence of direct effects of predation and competition. Ecology 76:2222-2229. Gascon, C., and J. R Bider. 1985. The effea of pH on buiifrog, Rana catesbeiana, and green fiog, Rana clamitans melanota, tadpoles. CanadianField Nat. 99:259-261. Gates, W R. 1983. A comparison of the embryonic and larval development of the southem chorus frog, Pm s* nigrita nigrita (LeConte), and the upland c h o m frog, P s e M s ~nieriataferarum(Baird), in Alabama. Thesis, Auburn University. Gatherer, D., and E. M. De1 Pino. 1992. Somitogenesis in the marsupial frog Gastrotheca riobambae. Intl. J. Develup. Bwl. 36:283-291. Gatten, R E., Jr., J. I? Caldweii, and M. E. Stockard. 1984. Anaerobic metabolism during intense swimming by anuran larvae. Hepetologica 40:164-169. Gatten, R E., Jr., K. Miiier, and R. J. F d . 1992. Energetics of amphibians at rest and during locomotion. pp. 314-377. In Environmental physwlogy of the amphibians, edited by M. E. Feder and W W Burggren. Chicago: University of Chicago Press. Gaupp, E. 1893. Beitrage zur Morphologie des Sch'adels. I: Primordial-Cranium und Kieferbogen von Ranafisica.Morphol. Arb. 2:275481.
'

371

. 1894. Beitrage zur Morphologie des Sch'adels. 11. Das Hyobranchial-Skelet der Anuren und seine Umwandlung. Morphol. Arb. 3:399437. . 1896. A. Ecker's und R. Wiedersheim's Anatomie des Frosches. Braunschweig: Vieweg und Sohn. Gaze, R. M., and P. Grant. 1992. Spatio-temporal pattems of retinal ganglion ceii death during Xenopus development. J. Comp.Neurol. 3 15:264-274. Gee, J. H., and R. C. Waidick. 1995. Ontogenetic buoyancy changes and hydrostatic control in larval anurans. Copeia 1995:861-870. Gegenbaur, C. 1898-1901. Kqkichende Anatomie &r Wirbel&e, mit Benksicbt&ung &r Wirbellosen, Vol. 2. Leipzig. Gehrau, A., and R. De Si. 1980. Comunicacion preliminar sobre larvas de L i m m d u s a mumoglossa (Amphibia:Leptodactyiidae). Res.J. Cienc.Nat. Montm'deo 1 :85-86. Grard, I?, and R. Cordier. 1933. Sur le mchanisme de transformation des nphrons ouverts en nphrons ferms chez Ia larve de DZFcoglossuspictus. C. R Acad. Roy. Bebique Cl. Sci. 19:508-512. . 1934a. Esquisse d'une histophysiologie compare du rein des vertbrs. Bwl. Rev. 9: 110-131. . 1934b. Recherches d'histophysiologie compare sur le pro- et le msonphros larvaires des anoures. Zeit. ZelIfmch. Mikros.Anat. 2 1:1-23. Gerhardt, E. 1932. Die Kiemenentwicklung bei Anuren (Pelobates fwcus, Hyla arborea) and Urodelen (Triton vubaris). Zool.Jahrb.Anat. 55 :173-220. Gerschenfeld, H. M., J. H. Tramezzani, and E. DeRobertis. 1960. Ultrastructure and function in neurohypophysisof the toad. Endom'nohgy 66:741-762. Gesner, C. 1551-1604. Htrcoria Animalium. Frankfurt: I. Wecheii. Ghate, H. V, and A. D. Padhye. 1988. Predation ofMimohyla tadpoles by Gambusia.J. Bombay Nat. H&. Soc. 85 :200-201. Giarretta, A. A. 1996. Reproductive specializations of the bromeiiad hylid frog Phyllodytes luteolus.J.Hepetol. 30:96-97. Giarretta, A. A., W C. A. Bokermann, and C. F. B. Haddad. 1993. A review of the genusMegaelosia (Anura: Leptodactylidae) with a description of a new species. J. Hepetol. 27:276-285. Giaretta, A. A., and L. K. Castanho. 1990. Nova espcie de Paratelmatobius (Amphibia, Anura, Leptodactylidae) da Serra do Mar, Brasil. Pap. Avul. Zool. 37:133-139. Gibley, C. W, Jr., and J. I? Chang. 1966. Fine structure of tubule ceiis in the functional pronephros ofRanapipiens. American Zool. 6:610. (Abstr.) Giesel, J. T. 1976. Reproductive strategies as adaptations to life in tempordy heterogeneous environments. Ann. Rev. Ecol. syst. 7:57-79. Giles, M. A., and E. M. Attas. 1993. Rare earth elements as intemal batch marks for rainbow trout -retention, distribution, and effects on growth of injeaed Dysprosium, Europium, and Samarium. Trans.A m e h n FZrh. Soc. 122: 289-297. Giii, D. E. 1978. The metapopulation ecology of the redspotted newt, Notophthalmus viridescens (Rafinesque). Ecol. Monogr 48: 145-166. Gines, H. 1958. Representantes de Ia familia Pipidae (Amphibia, Salientia) en Venezuela. Mem. Soc. Ciem. Nat. La Salh 18:5-18. Giorgi, I? I? 1974. Germ ceii migration in toad (Bufo bufo): Effea of ventral grafung of embryonic dorsal regions. J. Emblyol. Exp.Morphol. 31:75-87.

372

LITERATURE CITED

Gipouloux, J.-D. 1957. Le dveloppement prcoce du glornus est indpendant de Ia prsence du pronphros (Amphibiens, Anoures). C. R. Hebd. Seames Acad. Sci. Paris 244:10641065. . 1962. Mise en vidence du 'cytoplasrne gerrninal' dans I'oeuf et I'embryon du Discoglosse: Disco~lossuspic Otth. C. R. Hebd. SeancesAcad. Sci. Paris 254:2433-2435. . 1964. Une substance diffusible rnane des organes rnsodermiques dorsaux attire les ceilules germinales situes dans I'endoderme; dmonstration exprimentale chez le Crapaud commun, Bufo bufo L. C. R. Hebd. Seances Acad. Sci. Paris 259:3844-3847. . 1970. Recherches exprimentales sur Porigine, Ia migration des ceilules germinales, et I'dification des crtes gnitales chez les amphibiens anoures. Bull. Bwl. FranceBelgique 104:22-93. . 1975. "Cytoplasme germinal" et dtermination germinale chez les amphibiens anoures. Ann. Bwl. 14:475487. Gipouloux, J.-D., and R. Cambar. 1961. Recherches sur les facteurs dterminant Ia migration des cellules de l'uretre primaire (Amphibiens Anoures). C. R. Hebd. Seances A d . Sci. Paris 252:3643-3645. . 1970a. La migration des ceilules du blastrne de I'uretre primaire est indpendante de Ia prsence des somites, chez les embryons d'Amphibiens Anoures. C. R Hebd. Seancesdcad. Sci. Paris 270D: 372-373. . 1970b. Nouvelles prcisions sur Ia distance maximale laqueile I'uretre primaire exerce son action inductrice sur le blastrne msonphrogne chez les Amphibiens Anoures. C. R Hebd. SeancesAcad. Sci. Paris 270D:2329-2331. Gipouloux, J.-D., and M. Delbos. 1977. Aspects ultrastructuraux de I'interaction uretre-mesenchyme, au cours de Pinduction nphrogne, chez les amphibiens.J Emblyol. Exp. Mmphol. 41 :259-268. Gipouloux, J.-D., and C. Girard 1986. Organognse de I'appareil urognital et de I'organe interrnal. pp. 211-240. In Traite' de Zoolo~ie, Anatomie, Systematique, Bwlogie, Vol. 14, Fasc. lB, edited by P.-I? Grass and M. Delsol. Paris: Masson. Gipouloux, J.-D., C. Girard, and M. Delbos. 1978. Effets d'un traiternent par I'AMP cydique sur la migration des cellules gerrninales primordiales dans I'embryon de Xenupw W s . C. R. Hebd. SeancesAcad. Sci. Paris 287D: 1425-1427. Gipouloux, J.-D., and J. Hakim. 1976. Etude compare de Ia formation des bauches msonphrogenes rnddaire et interrnale c h e ~ ernbryons du crapaud comrnun, Bufo bufo les L., et du dactyltre du cap, Xenopus W Daudin. Bull. Bwl. France-Belgique 110:283-298. . 1978. Le dveloppernent de I'appareil uro-gnital aprs Ia greffe htrotopique du rnsoblaste latral chez Ia n e u d a du crapaud commun Rufo bufo L. (Amphibien Anoure). I. Inversions spares des axes antro-postrieur et dorsoventral de Ia lame latrale. J. Emblyal. Exp. Morphol. 43:137-146. Gitlin, D. 1944. The development of Eleuthwodacty1uspmtmicensis. Copeia 1944:91-98. Glaw, R., and M. Vences. 1992. Af;.Id&de to the amphibians and reptiles ofiL!ukgascar Leverkusen: Moos-Druck. Gluecksohn-Waelsch, S. 1987. Regulatory genes in development. Tvenak Genet. 3:123-127. Goater, C. E 1994. Growth and survival of postmetamorphic toads: Interactions among larval history, densiq, and parasitisrn. Ecology 75:2264-2274. Goater, C. P., and R. E. Vandenbos. 1997. Effects of larval his-

tory and lungworm infecxion on the growth and survival or juvende wood frogs (Rana sylvatica). Herpetalogica 53:331-338. Goette, A. 1874. Atlas zur Entwicklung der Unke. Arch. Mikros. Anat. 9. . 1875. D k Entwicklungsgeschicbte der Unke (Bombinator igneus). Leipzig: Verlag von Leopold Voss. Goicoechea, O., O. Garrido, and B. Jorquera. 1986. Evidence for a trophic paternal-larval relationship in the frog Rhinuderma danvinii.J. Herpetol. 20 :168-1 78. Goin, C. J. 1947. Studies on the life history of E l e u t h w o ~ l u s ricorditphnimstrti (Cope) in Florida: With special referente to the local distribution of an delomorphic color pattern. Univ.Florida Publ., Bwl. Scz. 4: 1-66. Goin, C. J., O. B. Goin, and G. R. Zug. 1978. IntroductMn to herpetology. San Francisco: W. H. Freeman. Goin, O. B., C. J. Goin, and K. Bachmann. 1968. DNA and amphibian life history. Copeia 1968:532-540. Goldacre, R. J. 1949. Surface fdms on natural bodies of water. J. Anim. Ecol. 18:36-39. G o h a n , B., and G. G o h a n . 1991. Embryoriic development of the mvobatrachine froes Geowinia k i s . Geocrinia victoriana, and their natural hybrids. ~m~hibia~Reptilia10312: 110. . 1993a. Hatching dynamics of larvae of Australian smooth froglets, Geocrinia l m s and Geocrinia victoriana (Amphibi& Anura, Myobatrachidae).Zool. Anz. 229: 191199. , 1993b. A laboratory experiment on interspeciic competition between tadpoles of Geocrinia victoriana and Pseudophlyne semimumata (Anura, Myobatrachidae). Amphibia-@tilia 14:349-356. . 1994. Life history variation across a hybrid zone in Geo&ia: Embryonic development and larval growth (Amphibia, Anura, Myobatrachinae). Acta Oecolo~ica15:247259. . 1995. Morphological variation in tadpoles of the Geocrinia h s cornplex: Regional divergente and hybridization (Amphibia, Anura, Myobatrachidae).J. Zool. Syst. Evol. Res. 33:32-41. . 1996. Geographic variation of larval traits in the Australian frog Geocrinia victm'ana. Herpetolooica 52: 181-187. Gona, A. G., K. F. Hauser, and N.J. Uray. 1982. Ultrastructural studies on Purkinje ceil maturation in the cerebellum of the frog tadpole during spontaneous and thyroxine-induced metamorphosis. Brain Behav. Evol. 20: 156-1 71. Goniakowska-Witainska,L. 1984. Tubular myelin structures in the lungs of Amphibia. The mode of formation. European J Cell Bwl. 33:127-133. . 1986. Lung of the tree frog, Hylaarborea L. A scanning and transmission electron microcopic study. Anat. Embryol. 174:379-389. Gonzalez, A., H. J. Ten Donkelaar, and R. De Boer-Van Huizen. 1984. Cerebeilar connections in Xenopw W s . An HRP study.Anat. Emblyol. 169:167-176. Goodyear, C. P., and R. Altig. 1971. Orientation of bulifrogs (Ranacutesbeiana)duringmetamorphosis.Copeia 1971:362264. Gordon, F. J. 1934. Preservation of s m d Amphibia in gelatin. Science 80:457. Gordon, J., and D. C. Hood. 1976. Anatorny and physiology of the frog retina. pp. 29-86. In The amphibian mm ystem: i 1 A multidisciplinaly approacb, edited by K. V. Fite. New York: Academic Press.
- -

LITERATURE CITED Gordon, M. S. 1962. Osmotic regulation in the green toad (Bufiviridis).J. Exp. Bwl. 39:261-270. Gordon, M. S., and V. A. Tucker. 1965. Osmotic regulation in the tadpoles of the crab-eating fiog (Ranacanm'vora).J Exp. Bwl. 42:437445. Gorham, E., S. J. Eisenreich, J. Ford, and M. V. Santelmann. 1985. The chemistry of bog waters. pp. 339-363. In Chemicnlprocesses in labes, edited by W. Stumm. New York: Wdey. Gorlick, D. L., and D. B. Keiiey. 1987. Neurogenesis in the vocalization pathway of Xenupus h i s . J. Comp. Neurol. 257:614-627. Gomer, P 1963. Untersuchungen zur Morphologie und Elek. trophysiologie des Seiteniinieorgans vom Kraenfrosches (Xenupushmis Daudin). Zeit. Ver~l. Physiol. 47:316-338. Gosner, K. L. 1959. Systematic variations in tadpole teeth with notes on food. Herpeto@ca 15:203-2 10. . 1960. A simpliied table for staging anuran embryos and larvae with notes on identification. Herpetoh~ica 16:183-190. Gosner, K. L., and I. H. Black. 1955. The effects of temperature and moisture on the reproductive cycle of Scaphiopw h. holbrooki. A m k a n Midl. Nat. 54: 192-203. . 1957a. Larvai development in New Jersey Hyiidae. Copeta 1957:31-36. . 195%. The effects of acidity on the development and hatching of New Jersey frogs. Eco&y 38:256-262. Gotte, S. W., and R. P Reynolds. 1998. Observations on the ef. fects of akohol vs. formalin storage of amphibian larvae. USGS Patuxent Wildlife Research Center homepage, <http: ,'/www.pwrc.nbs.gov/resshow/reynidlrs/amphhhd>. Gould, S. J. 1977. Ontogenyandphybeny. Cambridge: Harvard University Press. Gradwell, N. 1968. The jaw and hyoidean mechanism of the b&og tadpole during aqueous ventiiation. Canadian J Zool. 46:1041-1052. . 1969a. The function of the internai nares of the bullfiog tadpole. Herpeto@ca 25 :120-12 1. . 1969b. The respiratory importance of vascularization of the tadpole operculum in Rana catesbeiana Shaw. CanadianJ. Zool. 47: 1239-1243. 1 9 7 0 . The function of the ventral velum during gi irrigation in Rana catesbeiana. CanadianJ.Zwl. 48: 1179-1 186. . 1971a. rlscaphus tadpole: Experiments on the suaion and gi irrigation mechanism. Canadian J. Zool. 49:307332. . 1971b.Xempustadpole: On the water pumping mechanism. Herpetoh&ca 27:107-123. . 1972a. Comments on gdi irrigation in Rana&sc&uu. Herpetoh~ica 122-125. 28: . 1972b. Gdi irrigation in Rana catesbeiana. Part I. On the anatomicai basis. CanadianJ. Zwl. 50:481499. . 1972c. Gdi irrigation in Rana catesbeiana. Part 11. On the musculoskeletal mechanism. CanadianJ. Zool. 50:501521. . 1973. On the functionai morphology of suction and gdi irrigation in the tadpole ofrlscaphus,and notes on hibernation. Herpetoh~ica 84-93. 29: . 1974. Description of the tadpole ofl'hrynomerwannectem, and comments on its gdi irrigation mechanism. Herpeto150BKlt 30:53-62. . 1975a. The clinging mechanism of Pseuduphryne bibmni (Anura: Leptodactyiidae)to an alga on glass. Trans.Roy. Soc. South Australia 99:3 1-34. . 1975b. Experiments on oral suaion and gdi breathing

373

in five species of Australian tadpole (Anura: Hylidae and Leptodactylidae).J.Zool. 177:81-98. . 1975c. The bearing of iiter feeding on the water pumping mechanism of Xempus tadpoles (Anura: Pipidae). Acta Zool. 56: 119-128. Gradwell, N., and V. M. Pasztor. 1968. Hydrostatic pressures during normal ventiiation in the b&og tadpole. Canadian J. Zwl. 46: 1169-1174. Gradwell, N., and B. Waicott. 1971. Duai functionai and structurai properties of the interhyoideus musde of the bullfrog tadpole (Rana catesbeiana). Exp. Zwl. 176:193-218. J. Grafen, A. 1990. Do animais reay recognize kin?Anim. Behav. 39:42-54. Grandison, A. G. C. 1978. The occurrence of NectuphynuLhs (Anura: Bufonidae) in Ethiopia. A new concept of the genus with a description of a new species. Monit. Zool. Italiano 6: 119-172. . 1980. Aspects of breeding morphology in Mertensupbryne mimanotis (Anura: Bufonidae): Secondary sexual characters, eggs, and tadpole. BuU. British Mus. Nat. H&. (Zwl.) 39:299-304. . 1981. Morphology and phylogenetic position of the West African Didynam@usgoestedti Andersson, 1903 (Anura Bufonidae).Monit. Zool. Italiano (N.S.) Suppl. 15:187-21 5. Grant, E!, E. Rubin, and C. Cima. 1980. Ontogeny of the retina and optic nerve inXempus W s . I. Stages in the early development of the retina.J. Comp.Neurol. 189:593-613. Grant, S., and M. J. Keating. 1986. Ocular migration and the metarnorphic ari postmetamorphic maturation of the retinotectai system in Xempus W s : An autoradiographic and morphometric study.J. Embryol. Exp. MMmphol. 92:43-69. Grass, E!-P. 1986. La spermatognse. pp. 1-20. In Traitde Zwh~ie, Anatomie, Systematigue, Bw[oge, Vol. 14, Fasc. lB, edited by P.-P. Grass and M. Delsol. Paris: Masson. Grassi Miiano, E., and F. Accordi. 1983. Comparative morphology of the adrenai gland of anuran Amphibia. J. Anat. 136:165-174. . 1986. Evolutionary trends in the adrenai giand of anurans and urode1es.J. Morphol. 189:249-259. Grassi Miiano, E., F. Accordi, D. Antuzzi, and R. Rapisarda. 1979. Morphologic observations on the interrenai, Stiiiing and intermediary cells in the adrenai giand of Rana escuientacomplex. Riv. Bwl. 72:46-69. Gray, E. G. 1957. The spindle and extrafusai innervation of a frog muscle. Proc. Roy. Soc. h d o n 146B:416430. Gray, J. 1936. Studies in animal locomotion n The neuromus! d a r mechanism of swimming in the eel. J. Exp. BBi. 13:170-180. Gray, P 1930. The development of the amphibian kidney. Part . I. The development of the mesonephros of Rana temporavia. Quart.J. Micros. Sci. 73:507-546. . 1932. The development of the amphibian kidney. Part 11. The development of the kidney of Tviton vubavis and a comparison of this form with Rana temporavia. Quart. J. Micros. Sci. 75 :425465. . 1936. The development of the amphibian kidney. Part 111. The post-metamorphic development of the kidney, and the development of the vasa efferentia and seminal vesicles in Rana temporavia. Qwrt.J. Micros. Scz. 78 :445473. Graybeai, A., and D.C. Cannateiia. 1995. A new taxon of Bufonidae from Peru, with descriptions of two new species and a review of the phylogenetic status of supraspecific taxa. Herpetoh~0BKlt51:105-131. Green, N. B. 1938. The breeding habits of Pseudaris

374

LITERATURE CITED binocdar field and its tectal representation in Ranapipiens. J. Comp.Neurol. 190:175-185. Gromko, M. H., F. S. Mason, and S. J. Smith-Giii. 1973. Analysis of the crowding effect in Ranapipiens tadpoles. J. Exp. 2001. 186:63-72. Grover, B. G., and U. Grsser-Cornehls. 1984. Cerebellar afEerents in the frogs, Rana escubnta and Rana temporaria. CeU TZs. &S. 237:259-267. Grubb, J. C. 1972. DSerentiai predation by Gambusiaafinis on the eggs of seven species of anuran amphibians. American Midl. Nat. 88: 102-108. Gruberg, E. R., and S. B. Udin. 1978. Topographic projeaions between the nucleus isthmi and the tectum of the frog Rana pipiens.J.Comp.Neuml. 179:487-500. Grsser, O.-J., and U. Grsser-Cornehls. 1976. Neurophysiology of the anuran visual system. pp. 297-385. In Frog neumbwlogy, a handbook, edited by R. Llins and W. Precht. Berlin: Springer-Verlag. Guib, J., and M. Lamotte. 1958. Morphologie et reproduaion par dveloppement direct d'un Anoure du Mont Nimba, Arthrohptis erusculum Ang. Bull. Mus. d'Hst. Nat. Paris 30:125-133. Guilford, T. 1988. The evolution of conspicuous coloration. AmericanNat. 131S:7-21. Guillette, L. J., Jr. 1987. The evolution of viviparity in fishes, amphibians and reptiles: An endocrine approach. pp. 523562. In Hornwms and reproductwn in fishes, amphibians, and reptibs, edited by D. O. Norris and R. E. Jones. New York: Plenum. Gunasingh, A., S. Kasinathan, and S. Rarnakrishnan. 1982. Endocnne control of the cycle of oviduct and urogenitai papiiia of the tropical frog Rana hexadmtyla Lesson. IndianJ. Exp. Bwl. 20:285-289. Guraya, S. S. 1972. Histochemical observations of the testis of frog, Ranapipiens. Short communication.Acta Bwl.Acad. Sci. Hungaricu 23:327-329. . 1978. Maturation of the follicdar waii of nonmammalian vertebrates. pp. 261-330. In The vertebrate may. Comparative bwlogy and evolution, edited by R. E. Jones. New York: Plenum. Guaeit, E. 1889. Die Hornzahne der Batrachierlarven. Diss. Inaug. Zool. Univ. Albertina, Wilhelm Engelmann, Leipzig. Guyer, M. F. 1907. A simple method for rernoving the gelatinous coats of eggs. Amerieun Nat. 41 :400401. Guyetant, R. 1977a. Analysis of the group effect onAlytesobstetriuzns Laur. Ann. Sci. Univ.Besanum ( 14):11-21. - . 197%. Variations of the group-effect of Alytes obstetricans Laur. Ann. Sci. Univ.Besancon (14):23-38. Haas, A. 1992. Anatorny and nction of the buccal pump rnechanism in PhyUobates biwur (Anura: Dendrobatidae). Proc. Second Intem. Symp. Sondefmchungsbetzich 230: 87-92. . 1993. Das Cranium der Larve von PhyUobates bicour Bibron, 1841 (Anura: Dendrobatidae): Ein Beitrag zum phylogenetischen Systern der Anuran basierend auf Larvaimerkrnalen. Diplornarbeit, University of Tubingen. . 1995. Cranial features of dendrobatid larvae (Amphibia: Anura: Dendrobatidae).J. Mmphol. 224:241-264. . 1996a. Nonfeeding and feeding tadpoles of herniphractine frogs: Larval head morphology, heterochrony, and systernatics of Fbctonotztsgoeldii (Amphibia: Anura: Hylidae).J. Zool. Syst. Evol. Res. 34: 163-171. . 1996b. Das larvale Craniurn von Gastrotheeuriubambu und seine Metamorphose (Amphibia, Anura, Hylidae). Erh. Natunviss. Vm Hamburg 36:33-162.

brachyphona (Cope) with a descnption of the eggs and tadpole. Copeia 1938:79-82. Greil, A. 1905. Ueber die Genese der Mundhohlenschleimhaut der Urodelen. Anat. Anz. Suppl. 27, Verh.Anat. Ges. 19:2537. Greven, H. 1980. Ultrastructural investigations of the epidermis and the giii epithelium in the intrauterine larvae of Salamandra salamundra (L.)(Amphibia, Urodela). Zeit. Mikros. Anat. Fonch. 94: 196-208. GriKths, I. 1961. The form and h c t i o n of the fore-gut in anuran larvae (Amphibia, Salientia) with particular reference to the manicotto oglandulare. Proc. Zool. Soc. Lonhn 137:249283, 6 plates. . 1963. The phylogeny of the Salientia. Bwl. Rev. 38 :24 1-292. Grifiths, I., and A. L. De Carvalho. 1965. On the validity of employing larval characters as major phyletic indices in Amphibia, Salientia. Rev. Brasil. Bwl. 25: 113-121. Grifiths, R. A. 1986. Feeding niche overlap and food seleaion in srnooth and paimate newts, Tritumw b a k and ?: helveticu4 at a pond in mid-Wales.J. Anim. Ecol. 55:201-214. . 1991. Cornpetition between common frog, Rana temporaria, and natterjack toad, Bufo calamita, tadpoles: The effect of cornpetitor density and interaaion level on tadpole developrnent. Oikos 6 1:187-196. . 1995. Determining cornpetition rnechanisrns in tadpole assemblages. Herpeto1.J.5:208-210. Grifiths, R. A., and J. Denton. 1993. Interspecific associations in tadpoles. Anim. Behav. 44: 1153-1157. Grifiths, R. A., J. Denton, and A. L.-C. Wong. 1993. The effect of food level on competition in tadpoles: Interferente mediated by protothecan algae?J.Anim. Ewl. 62:274-279. GriKths, R. A., P. W. Edgar, and A. L.-C. Wong. 1991. Interspecific cornpetition in tadpoles: Growth inhibition and u growth retrieval in natterjack toads, B ? caiumita.J. Anim. Ewl. 60: 1065-1076. Grifiths, R. A., L. Schley, P. E. Sharp, J. L. Dennis, and A. Rornn. 1998. Behavioural responses of Mdorcan midwife toad tadpoles to natural and unnatural snake predators. Anim. Behav. 55 :207-214. Griiiitsch, B. 1992. Notes on the tadpole of Phynohyas resinzjctrjc (Goeldi 1907). Buccopharyngeal and external rnorphology of a tree hole dweiiing larva (Anura, Hylidae). Herpetoma 5:5 1-66. Griiiitsch, B., and H. Grillitsch. 1989. Teratological and ontogenetic alterations to extemal oral structure in some anuran larvae (Amphibia: Anura: Bufonidae, Ranidae). Fortschrit. Zool. 35:276-282. Griilitsch, B., H. Grillitsch, and H. Splechtna. 1988. Zur Morphologie der larve von Tomoptema nyptotis (Bodenger 1907)(Amphibia:Anura: Ranidae). Herpetowa 1:31-46. Griitsch, B., H. Grillitsch, M. Haupl, and F. Tiedemann. 1983. Lunhe und Kriechtiewe Neiabosteweichs. Wien: Facultas-Verlag. Griner, S. 1974. On the generation of locornotion in the spinal dogfish. l%p.Brain Res. 20:459470. Griner, S., and S. Kashin. 1976. On the generation and perforrnance of swimrning in fish. pp. 181-201. In Neural control ofucomutwn, edited by R. M. Herrnan, S. Griner, P. S. G. Stein, and D. G. Stuart. New York: Plenum. Grobstein, P., and C. Comer. 1983. The nucleus isthmi as an intertectal relay for the ipsilateral oculotectal projeaion in the frog, Ranapipiens. J. Comp.Neurol. 2 17:54-74. Grobstein, P., C. Comer, and S. Kostyk. 1980. The potential

LITERATURE CITED

375

. 1997a. The larval hyobranchial apparatus of discoglossoid fiogs: Its structure and bearing on the systematics of the Anura (Amphibia: Anura). J. 2001. Syst. Evol. Res. 35: 179-197. . 199%. Der larvale Hyobranchial-ApparatvonAscaphus tnui (Anura: Ascaphidae) und das Grundmuster der Anura. ~~crhahungen h-~eutschen Zoologischen Geselhchaft, Mainz 90: 172. Haertel, J. D., and R. M. Storm. 1970. Experimental hybridization between Ranapetwsa and Rana mcadue. Herpetologica 26:43M6. H&, J. H. 1974. Alkaline phosphatase activity in the developing pronephros and mesonephros of the frog Bombina orid KmeanJ. Zool. 17:177-184. . IIilman, J. P. 1976. Oii droplets in the eyes of adult anuran amphibians:A comparative study.J.Morphol. 148:453468. Hairston, N. G. 1989. Hard choices in ecological experimentation. Herpetologica 45 :119-122 Hakim, J., and J.-D. Gipouloux. 1978. Le dveloppement de Fappareil urognital aprs Ia greffe htrotopique du msobiaste latral chez la neuruia du crapaud commun Bufo bufo L (Arnphibien Anoure). 11. Inversion simultane des axes mtro-postrieur et dorso-ventral de Ia lame latrale. J. Emtnyol. l p Morphol. 44:113-119. %. HaIi, J. A., J. H. Larsen, Jr., and R E. Fitzner. 1997. Postembryonic ontogeny of the spadefoot toad, Scaphwpus intemnmnus (Anura: Pelobatidae): External morphology. Herpetol. Mimg1: (11):124-178. Eia4 J. A., J. H. Larsen, Jr., D. E. Miller, and R. E. Fitzner. 1995. Discrimination of kin- and diet-based cues by larval spadefoot toads, Scaphwpus intemntanus (Anura: Pelobatidae), under laboratory conditions.J. Herpetol. 29:233-243. IW1, J. C., and A. S. Feng. 1987. Evidence for pardel processing in the frog's auditory thalamus. J. Comp. Neurol. 258:407419. H d , R W. 1904. Development of the mesonephros and the Mullerian ducts of Amphibia. Bull. Mus. Comp. 2001. 45:29-125. Hamburger, V. 1988. The heritage o tzxperimental embryology: f H m Spemann and the organizei: New York: Oxford University Press. Eiamiiton, W D. 1964a. The genetical evolution of social bek o r I, 11.J. Thm. Bwl. 7: 1-52. . 1964b. The genetical evolution of social behaviour. I. J Bm Bwl. 7:l-16. e 1 9 6 4 ~The genetical evolution of social behaviour. I.J. . Thcor: Bwl. 7: 17-52. . 1971. Geometry for the selfish herd. J. Theol: Bwl. 31:295-311. Hmimerman, D. L. 1965. Development of the tongue of Rana dunitans. Ameriutn 2001.5 :250-25 1. (Abstr.) . 1969a. The frog tongue: I. General development and histogenesis of iiiform papillae and mucous giands in Rana atesbeiana. Acta 2001.50: 11-23. . 1969b. The frog tongue: 11. Histogenesis of fungform papillae in Rana catesbeiana. Acta 2001. 50:25-33. . 1969c. The frog tongue: 111. Histogenesis and regeneraion following complete and partial extirpation of anlagen. Ada 2001. 50:215-232. HImmerschlag, R., S. Maines, and M. Ando. 1989. Sensory &a fiom tadpoles but not adult bulifrogs synthesize heat &o&-like proteins in vitro at non-heat shock temperatures. J ATmrosci.Res. 23:4 16424. Ehnpton, S. H., and E. E Volpe. 1963. Development and in-

terpopulation variability of the mouthparts of Scaphwpus holbrooki.Ammican Midl. Nat. 70:3 19-328. Hanke, W. 1976. Neuroendocrinology. pp. 975-1022. In Frog neuroendominology,edited by R Llins and W. Precht. Berlin: Springer-Verlag. Hanken, J. 1983. Miniaturization and its etfects on cranial morphology in plethodontid salamanders, genus Thmius (Amphibia, Plethodontidae): 11. The fate of the brain and sense organs and their role in skul morphogenesis and evolution. J.Morphol. 177:255-268. . 1984. Miniaturization and its effects on cranial morphology in plethodontid salamanders, genus Thmius (Amphibia: Plethodontidae). I. Osteological variation. Bwl. J. Linnean SOL.23:55-75. . 1986. Developmental evidence for amphibians origins. pp. 389417. In Evolutwnary bwlogy, Vol. 20, edited by M. K. Hecht, B. Waace, and G. T. Prance. New York: Plenum. Hanken, J., and B. K. Ha. 1984. Variation and timing of the cranial ossification sequence of the Oriental fire-beliied toad, Bombinu mientalis (Amphibia, Discogiossidae). J. Movphol. 182:245-255. . 1988. Skul development during anuran metamorphosis: I. Early development of the first three bones to formthe exoccipital, the parasphenoid, and the frontoparietal. J. Mmphol. 195:247-256. Hanken, J., D. H. Jennings, and L. Olsson. 1997. Mechanistic basis of life-history evolution in anuran amphibians: Direct development.A k a n 2001. 160-171. 37: Hanken, J., M. W. Klymkowsky, C. H. Summers, D. W. Seufert, and N. Ingebrigtsen. 1992. Cranial ontogeny in the direct-developing frog, Eleutherodactylus cogui (Anura: Leptodactylidae), analyzed using whole-mount immunohistochemistry. J. Morphol. 21 1:95-118. Hanken, J., and C. H. Surnrners. 1988. Developmental basis of evolutionary success: Cranial ontogeny in a directdeveloping anuran. American Zool. 28: 12. (Abstr.) Hanken, J., and D. B. Wake. 1993. Miniaturization of body size: Organismal consequences and evolutionary significance. Ann. &v. Ewl. Syst. 24:501-5 19. Hardy, J. D., Jr. 1984. Frogs, egg teeth, and evolution: Preliminary conunents on egg teeth in the genus Eleutherodactylw. Bull. Maryiund Herpetol. Soc. 20: 1-1 1. Harkey, G. A., and R. D. Semiitsch. 1988. Effects of temperature on growth, development, and color polymorphism in the ornate c h o m fiog Pseudacris omata. Cupeia 1988:10011007. Harpur, R E 1968. Osmoregulation et metabolisme de I'uree: Cornparison entre Bufo vi& et Rana temporaria. Canudian J. 2001.46:295-301. Harris, R. N. 1987. Density-dependent paedomorphosis in the salamanderNotuphthalmusviridesgns abrsalis. Ewlogy 68:7057 1?

LL.

. 1989. Ontogenetic changes in size and shape of the facultatively paedomorphic salamander Notuphthalmus v i d a censhalis. Cupeia 1989:3542. Harris, R. N., R. A. Aiford, and H. M. Wiibur. 1988. Density and phenology of Notuphthalmusviridescem h a l i s in a natural pond. Herpetologia 44:234-242. Harris, R. N., R. D. Semiitsch, H. M. Wdbur, and J. E. Fauth. 1990. Local variation in the genetic basis of paedomorphosis in the salamander Am6ystoma talpoideum. Evolutwn 44: 1588-1603. Harrison, J. D. 1987. Food and feeding relations of common

376

LITERATURE CITED embryonic development in Sywhophus qst&nathoides campi (Anura: Leptodactylidae). Southwest. Nat. 35 :358-361. Hazard, E. S., 111, and V. H. Hutchison. 1978. Ontogenetic changes in erythrocyuc organic phosphates in the buiifrog, Rana catesbeiana.J.Exp. Zool. 206:109-1 18. Hearnden, M. N. 1992. The reproductive and larval ecology of Bufo marinus (Anura: Bufonidae). Dissertation, James Cook University, Townsville, Queensland, Australia. Heasman, J., R. O. Hynes, A. P. Swan, V. Thomas, and C. C. Wylie. 1981. Primordial germ cells of Xenopus embryos: The role of fibronectin in their adhesion during migration. Cell 27:437-447. Heasman, J., T. Mohum, and C. C. Wylie. 1977. Studies on the locomotion of primordial germ cells from &nopus laet.is in vitvo.J. Embryol. Exp. Mmphol. 42:149-161. Heasman, J., and C. C. Wylie. 1978. Elearon microscopic studies on the stnicture of motile primordial germ cells ofXenopus laeviis in vitvo.J. Embryol. Exp. Mmphol. 46: 119-133. Heasman, J., C. C. Wylie, and S. Holwill. 1985. The importance of basement membranes in the migration of primordial germ ceUs in Xenopus laevis. Develop. Bwl. 112:18-29. Heathcote, R. D., and A. Chen. 1991. Morphogenesis of adrenergic ceUs in a frog parasympathetic ganghon. J. Comp. Neurol. 308:139-148. Heatwole, H. 1962. Contributions to the natural history of Eieutherodactyluscornutus maussi. Stahlia (2):1-1 1. Heatwole, H., S. B. De Austin, and R. Herrero. 1968. Heat tolerantes of tadpoles of two species of tropical anurans. Comp. Bwchem. Physwl. 27:807-815. Hegner, R. W. 1922. The effects of changes in diet on the incidente, distribution and numbers of certain intestinal Protozoa of frog and toad tadpo1es.J. Parasitol. 9:51-67. Heisler, N. 1986. Acid-base regulation in fishes. pp. 309-356. In Acid-base reguiution in animalr, edited by N. Heisler. Amsterdam: Elsevier. Heisler, N., G. Forcht, G. R. Ultsch, and J. F. Anderson. 1982. Acid-base reguiation in response to environmental hypercapnia in two aquatic salamanders, Siren hvtz'na and Arnphiuma means. Respir: Physwl. 49:141-1 58. HeM, O. M. 1932. Studies on amphibian metamorphosis. X. Hydrogen ion concentration of the blood of anuran larvae during involution. Biol. Bu11. 63:405418. HeM, O. M., and M. C. Meilicker. 1941. Studies on amphibian metamorphosis. XM. Development of the tongue in Rana sylvatica, includig histogenesis of "premetamorphic" and filiform papillae and the mucous glands. AmericanJ. Anat. 68:339-369. HeH, O. M., and K. I. Stubblefield. 1931. The influente of oxygen tension on the oxygen consumption of Ranapipiens larvae. Physwl. Zool. 4:271-286. Hendricks, F. S. 1973. Intestinal contents of Rana pipiens Schreber (Ranidae) larvae. Southwest.Nat. 18:99-10 1. Hendrickson, A. E., and D. E. Kelly. 1971. Development of the amphibian pineal organ; ine structure during maturation. Anat. Rec. 170:129-142. Henrickson, B.-I. 1990. Predation on amphibian eggs and tadpoles by common predators in acidified lakes. HoiurcticEcol. 13:201-206. Hensley, F. R. 1993. Ontogenetic loss of phenotypic plasticity of age at rnetamorphosis in tadpoles. Ecology 74:2405-2412. Hero, J.-M. 1990. An illustrated key to tadpoles occurring in the central Amazona rainforest, Manaus, Amazonas, Brasil. Amazoniana 11:201-262.

frog and common toad tadpoles (Ranatempmaria and Bufo bufo) at a pond in mid-Wales. Herpeto1.J. 1:141-143. Harrison, P. H. 1990. Induction of locomotion in spinal tadpoles by exutatory amino acids and their agonists. J. Exp. Zool. 254:13-17. Harrison, R. G. 1904. Experimentelle Untersuchungen uber die Entwiddung der Sinnesorgane der Seitenlinie bei den Amphibien. Arch. Mik-is. Anat. Entwickl. 63:35-149. Hart, M. W. 1996. Evolutionary loss of larval feedig: Development, form and funaion in a facultatively feedig larva, Brisaster iutzj+ons.EvolutMn 50:174-1 87. Hartl, D. L., and A. G. Clark. 1989. Principies ofpopuhtwngenecics. Sunderland, MA: Sinauer. Harvey, P. H., and P. J. Greenwood. 1978. Anti-predator defense strategies: Some evolutionary problems. pp. 129-1 51. In Bebavioral ecology:A n evolutwnary approach, edited by J. R. Krebs and N. B. Davies. Oxford: Blackweil Sci. Publ. Harvey, P. H., and and M. D. Pagel. 1991. The comparative method in evolutwnary bwlogy. Oxford: Oxford University Press. Hassinger, D. D. 1972. Early life history and ecology of three congeneric species ofRana in New Jersey. Dissertation, Rutgers University. Hastings, D., and W. Burggren. 1995. Developmental changes in oxygen consumption regulation in larvae of the South African clawed frog Xenopus laet.is.J. Exp. Bwl. 198:2465-2475. Hathaway, E. S. 1927. Quantitative study of the changes produced by acclimatization in the tolerante of high temperatura by fishes and amphibians. Bu11. United States Bur: Fish. 43:169-192. Hauser, K. F., N. J. Uray, and A. G. Gona. 1986a. Stellate ceil development in the frog cerebellum during spontaneous and thyroxine-induced metamorphosis. J. Comp. Neurol. 244:229-244. . 1986b. Granule cell development in the frog cerebellum during spontaneous and thyroxine-induced metarnorph0sis.J. Comp.Neurol. 253:185-196. Havenhand, J. N. 1993. Egg to juvenile period, generation time, and the evolution of larval type in manne invertebrates. Mar. Ecol. Progr: Ser: 97:247-260. H a w h s , C. P., L. J., Gottschalk, and S. S. Brown. 1988. Densities and habitat of taiied frog tadpoles in s m d streams near Mt. St. Helens foilowing the 1980 eruption.J. NmthAmerican Benthol. Soc. 7:246-252. Hay, J. M., I. Ruvinsky, S. B. Hedges, andL. R. Maxson. 1995. Phylogenetic relationships of amphibian families inferred from DNA sequences of mitochondrial 12s and 16s ribosomal RNA genes. Mol. Bwl. Evol. 12:928-937. Hayes, T. B. 1995. Interdependence of corticosterone and thyroid hormones in larval toa& (Bufo boreas). 1. Thyroid hormone-dependent and independent effeas of cortiocosterone on growth and deve1opment.J Exp. Zool. 271:95-102. Hayes, T. B., R. Chan, and P. Licht. 1993. Interactions of ternperature and steroids on larvai growth, developrnent, and metamorphosis in a toad (Bufo boreas). J. Exp. Zool. 266:206-215. Hayes, T. B., and P. Licht. 1995. Faaors influencing testosterone metabolisrn in anuran 1arvae.J. Exp. Zool. 271:112-119 Hayes, T. B., and T. H. Wu. 1995. Interdependence of corticosterone and thyroid hormone in toad larvae (Bufoborem). 2. Regulation of corticosterone and thyroid hormones.J. Exp. ZOO!!. 271:103-111. Hayes-Odum, L. A. 1990. Observations on reproduction and

LITERATURE CITED . 1991. Predation, palatability, and the distribution of tadpoles in the Amazon rainforest. Dissertation, Griffith University, Brisbane, Queensland, Australia. Hero, J.-M., and A. Mijares-Urrutia. 1995. The tadpole of Scinax mstrata ( h a : Hylidae).$ Heyletol. 29:307-311. Hhn-Royer, and C. Van Bambeke. 1881. Sur les caractkres fournis par la bouche des tttards des batraciens anoures #Europe. BuU. Soc. Zol. France 6:75-81. . 1889. Le vestibule de la bouche chez les t&ards des batraciens anoures #Europe. Arch. B d . 9:185-309, plates 12-24. Herreid, C. F., 11, and S. Kinney. 1966. Survival of Alaskan woodfrog (Ram sylvatica) larvae. Ecology 47: 1039-1041. . 1967. Temperature and development of the woodfrog, Ram sylvatica, in Alaska. Ecology 48: 579-590. Herrick, C. J. 1909. The nervus terminalis (nerve of Pinkus) in the frog.$ Comp.Neurol. 19:175-190. . 1948. The brain ofthe t&wsalama& Chicago: University of Chicago Press. Hetherington, T. E. 1987. Timing of development of the middle ear of Anura (Amphibia). Zoommpholoogy 106:289300. . 1988. Metamorphic changes in the middle ear. pp. 339-357. In The evolutwnofthe amphibianauditolysystem, edited by B. Fritzsch, M. J. Ryan, W Wilczynski, T. E. Hetherington, and W Walkowiak. New York: Wiley. Heusser, H. 1970a. Spawn eating by tadpoles as possible cause of specific biotope preferences and short breeding times in European anurans (Amphibia Anura). Oemloogia 4:83-88. . 1970b. Laich-Fressen durch Kaulquappen als mogliche Ursache spezifischer Biotopprgerenzen und irurzer Laichzeiten bei europiiischen Froschlurchen (Amphibia, Anura). Oemloga 4:83-88. . 1971. Laich-Raubern und Kannibalismus bei sympamsher Anuran-Kaulquappen. Expen'entia 27:474475. Hemltt, J. 1927. Further descriptions of reptiles and batrachians from South Africa. &c. Albany M s . 3:371415. Hews, D. K. 1988. Alarm response in larval western toads, Bufo bmeas: Release of larval chemicals by a natural predator and its effect on predator capture efficiency. Anim. Behm. 36:125-133. . 1995. Overall predator feeding rates and relative susceptibility of large and small tadpoles to fish predation depend on microhabitat: A laboratory study. J. Hqetol. 29: 142-145. Hews, D. K., and A. R. Blaustein. 1985. An investigation of the alarm response in Bufo bmeas and Rana cuscadae tadpoles. Behav. Neural Bwl. 43 :47-57. Heyer, W. R. 1969. The adaptive ecology of the species groups of the genus Lepto+hs (Amphibia, Leptodactylidae). Emlutwn 23:421428. . 1971. Descriptions of some tadpoles from Thailand. Fkldiana Z d . 58:83-91. . 1973a. Systematics of the mamratus group of the frog genus Leptodmtylus (Amphibia, Leptodactylidae). Nat. Hist. Mw. b s Anogeles Co. Con&. Sci. (251):1-50. . 1973b. Ecological interactions of frog larvae at a seasonal tropical location in Thailand.$ He'petol. 7~337-361. . 1974. Niche measurements of frog larvae from a seasonal tropical location in Thailand. Ecology 55651-656. . 1976a. Studies in larval amphibian habitat partitioning. Smithsonian Contrib. Zool. (242) :1-27.

377

. 1976b. The presumed tadpole of Paratelmatobiuslutzi (Amphibia, Leptodactylidae). Pap. Aml. Zool. 30: 133-135. 1 9 7 9 . Annual variation in larval amphibian populations withii a temperate pond.$ WihinngtonAcad.Sci. 69:65-74. . 1983a. Variation and systematics of frogs of the genus Cycloramphus (Amphibia, Leptodactylidae). Arq. Zool. Mw. Unia. S& Paulo 30:235-239. . 1983b. Notes on the frog genus C~cIOPamphzrs(Amphibia: Leptodaaylidae), with descriptions on two new species. Pmc. BBi. Soc. Washinngton96:548-559. . 1985. Taxonomic and natural history notes on frogs of the genus Centmhelh (Amphibia: Centrolenidae) from southeastern Brasil and adjacent Argentina. Pap. A d . Zool. 36: 1-21. Heyer, W R., M. A. Domelly, R. W. McDiarmid, L.-A. C. . Hayek, and M. S. Foster, eds. 1994.ikfeasun'n~ m i & and z lrioloogical div*. w Standard methods fm amphibians. Biological Diversity Handbook Series. Washington, D.C.: Smithsonian Institution Press. Heyer, W R., R. W McDiarmid, and D. L. Weigmann. 1975. . Tadpoles, predation, and pond habitats in the tropics. Bw7:lOO-111. Heyer, W R., A. S. Rand, C. A. G. Da Cnrz, 0 . L. Peixoto, and . C. E. Nelson. 1990. Frogs ofBoracLia.Aq. Zool., Mus. Zool, Uiziv.S& Paulo 31:231410. Heyer, W R., and l? A. Silverstone. 1969. The larva of the frog . LL.po-lus @faedanylus (Leptodactylidae). Fkldiana Zool. 51:141-145. Hibiya, T. 1982.A n atlas offish his to loo,^. Normul andpatholo~ical features. Stuttgart: Gustav Fischer-Verlag. Higgins, G. M. 1920. The nasal organ in Amphibia. IZlimisBwl. Mmogr 6:3-90,6 plates. Hiltis, D. M. 1982. Morphological differentiation and adaptation of the larvae of Rana bwhndieri and Rana sphemcephala (Ranap@ns complex) in sympatry. Cqeia 1982:168-174. Hiitis, D. M., and R. 0.De SI. 1988. Phylogeny and taxonomy of the Rampalmt.pes group (Salientia: Ranidae). Hmpetol. Mamogr (2):1-26. Hillis, D. M., and J. S. Frost. 1985. Three new species of leopard frogs (Ramppiens complex) from the Mexican Plateau. O m . Pap. Mw. Nat. Hist., Unirz Kansas (1 :1-14. 17) Hinckley, M. H. 1880. Notes on the eggs and tadpoles o f H y h vmicoIm: Pmc. Boston Soc. Nat. Hist. 21: 104-107, plate 3. . 1881. On some differences in the mouth structure of the anourous batrachians found in Milton, Mass. Proc. Boston SOL. Hist. 21:307-315, plate 5. Nat. . 1882. Notes on the development ofRana sylvatica Leconte. Pmc. B o r n SOC. Hist. 22: 104-107. Nat. . 1884. Notes on the peeping frog, Hylapickeringi, Leconte. Mem. Boston Soc. Nat. Hist. 3~311-318,plate 28. Hirakow, R. 1989. Origin and differentiation of the chordate heart. pp. 261-263. In Tmnds in vevtebrate mmpholofiy.Pmogr Z d . 35, edited by H. Splechtna and H. Hilgers. Stuttgart: Gustav Fischer-Verlag. Hirakow, R., S. Komazaki, and T. Hirurna. 1987. Early cardiogenesis in the newt embryo. Scan. Minzrs. 1:1367-1376. Hirakow, R., and X Sugi. 1990. Intercellular junction and cytoskeletal organization in embryonic chick myocardial cells. pp. 95-1 13. In Dmelopmental cardwloogy: Mmphogenesis and fiknctian, edited by E. B. Clark and A. Takao. Mount Kisco, Fuma. Hird, D. W., S. L. Diesch, R. G. McKinnell, E. Gorham, F. B. Martin, S. W Kurtz, and C. Dubrovolny 1981. Aemmonas .

378

LITERATURE CITED

hydrophila in wild-caught frogs and tadpoles (Ranapiptevu) in Minnesota. Lab.Anim. Sci. 3 1: 166-169. Hirsch, G. C. 1938. Die Nieren der Arnphibien. pp. 775-805. In Handbuch der vey~hichenden Ana& der Wirbeltiere, Vol. 5, edited by E. B. Clark and A. Takao. Berlin: Urban and Schwarzenberg. Hisaoka, K. K., and J. C. List. 1957. The spontaneous occurrence of scoliosis in larvae of Rana sylaatica. Trans.American Mimos. Soc. 76:38 1-387. Hochachka, P. W. 1988. Metabolic suppression and oxygen availability. CanadianJ Zool. 66: 152-1 58. Hochachka, P. W., and G. N. Sornero. 1984.Bwchemdaduptatwn. Princeton: Princeton University Press. Hoegh-Guldberg, O., and R. B. Emlet. 1997. Energy use during the developrnent of a lecithotrophic and a planktotrophic echinoid. Bwl. Bull. 192:27-30. Hoff, K. 1986a. Undulatory swimming in anuran larvae without waves of muscular contraction.J.Bwmech. 19:478-479. . 1986b. Morphologicai correlates of fast-start perforrnance in anuran larvae. A m e k n Zool. 26: 132. (Abstr.) . 1988a. Evasive rnaneuvers of frogs and tadpoles in cornplex environrnents. Ammkan Zool. 28: 155 (Abstr.). . 1988b. Morphologicai determinants of fast-start perforrnance in anuran tadpoles. Dissertation, Dahousie University. . 1989. Changes in evasive rnaneuvers during rnetamorphosis in bullfrogs.American Zool. 29:43 (Abstr.). Hoff, K., and D. Ingle. 1988. Frogs have short terrn rnemory of obstacles. Soc. Neurosci.Abst7: 14:692. Hoff, K. vS., V. A. King, N. Huq, and R. J. Wassersug. 1989. Kinematics of larvai saiamander swimrning (Ambystornatidae). CanadianJ.Zool. 67:2756-2761. Hoff, K., and R Wassersug. 1985. Do tadpoles swirn like fishes? pp. 31-34. In Functional mrphology in vertehates. Fmtschrit. Zool. 30, edited by H.-R. Duncker and G. Fleischer. Stuttgart: Gustav Fischer-Verlag. . 1986. The kinernatics of swimming in larvae of the clawed frog, Xempus luevis.J. Exp. B d . 122:1-12. Hofrnann, M. H., and D. L. Meyer. 1989. Central projections of the nervus terminalis in four species of amphibians. Brain Behav. Evol. 34:301-307. Hokit, D. G., and A. R. Blaustein. 1994. The effects of kinship on growth and developrnent in tadpoles of Rana ms&. Evolutwn 48:1383-1388. . 1995. Predator avoidance and aiarm-response behaviour in kin-discriminating tadpoles (Rana cm&). Etholagy 101:280-291. . 1997. The effects of kinship on interaction between tadpoles of Rana cmcadue. Emlofiv 78: 1722-1 735. Holder, N.J., I. D. W. Clarke, and D. Tonge. 1987. Pathfkding by dorsal column axons in the spinai cord of the frog tadpole. Devehpent 99:577-587. Hoiyfield, J. G. 1966. The origin of erythroblasts in Ranapip n s tadpoles. Develop. Bwl. 14:461480. Holornuzki, J. R. 1998. Grazing effects by green frog tadpoles (Ranaclamitavu)in a woodand p0nd.J Freshw.E d . 13:1-8. Honda, E., T. Hirakawa, and S. Nakamura. 1992. Tactile and taste sensory rnaturation in the frog lamae tongue. Brain Res. 595:74-78. Honegger, R. E., C. Schneider, and E. Zimmerrnan. 1985. Notizen zur aufiucht von Schmuckhornfroschen Ceratophys omata (Beli 1843). Salamandra 21 :70-83. Hoogmoed, M. S., and J. E. Cade. 1991. Natural history and

distribution of &dychnis crmpehpus (Funkhouser 1957) (Amphibia: Anura: Hylidae). Zool. Med. 65:129-142. Hoppe, D. M. 1978. Therrnai tolerante in tadpoles of the chom frog Pseudumis triim'ata. Herpetolofiica 34:3 18-321. Hora, S. L. 1923. Further obsemations on the oral apparatus of the tadpoles of the genus Mgophys. Rec. Indian Mus. 30: 139-145. Horat, E, and R. D. Semlitsch. 1994. Effects of predation risk and hunger on the behaviour of two species of tadpoles. Behav. Ecol. Socwbwl. 34:393-401. Horiuchi, S., and Y Koshida. 1989. Effects of foodstuffs on . intestinal length in larvae of Rhacophorus arboreus (Anura: Rhacophoridae). Zool. Sci. 6:321-328. Horsernan, N. D., A. H. Meier, and D. D. Cuiiey, Jr. 1976. Daily variations in the effects of disturbance on growth, fattening and rnetamorphosis in the buiifrog (Ranacatesbeiana) tadpo1e.J. Exp. 2001. 198:353-357. Horton, J. D. 1971a. Ontogeny of the imrnune systern in amphibians. A m C a n Zool. 11:219-228. . 1971b. Histogenesis of the lymphornyeloid cornplex in the larvai leopard frog, Ranaptpiens. J Morphol. 134:l-20. Horton, P. 1983. Reproduaive systern. pp. 84-92. In Thegmtri'c hoodinfifrog, edited by M. J. Tyler. Beckenham: Croom Helm, Ltd. Hosoi, M., S. Niida, Y. Yoshiko, S. Suernune, and N. Maeda. 1995. Scanning elearon microscopy of horny teeth in the anuran tadpole Rhacophoridae, Rhacophams arbmeus and Rhacophams schhgelii.J. Ekctron Mimos. 44:351-357. Hota, A. K., and M. C. Dash. 1981. Growth and rnetamorphosis ofRana&>rina larvae: Effects of food leve1 and larvai density. Oihos 37: 349-352. . 1983. Evidence of interspecific predation among larvai anurans: Predation of Rana &@na larvae on Buf melanostictus larvae. Bwl. Bull. India 5 :54-55. . 1986. Growth and rnetamorphosis of anuran larvae: Effect of diet and temperature.Alpes 51165-172. Hourdry, J. 1974. Etude des branchies "internes" puis de leur rgression au rnornent de Ia mtamorphose, chez Ia larve de Dismglossus pictus (Otth), amphibien anoure. J Micros. 20: 165-182. . 1993. Passage to the terrestrial life in amphibians. 2. Endocrine detenninism. Zool. Sci. 10:715-73 1. Hourdry, J., and R. Guyetant. 1979. Group breeding effect on the iodine concentration in the coiioid of thyroid foiiicles of Alpes obst&ns tadpole, and evaiuation of the iodine content of the thyroid coiioid. C.R. Hebd. SeamsAcad. Sci. Paris 289D:323-326. Hourdry, J., L. Herrnite, and R. Ferrand. 1996. Changes in the digestive tract and feeding behavior of anuran amphibians during metamorphosis. Physwl. Zwl. 69:219-25 1. Hower, R. 0 . 1967. The freeze-dry presemation of biological specimens. Proc. United StatesNatl.Mus. 119: 1-24. Hoyer, M. H. 1905a. Untersuchungen uber das Lymphgefassystem der Froschiarven. I. Teil. Bull. Acad. C r m . 1905:417-430. . 1905b. Untersuchungen uber das Lymphgefassystern 1 der Froschiarven. 1 .Teil. Bull.Acad. C r a m 1905:451-461. Hoyer, W. J., J. N. Shafer, J. E. Mauldin, and H. T. Corbett. 1971. Discriminated avoidance and escape conditioning with the tadpole (Ranapipiens).Psychon. Sci 24:247-248. Hrbek, J. 1950. On the flight reaction of tadpoles of the common toad caused by chernicai substances. Expenbtia 6: 100-101.

LITERATURE CITED E b q R B. 1980. Sprint velocity of tadpoles (Bufo boreas) through rnetamorphosis. Copeia 1980:537-540. R B., and R. D. Stevenson. 1979. Integrating therrnai ph?xiology and ecology of ectotherms: A discussion of appnraches. Amen'cun Zool. 19:357-366. E+-. R B., and T. P. Webster. 1976. Thermai biology ofAnolis Inards in a cornplex fauna: The cvistatellus group on h e r t o Rioo. ECO& 57:985-994. Ebgks, k 1957. The developrnent of the primary sensory systini inXlpnopus Lami (Daudin).J. Anat. 91 :323-338. . 1959. Studies in embryonic and larvai developrnent in -3mphibia. I. The embryology of Elezttherodactylus ricordii dspeciai referente to the spinai c0rd.J. Embryol.Exp. MorH. 722-38. . 1961. Ceil degeneration in the larvai ventral horn of Xmopshmi (Daudin).J. Embyol. Exp. Morphol. 9:269-284. . 1962. An experimental study of the relationships bem-een h b and spinai cord of the ernbryo of Ehthevwdactyhrr martiniansis.J.Embryol. Exp. Morphol. 10:575-601. . 1966. The thyroid and the developrnent of the nervous -em in Eleutherodactylus martinicensis: An experimental &.J. Embyol. Exp.Morphol. 16:401-430. . 1976. Metamorphic changes in the brain and spinai d pp. 856-863. In Fmg neurobwlogy, a handbook, edited . R Llins and W. Precht. Berlin: Springer-Verlag. iingk, A., and M. C. Prestige. 1967. Developrnent of behaviour in the hindlimb ofXenopus 1mi.J. 152:347-359. Zool. Ebgks, A., and P. Reier. 1972. A prelirninary study of the cdeirs of bovine prolactin on embryos of Ekutherodactylus d i i . Gen. Comp. E n d o h l . 19:304-312. Ehkebus, J., and E. S. Farrar. 1985. Insulin-like irnrnunoreacmity in serum and pancreas of metamorphosing tadpoles. Grn. Comp. Endoc~z'ml. 58:114-119. Hndey, M. B. 1958. The role of the thyroid in kidney development in the tadpole. W h 22:125-166. H.nxhke, E. 1826. Uber die Urnbildung des Darmkanais und der Kiernen der Froschkaulquappen. Isis von Oken: 613-627. Y ' on, T. 1796. The natural history of theflofifih of Surinum. 8 p., 4 plates. York: G. Peacock. Hinchison, V. H., and R K. Dupr. 1992. Therrnoregulation. pp. 206-249. In Environmental physi4@y of the amphibians, cdited by M. E. Feder and W. W. Burggren. Chicago: University of Chicago Press. Hmchison, V. H., and C. G. Hill. 1978. Therrnai selection of bullfrog tadpoles (Rana catesbeiana) at different stages of developrnent and acclirnation temperatura. J. Them. Bwl. 357-60. Hutchison, V . H., and J. D. Maness. 1979. The role of behavior i ternperature acclimation and tolerante in ectotherrns. n h m i c a n Zool. 19:367-384. H d e y , J. 1942. Evolution: The m d e m syzthesis. New York: Harper and Brothers. H d e y , T H. 1858. On the theory of the vertebrate s k d . Proc. Roy. Soc.Lonhn 9:381457. I d o , J. M. 1997. Morphogenesis of vertebrate hearts. pp. 114-1 26. In Deve@ment of cardiovascular systems.Molecules to qanisms, edited by M. E. Feder and W. W. Burggren. Cambridge: Cambridge University Press. &ya-Oka, A., and A. Shimozawa. 1987. Developrnent of the comective tissue in the digestive tract of the larvai and metamorphosingXenopus W s .Anat. Anz. 164:8 1-93. Ideker, J. 1976. Tadpole therrnoregulatory behavior facilitates grackle predation. TmasJ. Sci. 27:244-245.

379

w.

'

Ijiri, K. L., and N. Egami. 1975. Mitotic activity of gerrn c e h during normal developrnent of Xenopus l m i tadpoles. J. Embyol. Exp. M M w p 34:687-694. Ikenishi, K. 1980. Corticai granules persisting in gerrn ceils of Xenopus W s after the cleavage stages. Deve@. Growth Dzffm 22:669-678. . 1982. A possibility of an in vitro differentiationof primordiai germ ceils (PGCs) frorn blastorneres containing "germinal plasrn" of early cleavage stage inXenopw hmi. Deve@: GrowthDzffeel: 24:205-2 15. Ikenishi, K., and M. Kotani. 1975. Ultrastnicture of the "gerrninai plasrn" in Xenopus ernbryos after cleavage. Deve@. Growth Dzffeel: 17:101-1 10. Ikenishi, K., M. Kotani, and K. Tanabe. 1974. Ultrastnicturai changes associated with UV irradiation in the "genninai plasrn" ofxenopus hmi. Deve@. BBi. 36:155-168. Infantino, R. L., Jr. 1989. Ontogeny of gill and lung ventilatory responses to oxygen and carbon dioxide in the bullfrog, Rana cutesbeiana. American Zool. 29:57. (Abstr.) Inger, R. F. 1954. Systernatics and zoogeography of Philippine Amphibia. Fieidiana Zool. 34:389-424. . 1956. Some amphibians frorn the lowlands of North Borneo. Fieidiana Zool. 34:389424. . 1966. The systernatics and zoogeography of the Amphibia of Borneo. Fieldiana Zool. 52:1-402. . 1967. The developrnent of a phylogeny of frogs. Evolution 21:369-384. . 1969. Organization of communities of frogs aiong srnd rainforest streams in Sarawak.J. Anim. Ecol. 38:123148. . 1983. Larvae of southeast Asian species of LeptobraEhium and Leptobracbella (Anura: Pelobatidae). pp. 13-32. In Advances in herpetology and evolutwnary bwlogy, edited by A. Rhodin and K. Miyata. Cambridge, MA: Mus. Cornp. Zool. . 1985. Tadpoles of the forested regions of Borneo. Fieldiana Zool. (N.S.) (26):1-89. . 1986. Diets of tadpoles living in a Bornean rain forest. Alytes 5:153-164. . 1992a. Variation of apornorphic characters in streamdweiling tadpoles of the bufonid genus AnsonZa (Amphibia, Anura). Zoo1.J.Linnean Soc. 105:225-237. . 1992b. A bimodal feeding systern i a stream-dwelling larva of Rha~ophmus frorn Borneo. Copeia 1992:887-890. . 1996. Commentary on a proposed classiication of the farnily Ranidae. HeipetolofiiEa 52:241-246. Inger, R. F., and K. J. Frogner. 1979. New species of narrowrnouth frogs (genus Micyohyla) frorn Borneo. Sarawak Mus. J. 27:311-322. Inger, R. F., and P. A. Gritis. 1983. Variation in Bornean frogs of the A m @ s jerboa species group, with descriptions of two new species. Fieldiana Zool. (N.S.) (19):1-13. Inger, R. F., and F. L. Tan. 1990. Recently discovered and newly assigned frog larvae (Ranidae and Rhacophoridae) frorn Borneo. Ra@s Bull. Zool. 38:3-9. Inger, R. F., and H . K. Voris. 1988. Taxonornic status and reproductive biology of Bornean tadpole-carrying frogs. Copeia 1988:1060-1061. Inger, R. F., H. K. Voris, and K. J. Frogner. 1986a. Organization of a cornrnunity of tadpoles in rain forest streams in B0rneo.J. TV. Ecol. 2:193-205. Inger, R. F., H. K. Voris, and P. Walker. 1986b.Larva1 transport in a Bornean ranid frog. Copeia 1986:523-525.

380

LITERATURE CITED

Inger, R. F., and R. J. Wassersug. 1990. A centrolenid-like anuran larvae from southeast Asia. Zool. Sci. 7:557-561. Ingerman, R. L. 1992. Maternal-fetal oxygen transfer in lower vertebrates.American Zool. 32:322-330. Ingle, D. 1976. Behavioral correlates of central visual funaion in anurans. pp. 435-451. In Frog neurobwlagy, a handbook, edited by R. Llins and W. Precht. Berlin: Springer-Verlag. Ingram, G. J., M. Anstis, and C. J. Corben. 1975. Observations on the Australian leptodactylid frog, Asa darlingtoni. Hwpetolagku 31 :425429. Ingram, V N . 1972. Embryonic red blood cell formation. Nature 3351339. Inokuchi, T., K. I. Kobayashi, and S. Horiuchi. 1991. Acid proteinases of the fore-gut in metamorphosing tadpoles ofRana catesbeiana. Comp.Bwchem. Physwl. 99B:653-662. Ishizuya-Oka, A,, and A. Shimozawa. 1990. Changes in lectinbinding pattern in the digestive tract of Xenopw h&s during metamorphosis. 11. S m d intestine.J. Morphol. 205:9-15. Ison, R. E. 1968. The development of the heart of Rana tempuraria L.J. Nat. Hist. 2 9 9 4 5 7 . Istock, C. A. 1967. The evolution of complex life cycle phenomena: An ecological perspective. Evolution 21 :592-605. . 1984. Boundaries to life histoty variation and evolution. pp. 143-168. InA new ewlagy: Nmel appvomhes to intwactive qtems, edited by P. W. Price, C. N. Slobodchikoff, and W S. Gaud. New York: Wiley. Iwasawa, H., and M. Kobayashi. 1976. Development of the testis in the frog Rana nigrommulata, with special referente to germ ceil maturation. Copeia 1976:461-467. Iwasawa, H., and K. Yamaguchi. 1984. Ultrastructural study of gonadal development in Xenopus h&s. Zool. Sci. 1:591600. Izecksohn, E. 1997. Novo enero de Hylidae Brasileiro (Arnphibia, Anura). Rev. Uniu. Rural, S Cinc. Vida 18:47k 52. Izecksohn, E., and C. A. G. Da Cruz. 1972. Notas sobre os girinos de Dendrophryniscus leucomystax Izecksohn e D. brevipollicatus Espada (Amphibian, Anura, Bufonidae).Arq. Univ.Fed. Rural Rio deJaneiro 2:63-69. . 1976. Nova espcies de Phyilomedusinae do Estado do Esprito Santo, Brasil (Amphibia, Anura, Hylidae). h Bra. sil. Biol. 36:257-261. Izecksohn, E., C. A. G. Da Cruz, and 0. L. Peixoto. 1979. Notas sobre o girino de Proceratophrys boiei (Wied)(Amphibia, Anura, Leptodactylidae). Rev. Brasil. Bwl. 39:233-236. Izecksohn, E., and S. T. De Aibuquerque. 1972. Algumas observaes sobre o desenvolvimento de Eleuthwodactylus venancwi B. Lutz (Amphibia, Anura). Arq. Univ. Fed. Rural Rio deJaneiro 2:13-15. Izecksohn, E., J. Jim, S. T. De Aibuquerque, and W F. De Mendona. 1971. Observaes sobre o desenvolvimento e os hbitos de Myeniella szlbnigra (Miranda-Ribeiro).Arq. Mus. Nm. Rio deJaneiro 54:69-73. Jacobson, A. G., and J. T. Duncan. 1968. Heart induction in salamanders.J Exp. Zool. 167:79-103. Jacobson, C. M. 1968. The development of the chondrocranium in two species of the Australian anuran genus Pseuabphryne Fitzinger.AustralianJ.Zool. 16:1-15. Jacoby, J., and K. Rubinson. 1983. The acoustic and lateral line nuclei are distinct in the premetamorphic frog, Rana catesbeiuna.J. Comp.Neurol. 216:152-161. Jaeger, C. B., and D. E. H i h a n . 1976. ~Morphology gustaof tory organs. pp. 588-606. In Frg neurobwhgy, a handbook, edited by R. Llins and W. Precht. Berlin: Springer-Verlag.

Jaeger, R. G., and J. E Hailman. 1976. Ontogenetic shift of spectral phototaaic preferences in anuran tadpo1es.J. Comp. Physiol. Psychol. 90:930-945. Jaeger, R. G., and S. C. Wds. 1989. On salamander guilds and ecological methodology. Herpetolagica 45:11 1-1 19. Jaffee, O. C. 1954a. Morphogenesis of the pronephros of the leopard frog (Ranaptptens).J. q h o l . 95 :109-123. M . 1954b. Phenol red transport in the pronephros and mesonephros of the developing frog (Ranapipiens).J. Cell. Comp.Physwl. 44:347-364. . 1963. Cellular differentiation in the anuran pronephros. Anat. Rec. 145:179-18 1. Jagersten, G. 1972. Evobtwn of the metawan lye ycle: A comprehensive theory. London: Academic Press. Jande, S. S. 1966. Fine structure of lateral-line organs of frog tadpo1es.J. U l t r m c t . Res. 15 :496-509. Janes, R. G. 1934. Studies on the amphibian digestive system. I. Histological changes in the alimentary tract of anuran larvae during invo1ution.J. Exp. Zool. 67:73-91. Jasienski, M . 1988. Kinship ecology of competition: Size hierarchies in kin and nonkin laboratoty cohorts of tadpoles. Oecolagia 77:407413. Jaslow, A. P., T. E. Hetherington, and R. E. Lombard. 1988. Structure and function of the amphibian middle ear. pp. 6991. In The evolutwn of the amphibian auditmy ystem, edited by B. Fritzsch, M. J. Ryan, W. Wdczynski, T. E. Hetherington, and W. Walicowiak. New York: Wiley. Jayatilaka,A. D. P. 1978. An ultrastructural study of the thyroid gland in the pre-metamorphic Xenopus laevik (Daudin) tadpo1e.J. A m t . 125:579-591. Jennings, D. H. 1994. Thyroid hormone mediation of embryonic development in a non-metamorphosingfrog, Eleutherodactylus coqui.J.M q h o l . 220:250. Jennings, D. H., and J. Hanken. 1998. Mechanistic basis of life history evolution in anuran amphibians: Thyroid gland development in the direct-developingfrog, Ehthemda~tylus coqui. Gen. Comp.Enhcvinol. 111:225-32. Jennings, R. D., and N.J. Scott, Jr. 1993. Ecologicdy correlated rnorphological variation in tadpoles of the leopard frog, Rana chiricahuensis.J. Herpetol. 27~285-293. Jenssen, T. A. 1967. Food habits of the green frog, Rana clamitans, before and during metamorphosis. Copeia 1967:214218. Jia, X. X., and W. W. Burggren. 1997a. Developmental changes in chemoreceptive control of giil ventilation in larval bullfrogs (Rana catesbeiana). I. Reflex ventilatoty responses to ambient hyperoxia, hypoxia, and NaCn. J. Exp. Bwl. 200:2229-2236. . 199%. Developmental changes in chemoreceptivecontrol of gill ventilation in larval bullfrogs (Rana catesbeiana). 1 . Site of O,-sensitive chemoreceptors. J. Exp. Bwl. 1 200:2237-2248. Jiang, T., and A. Hoiley. 1992. Morphological variations among output neurons of the olfactoty bulb in the frog (Ranaridibunda).J.Comp.Neuol. 32036-96. Johansen, K., and C. Lenfant. 1972. A comparative approach to the adaptability of O,-HB aanity. pp. 750-783. In Oygen a$nity of hemoglobin and red cell acid bme status.Alfred Benwn Symp., edited by M. Rorth and P. Astrup. New York: Academic Press. John, K. R., and D. Fenster. 1975. The effects of partitions on the growth rates of crowded Ranapipiens larvae. American Midl. Nat. 93:123-130. John, K. R., and J. M. Fusaro. 1981. Growth and metamorpho-

LITERATURE CITED

381

sis of solitary Ranapipiens tadpoles in conined space. Copeia 1981:737-741. John-Alder, H. B., and P. J. Morin. 1990. Effects of larvai density on jumping ability and stamina in newly metamorphosed Bufo woodhousiifowlen'. Copeia 1990:856-860. Johnson, D. S. 1967. On the chemistry of freshwaters in southern Malaya and Singapore. A&. Hydrobwl. 63:477-496. Johnson, L. M. 1991. Growth and development of larvai northern cricket frogs (Akscrepitans)in relation to phytoplankton abundance. Fresh. Bwl. 25:51-59. Johnston, G. F. 1982. Functions of the keratinized oral features in anuran tadpoles with an analysis of ecomorphological tadpole types. Thesis, Mississippi State University. . 1990. Feeding function and anatomy of the oral apparatus of anuran larvae. Dissertation, Southern Iliinois University. Johnston, G. F., and R. Altig. 1986. Identification characteristics of anuran tadpoles. Herpetol. Rev. 17:36-37. Johnston, L. M. 1991. Growth and developrnent of larval northern cricket frogs ( A hcrepitans) in relation to phytoplankton abundance. Fresh. Bwl. 25:51-59. Jolivet-Jaudet, G., and J. Leloup-Hatey. 1984. Variations in aldosterone and corticosterone plasma levels during metamorphosis in Xenopus l m i tadpoles. Gen. Comp. Endomiml. 56:59-65. Jollie, M. 1982. Ventral branchial musculature and synapomorphies questioned. Zoo1.J. Linnean Soc. 74:3 5 4 7 . Jonas, L., and P Rohlich. 1970. Elektronenmikroskopischer . Nachweis saurer Mucopolysaccharidein den Flaschenzellen der Xenopus-Niere. Zeit. Zellforsch.Mikros. Anat. 104:56-68. Jonas, L., and L. Spannhof. 1971. Elektronenmikroskopischer Nachweis von Mucoproteiden in den Flaschenzellen der Urniere von Xenopus h u i s Daudin mittels Phosphorwolframsaurefkbung. Acta Himchem. 41:185-192. Jones, R. E., A. M. Gerrard, and J. J. Roth. 1973. Estrogen and brood pouch formation in the marsupial frog Gwotheca rMbamba.J. Exp. 2001. 177-184. 184: Jones, R. M. 1980. Nitrogen excretion by Scaphwpus tadpoles in ephemeral ponds. Physiol. 2001. 53:26-3 1. Jordan, H. E. 1933. The evolution of blood-forming tissues. Quart. Rev.Bwl. 8:58-76. Jmgensen, C. B. 1973. Mechanisms regulating ovarian function in amphibians (toads). pp. 133-15 1. In The akveiupment and maturatwn of the ovary and i t s f u h n s , edited by H. Peters. Amsterdam: Excerpta Medica. . 1986. Effects of fat body excision in female Bufo bufo on the ipsilateral ovary, with a discussion of fat body-gonad relationships. Acta 2001. 5-10. 67: Jmgensen, C. B., L. O. Larsen, and B. Lofts. 1979. Annual cycles of fat bodies and gonads in the toad Bufo bufo (L.), compared with cycles in other temperate zone anurans. Biol. Skr. (Copenhagen) 22:1-27. Jorquera, B., O. Garrido, and E. Pugn. 1982. Comparative studies of the digestive tract development between Rbinodema dunvinii and R. ntfum. J. Herpetol. 16:204-214. Jorquera, B., and L. Izquierdo. 1964. Tabla de desarrollo normal de Calyptocephaiellagayi (Rana chilena).BwlogiEa 36:4353. Jorquera, B., E. Pugn, and 0.Y Goicoechea. 1972. Tabla de . desarrollo normal de Rbinodanza danvi'nii. Arch. Med. V i . 4:5-19. Joseph, B. S., and D. G. Whitlock. 1968. Central projections of selected spinal dorsal roots in anuran amphibians. Anat. Rec. 160:279-288.

Junc, F. A., R. Altig, and C. Gascon. 1994. Breeding biology of Colostethus stepheni, a dendrobatid frog with a nontransported nidicolous tadpole. Copeia 1994:747-750. Jungfer, K.-H. 1985. Beitrag zur Kenntnis von Dendvabates speciosus 0. Schmidt, 1857 (Saiientia: Dendrobatidae). Salamanha 21 :263-280. . 1996. Reproduction and parental care of the coronated treefrog, Anotheca coronata (Steindachner 1864)(Anura: Hylidae). Herpetologca 52:25-32. Jungfer, K.-H., and L. C. Schiesari. 1995. Description of a central Amazonian and Guianan tree frog, genus Osteocephalus (Anura, Hylidae), with oophagous tadpoles. Alpes 13:l-13. Junqueira, L. C. U., P. P Joazeiro, S. M. O. Vieira, A. B. Costa . Silva, A. Cais, and G. S. Montes. 1984. S p e d c attachment of morphologicaily-distinct intermediate filaments to desmosomes and hernidesmosomes in the epidermis of the tadpole of the anuran, Pseudis paradoxus. J. Submicros. Cpol. 16:643-648. Jurgens, J. D. 1971. The morphology of the nasal region of Amphibia and its bearing on the phylogeny of the group. Ann. Uniu. Stelienbosch 46A:1-146. Just, J. J., and B. G. Atkinson. 1972. Hemoglobin transitions in the bullfrog, Rana catesbeiana, during spontaneous and induced metamorphosis.J. Exp. 2001. 182:271-280. Just, J. J., R. N. Gatz, and E. C. Crawford. 1973. Changes in respiratory functions during metamorphosis of the bullfrog, Rana catesbeiana.Respir.Physwl. 17:276-282. Just, J. J., R. Sperka, and S. Strange. 1977. A quantitative analysis of plasma osmotic pressure during metamorphosis of the bullfrog, Rana catesbeiana.Expehntia 33 :1503-1504. Justis, C. S., and D. H. Taylor. 1976. Extraocular photoreception and compass orientation in larval bullfrogs, Rana catesbeianu. Copeia 1976:98-105. Kaiser, H., and R. Altig. 1994. The atypical tadpole of the dendrobatid frog, Colostethus chalcopis, from Martinique, French Antilles. J. Herpetol. 28:374-378. Kait, M. R. 1973a. Uitrastructurd obse~ations the germ on line of Xenopus h u i s . Zeit. Zellforsch. Mibos. Anat. 138:4162. . 1973b. Morphology and kinetics of spermatogenesis in Xenopus h u i s . J . Exp. 2001. 195:393408. Kalt, M. R., and J. G. Gaii. 1974. Observations on early germ cell development and premeiotic ribosomal DNA amplification in Xenopus 1mi.J. Bwl. 62:460-472. Cell Kam, Y.-C., Z.-C. Chuang, and C.-F. Yen. 1996. Reproduction, oviposition-siteselection, and tadpole ecology of an arboreal nester, ChirUcalus eifingeri (Rhacophoridae), from Taiwan. J. Herpetol. 30:52-59. Kamat, N. D. 1962. On the intestinal content of tadpoles and algae of smaii ponds. Cum Sci. 321:300-310. Karnimura, M., K. Ikenishi, M. Kotani, and T Matsuno. 1976. Observations on the migration and proliferation of gonocytes in Xenopus h u i s . J. Embryol. Exp. Murphol. 36:197-207. Kamimura, M., M. Kotani, and K. Yamagata. 1980. The migration of presumptive primordial germ cells through the endodermal cell mass inXnopush i s . A light and electron microscopic study.J. Embryol. Exp. Morphol. 59:1-17. Kampmeier, O. F. 1915. On the origin of lymphatics in Bufo. AmericanJ. Anat. 17:161-209. . 1922. The development of the anterior lymphatics and lymph hearts in anuran embryos. AmekanJ. Anat. 30:61131. . 1958. On the lymphatics system ofhaphus: Its evolutionary significance.Anat. Rec. 132:343-364.

382

LITERATURE CITED estados larvales de Bufo arenarum en condiciones naturales. Neotrpica 36:67-81. Kehr, A. I., and N. G. Basso. 1990. Description of the tadpole of Lysapsus Limellus (Anura, Pseudidae) and some considerations on its biology. Copeia 1990:573-575. Kehr, A. I., and J. A. Schnack. 1991. Predator-preyrelationship between giant water bugs (Beiostomaoxyztrum) and larval anurans (Bufoarenarum).Alpes 9:61-69. Kehr, A. I., and J. D. Wdiams. 1990. Larvas de anuros de Ia Republica Argentina. Cuad. Herpetol.Monoj1: (2):1-44. Keiffer, H. 1888. Recherches sur la structure et Ie dveloppement des dents et du bec coms chez Alytes obstetn'mns.A&. Bwl. 9:55-81. Keiier, R. 1991. Early embryonic development ofXenUpusk s . pp. 62-1 13. In Xempw k s : Practical uses in ull and molecular bwlojy.Metb. CellBwl. Vol. 36, edited by B. K. Kay and H. B. Peng. New York: Academic Press. Kemali, M. 1974. Ultrastructural asymmetry of the habenular nuclei of the frog. J. Hirnforsch. 15:419426. Kemali, M., and V. Braitenburg. 1969. A t h of thefroj's brain. Berlin: Springer-Verlag. Kemali, M., and V. Gughelmotti. 1977. An electron microscope observation of the right and the two left portions of the habenular nuclei of the fr0g.J. Cmp. Neurol. 176:133-148. . 1987. A horseradish peroxidase study of the olfaaory system of the frog, Rana escuhnta. J. Comp. Neurol. 263:400417. Kemp, A. 1977. The pattem of tooth plate formation in the Australian lungfish, Neoceratodusforstmm (Krefft). Zoo1.J Lznnean Soc. Lona'on 60:223-258. Kemp, N. E. 1949. Development of intestinal coiling in the anuran tadpole. Anat. Rec. 105:88. (Abstr.) . 1951.Development of intestinal coiling in anuran larvae.J. Exp. Zool. 116:259-287. Kemp, N. E., and J. A. Hoyt. 1969. Sequence of ossification in the skeleton of growing and metamorphosing tadpoles of 2bznapipiens.J. Morphol. 129:415-444. Kenny, J. S. 1969a. The Amphibia of Trinidad. Stud. Fauna Curapo Caribbean Isl. 29: 1-78. . 1969b. Pharyngeal mucous secreting epithelia of anuran larvae. ActaZool. 50: 143-153. . 1969c. Feediig mechanisms in anuran 1arvae.J Zool. 157:225-246. Kenward, R. E. 1978. Hawks and doves: Factors affecting success and selection in goshawk attacks on woodpigeons. J.Anim. Ewl. 47:449-460. Kerr, J. B., and K. E. Dixon. 1974. An ultrastructural study of germ plasm in spermatogenesis of &nupus hn&. Embvyol. J. Exp. Morphol. 32: 573-592. Kerr, J. G. 1905. The embryology of certain of the lower fishes, and its bearing upon vertebrate morphology. Proc. Roy. Soc. Edinburgh 1905-1906, Vol. 16. . 1919. Tmbook of embryology. London: MacMiUan Publ. Co. Kessel, R. G., H. W. Beam, and C. Y. Shih. 1974. The origin, distribution and disappearanceof surface cilia during embryonic development of Rana pip'ens as revealed by scanning electron microscopy.A m k a n J .Anat. 141:34 1-360. Kesteven, H . L. 1944. The evolution of the s k d and the cephalic musdes: A comparative study of the development and adult morphology. Part 11. The Amphibia. Mem. Aurtralian Mus. Sydney 8:133-236. Kett, N. A., and E. D. Poiiack. 1985. Retention of lateral motor

. 1969. Evolutwn and cmparative morphology ofthe lymphaticsystem. Springfield, IL: Charles C. Thomas. Kanamadi, R. D., and S. K. Saidapur. 1988. Effea of fat body ablation on spermatogenesis in the Indian skipper frog Rana cyamph&ctis Schn. Zool. Anz. 220: 79-8 1. Kaplan, M. 1994. A new species of frog of the genus Hyh from the Cordillera Oriental in northern Colombia with comments on the taxonomy of Hyh minuta. J. Herpetol. 28: 79-87. Kaplan, R. H. 1992. Greater maternal investment can decrease offspring survival in the frog Bombina orientalis. Ewlogy 73:280-288. Karlson, P. 1972. Kunes Lehrbuch der Biochemie. Stuttgart: Georg Thieme-Verlag. Karlstrom, E. L. 1962. The toad genus Bufo in the Sierra Nevada of California: Ecological and systematic relationships. Univ. C a l f m i a Publ. Zool. 62: 1-104. Karnovsky, M. J. 1965. A forrnaldehyde-glutaraldehydefixative of high osmolality for use in electron microscopy. J. Cell. Bwl. 27A:137-138. Kasinathan, S., A. Gunashing, and S. L. Basu. 1978. Fat bodies and spermatogenesis in South Indian green frog &na hma+h Lesson. Boll. Zool. 45: 15-22. Kats, L. B., J. W. Petranka, and A. Sih. 1988. Antipredator defenses and the persistente of amphibian larvae with fishes. Ecology 69: 1865-1870. Katz, L. C., M. J. Potel, and R J. Wassersug. 1981. Structure and mechanisms of schooiing in tadpoles of the dawed frog, XenUpus hn&. Anim. Behav. 29:20-33. Katz, M. J. 1980. Allometry formula: A cellular model. Growth 44:89-96. Kaung, H. C. 1975. Development of the horny jaws of Rana pipiens larvae. Develop. Bwl. 11: 2 5 4 9 . . 1981. Immunocytochemical localization of pancreatic endocrine ceiis in frog embryos and young larvae. Gen. C m p . Endom'ml. 45 :204-2 11. . 1983. Changes of pancreatic beta cell population during larval development in Ranapipiens. Gen. Comp.Endocrinol. 49:50-56. Kaung, H. C., and J. J. Kollros. 1976. CeU turnover in the beak of Ranapipiens.Anat. Rec. 188:361-370. Kawada, J., R. E. Taylor, Jc, and S. B. Barker. 1969. Measurement of Na-K-ATPase in the separated epidermis of Rana catesbeianu frogs and tadpoles. Comp. Bwchem. Physwl. 30:965-975. Kawai, A., J. Ibeya, and K. Yoshizato. 1994. A three-step mechanism of action of thyroid hormone and mesenchyme in metamorphic changes in anuran larval skin. Develop. Bwl. 166:477488. Kawamura, T., and M. Kobayashi. 1960. Studies on hybridization i amphibians. VII. Hybrids benveen Japanese and Eun ropean brown frogs.J. Sci. Hiroshima Univ. 18B:221-238. Keating, M. J., and R. M. Gaze. 1970. The ipsilateral retinotectal pathway in the frog. Quart.J.Exp. Physiol. 55:284292. Kehr, A. I. 1989. Factores dependientes de Ia densidad y influencia sobre e1 crecimiento individual de 10s estados larvales de Hyh pukhelh p u e (Amphibia Anura). Limno~ W 2:757-761. S . 1997. Stage-frequencyand habitat seleaion of a cohort of Pseuhmi ocuhris tadpoles (Hyiidae: Anura) in a Florida temporary pond. Herpetol.J. 7: 103-109. Kehr, A. I., and E. O. Adema. 1990. Crecimento corporal y analises estadistico de Ia frequencia por clases de edades de 10s

LITERATURE CITED column neurons during the phase of rapid ceii loss after limb amputation in Ranapipns tadpo1es.J Ekp. Zool. 236: 59-66. E;han, M. S. 1982a. Key to the identification of amphibian tadpoles from the plains of Pakistan. PakZrcan J. Zool. 14:133-145. . 1982b. Coilection, preservation, and identification of amphibian eggs frorn the plains of Pakistan. PakistanJ.Zool. 14:241-243. b, S., and S. A. Malik. 1987. Buccopharyngeal rnorM. phology of the tadpole ofRanahazarensis Dubois and Khan 1978, and its tomenticole adaptations. Biologia (Lahme) 33:45-60. Khan, M. S., and S. A. Mufti. 1994. Oral disc rnorphology of amphibian tadpole and its functional correlates. Pakistan J. Zool. 26~25-30. Kdter, E. 1979. Some telencephalic comections in the frog, Ranupipns. J Comp.Neurol. 185:75-86. Kdter, E., and S. O. E. Ebbesson. 1976. Organization of the 'non-olfactory" telecephalon. pp. 946-972. In Frog neurobi& I , a handbook, edited by R. &s and W. Precht. Berlin: Springer-Verlag. &&ter, E., and R. G. Northcutt. 1975. Ascending derents to the telencephalon of ranid frogs: An anterograde degeneration study.J. C m p . Neurol. 161:239-254. k c k e r , J. 1996. pH-mediated predator-prey interactions bemeen Ambystoma tgrinum and PseudL1E?.striimata. Ecol. 4 1 . 6: 1325-1331. Kiiham, P. 1982.The biogeochernistry of bog ecosystems and the chemical ecology of S p h a p m . MiEhigan Bot. 21: 159-168. Edahi, M. 1938. Zur Entwicklung der Exkretionsorgane von Dipnoern und Amphibien rnit Anmerkungen bezuglich Ganoiden und Teleostier.Acta Zool. 19:1-190. lkdred, J. E. 1927. Ceil division and ciliogenesis in the ciliated epithelium of the pharynx and esophagus of the tadpole of &e green frog, Rana clamitans.J. Morphol. 43:267-297. i&g, M. 1981. A cytotaxonomic analysis of Australian hylid frogs of the genus Litoria. Proc. Hepetol. Symp., Melbourne 169-1.75. -. W b u r y , B. F. 1895. The lateral line system of sense organs in some American Amphibia, and cornparison with the dipmans. Trans.A m ' m n MiFIos. SOL. 17:115-154. b e y , E. C. 1964. Extent of acid mine poilution in the United States decting fish and wildlife. United States Fkh Wildl. Cin. 191. 27 p. Kinisinghe, P. 1957. Amphibia of Ceyon. Published privately, Colornbo. . 1958. Some hitherto undescribed anuran tadpoles. CeyhJ. 1:171-176 Sci. Eseleva, E. I. 1984. Responses of behavioral reactions in the Rama tempmaria tadpoles to chemical stimuii under the conditions of pretreatrnent. 2 1 0 Zhur 63: 1046-1054. . . 1992. h i n o acid sensitivitv in tad~oies the s ~ a d e of foot toad Peobatesfuscus. J. ~voL.'~iocheh. Pbysiol. 28r527330. . 1997. Aspects of the chemicai ecology and chemosensoy guided behavioral reactions in anuran tadpoles. Adv. h p h i b . Res. Fom. Soviet Unwn 2:95-101. &eleva, E. I., and Y. B. Manteifel. 1982. Behavioral reactions of Bujb bufo and Rana tempmaria tadpoles to chemical stimuli. Zool. Zhur 61 :1669-1681. Kkm, S. L., and l? l?C. Graziadei. 1983. The differentiation of the olfactory placode in Xenupus L : & A light and electron microscope study. J. Comp.Neuml. 217: 17-30.
-~ -

383

Kiuge, A. G. 1988. The characterization of ontogeny. pp. 5781. In Ontogeny and ystematics, edited by C. J. Humphries. New York: Columbia University Press. . 1989. A concern for evidence and a phylogenetic hypothesis of relationships among Epzirates (Boidae, Serpentes). Syst. Zool. 38: 7-25. Kiuge, A. G., and J. S. Farris. 1969. Quantitative phyletics and the evolution of anurans. Syst. Zml. 18:1-32. Kiuge, A. G., and R. E. Strauss. 1985. Ontogeny and systernatics. Ann. Rev. Ecol. Syst. 16:247-268. Kiuger, M. J. 1978. The evolution and adaptive value of fever. American Sci. 66: 38-43. Kiymkowsh M. W., and J. Hanken. 1991. Whole-rnount staining of Xenupus and other vertebrates. Meth. Cell Bwl. 36:419-441. Kobayashi, M. 1975. Sexual and bilateral d8erences in germ ceil numbers in the course of the gonadal developrnent of Rana n&omb~uhta. Zool. Mag. (Tokyo) 84: 109-1 14. Kok, D., L. H. Du Preez, and A. Channing. 1989. Channel construction by the African bullfiog: Another anuran parenta1 care strategy.J. Hepetol. 23:435437. Koilros, J. J. 1961. Mechanisrns of amphibian rnetamorphosis: Hormones. American Zool. 1:107-114. . 1981. Transitions in the nervous systern during amphibian metamorphosis. pp. 445459. In Metamo?.phosk:A problem in dmehpmental bwlogy. 2d ed., edited by L. I. Gilbert and E. Frieden. New York: Plenum. . 1984. Growth and death of ceils of the rnesencephalic fifth nucleus in Rana pipiens larvae. J. Comp. Neurol. 224:386-394. Koiiros, J. J., and i? McMurray. 1956. The rnesencephalic V M. nucleus in anurans. 11.The iduence of thyroid hormone on cell size and ceil number.J. Exp. Zool. 131:1-26. Koiiros, J. J., and M. L. Thiesse. 1985. Growth and death of ceils of the mesence~halic Mih nucleus in Xe-s hmis larvae.J. C m p . Neurol. 233:481489. Kopsch, F. 1952. Die Enhvicklung des braunen Grasf?.osches Rana fusca Roesel (DargereUt in der Art der Nwmentafeln zur Enhvicklun.eschiEhte der Wirbeltiere). Stuttgart: Georg Thieme-Verlag. Korky, J. K., and R. G. Webb. 1973. The larva of the Mexican toad Bufo caazfroons W1egrnann.J Hepetol. 7:4749. . 1991. Geographic variation in larvae of rnountain treefrogs of the Hyh mimia group (Anura: Hylidae). Bull. New J e q A c a d . Sci. 36:7-12. . 1994. A scanning electron microscope study of the larval labial teeth of Rama montezum. Bull.Maryland Hepetol. Soc. 30: 126-131. . 1996. Morphological variation in larvae of Rana tempmaria L. (Anura: Ranidae) in the Republic of Ireland. I& Biqgeograph. Soc. Bull. (19):159-181. Koskela, P. 1973. Duration of the Iamal stage, growth, and migration in Rana temporaria L. in two ponds in northern Finland in relation to environrnental factors. Ann. Zool. Fennici 10:414-418. Kothbauer, H., and H. Schenkel-Brunner. 1981. Immunochemicai investigations on fish eggs and toad spawn: On the appearance of glycosidases during the hatching process. Zool. Anz. 206:354-360. Kotthaus, A. 1933. Die Entwicklung des Primordial-cranium von Xenupus hmis bis zue Metarnorphose. Zeit. Wiss. Zool. 144:510-572. Kraemer, M. 1974. La morphogense du chondrocrne de DkA -

384

LITERATURE CITED (Amphibia, Leptodactylidae) en e1 Paran medio, Argentina. Stud. Neotmp. Fauna Environ. 29: 55-61. Lajmanovich, R. C., and J. Faivovich. 1998. Dieta larval de Phyllomedusa tetraploidea Pombal and Haddad, 1992 en Ia provincia de Misiones (Argentina).Alytes 15:137-144. Lamar, W. W., and E. R. Wiid. 1995. Cornments on the natural history of Lithodytes lineatus (Anura: Leptodactylidae), with a description of the tadpole. Herpetol. Nat. Hist. 3: 135142. La Marca, E. 1985. Systematics and ecological observations on the nebtropical frogs Hyhjahni and ~~Lp1aqdactyh.J. Her~etol. 19:227-237. L a ~ a r c aE., and A. Mijares-U. 1988. Description of the tad, pole of Colostethus mayorgai (Anura: Dendrobatidae) with preliminary data on the reproductive biology of the species. Buli. Malyhnd Herpetol. Soc. 24:47-57. Lamb, A. H. 1981. Target dependency of developing motoneurons in X e q u s 1 m j . J . Comp.Neurol. 203: 157-1 71. Lamb, A. H., M. J. Ferns, and K. Klose. 1989. Peripheral competition in the control of sensory neuron numbers inXen~pw frogs reared with a single biiateraily innervated hindlimb. Develop. Brain Hes. 45: 149-154. Lamberti, G. A., S. V. Gregory, C. P. Hawkins, R. C. Wiidman, L. R. Ashkenas, and D. M. Denicola. 1992. Plant-herbivore interactions in streams near Mt. St. Helens. Freshw. Bwl. 27:237-247. Lambertini, G. 1929. I1 manicotto glandulare di Rana esculenta. Ric. Mmfol. Roma 9: 71. Lambiris, A. J. L. 1988a. A review of the amphibians of Natal. L a m ~ g e y e r :1-2 10. (39) . 1988b. Frogs and to& of the Natal Drakensberg. Ukhahlamba Series no. 3. Pietermaritzburg: University of Natal Press. . 1989. The frogs of Zimbabwe. Mus. RegMn. Sci. Nat. TorinoMonog~(10):1-246. Larnborghini, J. E. 1980. Rohon-Beard cells and other large neurons in X e q u s embryos originate during gastrulation. J. Comp.Neurol. 189:323-333. . 1987. Disappearance of Rohon-Beard neurons from the spinal cord of larval Xenopus lmis. J. Comp. Neurol. 264:47-55. Lametschwandtner, A., U. Albrecht, and H. Adam. 1978. The vascularization of the kidneys in Bufo bufo (L.), Bombina varigata (L.), Rana ridibunda (L.) and Xenopus h i s (D.) (Amphibia, Anura) as revealed by scanning electron microscopy of vascular corrosion casts.Acta Zool. 59: 11-23. Lametschwandtner, A., A. Laminger, and W. Slattenschek. 1982. Eine experirnentelle Anordunung zur Erfassung des Crowd-Effektes bie Kaulquappen des Suedafrikanischen Krallenfrosches X e q u e laepis (Daudin). Zool. Anz. 209: 145-155. Lamotte, M. 1961. Contribution l'tude des Batraciens de I'Ouest africain. XII. Les formes larvaires de Cardwglossa leucomystax Blgr. Bull. Inst. France Afizque Noire 23 :211-216. . 1963. Contribution l'tude des Batraciens de l'Ouest africain. XVII. Le dveloppement larvaire de Hymenochirus (Pseudhymenochivus) merlini Chabanaud. Bull. Inst. France Afique Noire 25:944953. . 1964. Contribution I'tude des Batraciens de I'Ouest africain. XVIII. Le dveloppement larvaire de Phlynomantis mzcrgs Peters. Bull. Inst. FranceAfiqueNoire 26:228-237. Lamotte, M., and S. Dzieduszycka. 1958. Contribution l'tude des Batraciens de I'Ouest africain. VII. Le dveloppe-

coglosus pictus Otth. (Amphibien, Anoure). Bull. Bwl. 108:211-228. Kramer, D. L. 1987. Dissolved oxygen and fish behavior. Environ. Bwl. Fish. 18:81-92. Kramer, D. L., and J. B. Graham. 1976. Synchronous air breathing, a social component of respiration in fishes. Copeia 1976:689-697. Kramer, H., and G. M. Windrum. 1955. The metachromatic staining reaai0n.J. Histochem. Cytochem. 3 :227-237. Kratochwill, K. 1933. Zur Morphologie und Physiologie der Nahrungsaufnahme der Froschlarven. Zeit. Wiss. Zooi. 144:421468. Kress, A., and U. M. Spornitz. 1972. Ultrastructural studies of oogenesis in some European amphibians. I. Rana esculenta and Rana temporaria. Zeit. Zellforsch. Mikros. Anat. 128:438456. Krgel, P. 1993. Biologie und Okologe der Bromelienfauna von Guzmania weberbaureri im amazonischen P m e@nzt durch eine emfmsende. Biblwgraphie der Bromelien-Phytotelmata. Wien: Osterreichische Akad. Wissenshaft. Krgel, P., and S. Richter. 1995. Synccrpe antenori- a bromeliad breeding frog with free-swimming, non-feediig tadpoles (Anura, Microhylidae). Copeia 1995:955-963. Kmjitzer, E. M. 1931. De ontwikkeling van het chondrocranium en enkele kopzenuwen van Mgalophlys montana. Dissertation, Leiden. Kme, K. C., and M. G. Francis. 1977. A predation deterrent in larvae of the bullfrog, Rana catesbeiana. Trans. A m ' c a n Fish. Soc. 106:248-252. Kunst, J. 1936. Vergleichende Untersuchungen an Anurennieren insbesondere an der Niere von Xenopus laevis Daud. Zool.Jahrb. Abt. Anat. Ontog. Tiere 6 1:5 1-76. Kunte, K. 1998. Ahaetulh nasuta feeding on tadpoles. Hamadlyad 22: 124-125. Kuntz, A. 1924. Anatomical and physiological changes in the digestive system during metamorphosis in Ranapipiens and Amblystoma tigrinum.J.Morphol. 38: 58 1-598. KUnzle, H. 1983. Spinocerebellar projeaions in the turtle. Observations on their origin and terminal organization. Exp. BrainRes. 53:129-141. Kupferberg, S. J. 1997a. The role of l a ~ adiet in anuran metal morphosis. American Zool. 37: 146-159. . 1997b. Facilitation of periphyton produaion by tadpole grazing: Funaional differences between species. Freshw. Bwl. 37:427-439. . 1997c. Bullfrog (Rana catesbeiana) invasion of a California river: The role of larval competition. Ecology 78:1736-1751. . 1998. Predator mediated patch use by tadpoles (Hyla regilia):Risk balancing or consequences of motionlessness? J. Herpetol. 32: 84-92. Kupferberg, S. J., J. C. Marks, and M. E. Power. 1994. Effects of variation in natural algal and detrital diets on larval anuran (Hyh regilh) life-history traits. Cupeia 1994:446457. Kuramoto, M., and C.-S. Wang. 1987. A new rhacophorid treefrog from Taiwan, with comparisons to Chirixalus eifingeri (Anura, Rhacophoridae). Copeia 1987:931-942. Kuzrnin, S. L. 1995. The problem of food competition in amphibians. Herpeo1.J. 5:252-256. Laird, A. K., S. A. Tyler, and A. D. Barton. 1965. Dynarnics of normal growth. Grawth 29:233-248. Lajmanovich, R. C. 1994. Contribucin al conocimiento de Ia aiimentacin de larvas de la rana criolla Leptoda~$us ocelhtus

LITERATURE CITED

385

Bull. Inst. France ment larvaires de Phrynobatra~husfiancisci. AfiqueNoire 20: 1071-1086. Larnotte, M., S. Dzieduszydra, and G. Lauwarier. 1958. Contribution a l'etude des Batrauens de l'Ouest afncain. VIII. Les formes larvaires de l'tychadenasubmasuweniensis,Pt. tournien et Pt. tnnodis. BuU. Inst. France Afique Noire 20:14641482. Lamotte, M., R Gaon, and F. Xavier. 1973. Recherches sur le dveloppement embryonnaire de Nectophrynoides occdentalis Angel arnphibien anoure vivipare. 2. Le dveloppement des gonades. Ann. Embyol. Mo'phogen. 6:271-296. Lamotte, M., and J. Lescure. 1977. Tendances adaptives a l'afianchissement due miiieu aquatique chez les arnphibiens anoures. Tewe Vie 31:225-312. . 1989. Les ttards rhophiles et hygroptriques de I'Ancien et due Nouveau Monde. Ann. Sci. Nat., Zool., Paris 1O:lll-122,125-144. Larnotte, M., and J.-L. Perret. 1961a. Contribution I'tude des Batraciens de I'Ouest africain. XI. Les formes larvaires de trois espcies de l'tychadena:Pt. maccarthyensis And., Pt. perreti G. et L., et Pt. mascareniensis D. et B. Bull. Inst. France Afnsue Noire 23A: 192-210. . 1961b. Contribution I'tude de Batraciens de l'Ouest africain. XIII. Les formes larvaires de quelques especes de Leptopelis: L. aubryi, L. viridis, L. anchietae, L. oceliutus et L. calcaratus. Bull. Inst. France Afique Noire 23:855-885. . 1963a. Contribution l'tude de Batraciens de l'Ouest africain. XIi? Les dveloppement larvaire de Chiromantis mfescens Gnther. Bull. Inst. F ~ r a Afnsue Noire 25:266n 276. . 1963b. Contribution I'tude de Batraciens des I'Ouest africain. XV. Le dveloppement direct de l'espce Arthroieptis poecilonotus Peters. Bull. Inst. France Afique Noire 25:277284. . 1963c. Contribution l'tude de Batraciens de I'Ouest africain. XVI. Les formes larvaires de cnq espcies d'Hyperolius de Cameroun: H. hierogl'ycus, H. steindachm'pardulir, H. uutiro~tni,H. vi~dzfivus,H. tuberculatus. BuU. Inst. FmnaAfique Noire 25: 545-558. Larnotte, M., J.-L. Perret, and S. Dzieduszycka. 1959a. Contributions l'tude des Batraciens de I'Ouest africain. IX.Les formes larvaires de Petropedetes palmiper, Conrarra goliath et Acanthij,alus spinosus. Bull. Inst. France Afnsue Noire 2 1:762-776. . 1959b. Contributions I'tude des Batraciens de I'Ouest africain. X. Les formes larvaires de Cyptothykzxgresho$, Lqptopelis notatus et Ptychadenu tmnwcelk. Bull. Inst. FmeAfiqueNoire 21:1336-1350. Lamotte, M., and F. Xavier. 1972. Recherches sur le dveloppement embryonnaire de Nectophrynoiaes occidentalk Angel, amphibien anoure vivipare. I. Les principaux traits morphologiques et biometriques de dveloppement. Ann. Embryol. Morphol. 5:315-340. . 1973. Le dveloppement des gonades et Ia difFrentiation sexuelle chez I'embryon de Neccophrynoides occidentalis Angel. Symbwsis 4:259-261. Lamotte, M., and M. Zuber-Vogeli. 1954a. Contribution I'tude des Batraciens de I'Ouest africain. 11. Le dveloppement larvaire de Bufo rguaris Reuss, de Rana occz'dentalis Gnther et de Rana crassipes (Buch. et Peters). Bull. Inst. France A f i q w Noire 16:940-954. . 1954b. Contribution I'tude des Batraciens de I'Ouest-africain. 111. Les dveloppement larvaire de deux es-

pcies rhophiles, Astylosternus diadematus et Petropedetes nutatol: Bull. Inst. France Afjtgue Noire 16:1222-1233. . 1956. Les formes larvaire de Chiromuntis mfescens, Ajkalus ~tosomus filvovittatus, and &sina senegaiensis. Bull. Inst. France Afique Noire 18:863-876. Lancaster, J., and A. G. Hiidrew. 1993. Characterizing instrearn f l o refugia. CanadianJ. Fish. Aqrrat. Sci. 50: 1663~ 1675. Lannoo, M. J. 1985. Neuromast topography inAmbystoma larvae. Copeia 1985:535-539. . 1987. Neuromast topography in anuran arnphibians. J Morphol. 191:115-129. . 1988. The evolution of the arnphibian lateral line system and its bearing on arnphibian phylogeny. Zeit. Zool. Syst. Ev01.-Fmsch.26: 128-134. Lannoo, M. J., and S. C. Smith. 1989. The lateral line system. pp. 176-186. In Deve@mental biology of the axolotl, edited by J. B. Armstrong and G. M. Malacinski. Oxford: Oxford University Press. Lannoo, M. J., D. S. Townsend, and R. J. Wassersug. 1987. Larval iife in the leaves: Arboreal tadpole types, with special attention to the morphology, ecology, and behavior of the oophagous Osteopilus bmnneus (Hyiidae) larva. Fieldiana 2001. (N.S.) (38):1-31. Lanza, B. 1983. Guideperi1 riconoscimnto delle specie animali delle q u e i n t m Italiane 27. Anfibi, Rettili. Consiglio Nazionale delie Ricerche. Largen, M. J. 1975. The status of the genus &sina (Amphibia Italiano Anura Hyperoliidae) in Ethiopia. Monit. 2001. Su.1. 6: 1-28. Larseli, 0.1925. The development of the cerebellum in the frog (Hyhregilla) in relation to the vestibular and lateral-line systems.J. Comp.Neurol. 39:249-289. . 1934. The difFerentiation of the peripheral and central acoustic apparatus in the frog. J. Comp. Neurol. 60:473527. . 1967. The wmparative anutomy and histology of the emebellumf?om myxinoidr through birak. Edited by J. Jansen. Minneapolis: University of Minnesota Press. Larson, , and R O. De S. 1998. Chondroaanial morphol! I ogy of Leptoductylus larvae (Leptodactylidae: Leptodactylinae): Its utility in phylogenetic reconstruction. J. Morphol. 238:287-305. Larson-Prior, L. J., and W. L. R Cmce. 1992. The red nucleus and mesencephalic tegmentum in a ranid arnphibian: A cytotechtonic and HRP conneaional study. Bmin Behav. Evol. 40:273-286. Lataste, F. 1879. tude sur le Discoglosse. Actes Soc. Linneenne Bordeaux 33: 275-341, plates 3-5. Latsis, R. V., and N. Y. Saraeva. 1980. Development of hemopoietic tissue in Rana temporaria. Arkh. Anat. Gistol. Emb h l . 78: 100-105. Lauder, G. i?1981. Form and funaion: Structural analysis in evolutionary morphology. Paieobwlogy 7:430442. Laurens, H. 1914. The reactions of normal and eye-less amphibian larvae to 1ight.J. E q . Zool. 16:195-210. Lauriia, A., and T Aho. 1997. Do female cornmon fiogs choose their breeding habitat to avoid predation on tadpoles? Oikos 78:585-591. Lauriia, A., J. Kujasalo, and E. Ranta. 1997. Ditferent antipredator behavior in two anuran tadpoles: Effects of predator diet. Behav. Ewl. Socwbwl. 40:329-336. Lavia, E. 0. 1984a. Larvas de Telmutobius (Anura: Leptodac-

386

LITERATURE CITED

tylidae) de la Provincia de Tucuman (Argentina). Acta Zool. Lilloana 38 :69-79. . 1984b. Redescripcion de larvas de Hylapulchella andina (Anua: Hylidae) con un analisis de la variabilidad interpoblacional. Neotrpua 30 :19-30. . 1985. Diagnosis generica y agrupacion de Ias especies de Telmatobius (Anua: Leptodactylidae) en base a caracteres larvales. Physis 43 :63-67. 1 9 8 7 . La larva de Rhinohmzadanvinii D. & B. (Anua: Rhinodermatidae).Acta Zool. Lilloana 39: 8 1-88. . 1988. Lower Telrnatobiinae (Anura: Leptodaaylidae): Generic diagnoses based on larval characters. Occac. Pap. Mus. Nat. Hist., Univ.Kansac (124):1-19. . 1990. The tadpole of Hyla nana (Anura: Hylidae). J. H~~/~etol. 24:207-209. . 1991. Condrocneo esqueleto visceral en larvas de Cycloramphus stejnepri (Le~tod&tylidae). Amphibia-Reptilia 12:33-38. . 1992a. Estmctura dei condrocrneoy esqueleto viscerd de larvas de Ahohs b a h i (Anua: Leptodaaylidae). Acta Zool. Lilloana 42: 13-17. . 1992b. The tadpole of Dmatonotus muehri (Anura: Microhylidae). Boll. Mus. Regwn. Sci. Nat. 10:63-71. Laviiia, E. O., and I. De la Riva. 1993. La larva de Telmatobius bolivianus (Anura, Leptodactylidae).Alpes 11:37-46. Laviiia, E. O., and M. Fabrezi. 1987. Anatomie de larvas de Hyla pulchella andina (Anura: Hylidae). Physis 45B :77-82. 1 9 9 2 . Anatomia craneal de larvas de iepiabatrmhusilunensis y Ceratophrys cranwelli (Anua: Leptodactylidae).Acta Zool. Lilloana 42 :5-1 1. Lavilla, E. O., and J. A. Langone. 1991. Ontogenetic changes in spiracular and proctodeal tube orientation in ElaEhistocieis buolor (Anura: Microhylidae).J. Hqetol. 25 :119-12 1. Lawler, K. L., and J.-M. Hero. 1997. Palatability of Bufo marinus tadpoles to a predatory fish decreases with development. Wiidl.Res. 24:327-334. Lawler, S. P. 1989. Behavioral responses to predators and predation risk in four species of larval anurans. Anim. Behav. 38: 1039-1047. Lawler, S. P., and P. J. Morin. 1993. Temporal overlap, competition, and priority effects in larval anurans. Ecology 74: 174-182. Lawson, D. P. 1993. The reptiles and amphibians of Komp National Park Project, Cameroon. Herpetol. Nat. Hist. 1:27-93. Lzr, G. Y., Z. S. Liposits, P. Tth, S. L. Trasti, J. L. Maderdrut, and I. Merchenthaler. 1991. Distribution of galanin-likeimmunoreactivity in the brain of Rana escuienta and Xenopus laevis.] Comp.Neurol. 3 10:45-67. . Lzr, G. Y., P. Tth, G. Y. Csank, and E. Kicliter. 1983. Morphology and location of tectal projection neurons in frogs: A study with HRP and cobalt-illing. J. Comp. Neurol. 215:108-120. Leaf, A. 1965. Transepithelial transport and its hormonal control in toad bladder. Physwl. Bwl. Chem. Exp. Phamzml. 56:216-263. Lebedinskaya, I. I., T. L. Padzyukevich, and G. A. Lasledov. 1989. Morphofunctional characteristics of myotomal muscle fibers from the tails of Rana temporaria tadpo1es.J. Evol. Bwchem. Physwl. 25:223-228. Lee, A. K. 1967. Studies in Austraiian Amphibia 1 .Taxonomy, 1 ecology, and evolution of the genus Heleiapms Gray (Anura: Leptodactylidae).AurtralianJ. Zool. 15:367-439. Lee, J. C. 1982. Accuracy and precision in anuan morphometrics: artifacts of preservation. Syst. Zool. 31 :266-28 1.

. 1996. The amphibians and reptiies ofthe Yucatn Peninsula. Ithaca, NY: Comstock Publ. Assoc. Lee, R. K. K., and R. C. Eaton. 1989. Segmental template for reticulospinal organization. Soc. Neurosci. Abrtr. p. 32. Lefcort, H. 1996. Adaptive, chemically mediated flight response in tadpoles of the southern leopard frog, Rana utricularia. Copeia 1996:455-459. . 1998. Chemically mediated fright response in southern toad (Bufotewescris) tadpoles. Cripeia 1998:45-450. Lefcort, H., and A. R. Blaustein. 1995. Disease, predator avoidance, and vulnerability to predation in tadpoles. Oikos 74:469474. Lefcort, H., and S. M. Eiger. 1993. Antipredatory behaviour of feverish tadpoles -implication for pathogen transmission. BehavMur 126:13-27. Leibovitz, H. E., D. D. Culiey, Jc, and J. P. Geaghan. 1982. Effects of vitamin C and sodium benzoate on survival, growth, and skeletal deformities of intensively cultured bulifrog larvae Rana catesbeiana reared at two pH 1evels.J. Worid Maricult. SOL. 13:322-328. Leips, J., and J. Travis. 1994. Metamorphic responses to changing food levels in two species of hylid frogs. Ecology 75: 1345-1356. Leloup-Hatey, J., M. Buscaglia, G. Jolivet-Jaudet, and J. Leloup. 1990. Interrenal function during the metamorphosis in anuran Amphibia. pp. 139-154. In Bwlogy andphyswlogy ofamphibians. Prgress in Zoology 38.Pmc. FintIntl. Symp. Bwl. Physwl. Amphibians (Karlsruhe, Federal Republic of Germany), edited by W Hanke. Stuttgart: Gustav Fischer-Verlag. Lemckert, F. 1996. An observation of aggregation behaviour by tadpoles of the great barred river frog (Mkophyes facciolatus). Herptofauna 26:43-44. Leon-Ochoa, J., and R. Donoso-Barros. 1970. Desarrollo embrionario y metamorfosis de Pieurohma brachyops (Cope) (Saliatia-Leptodaaylidae). Bol. Soc. Biol. Concepcion 42: 355-379. Le Quang Trong, Y. 1973. Stnicture et dveloppement de la peau et des glandes cutanes de Bufo regularis Reuss. Bull. SOL.Zool. Fr5nce 98:449485. . 1974. Etude de la peau et des glandes cutanes de&pus tropicalis Gray. Bull. Inst. France Afique Noire 26:407426. Lescure, J. 1968. Lc comportement social des batraciens. Rev. Comport.Anim. 2: 1-33. . 1976. Etude de deux ttards de Phyllobates (Dendrobatidae): Pfemoralis (Boulenger) et Ppictus (Bibron). Bull. Soc. Zool. France 101:299-304. . 1981. Contribution I'tude des amphibiens de Guyane franaise. IX. Le ttard gastromyzophore dXteiupus ~%vescens Dumril et Bibron (Anua, Bufonidae).AmphibiaReptilia 2:209-215. . 1984. Las larvas de dendrobatidae [sic]. (Proc.]II. Reunwn Iberoamer. Cons. Zool. Vert. 37-45. Leucht, T. 1987. Effects of weak magnetic fields on background adaptation inXenupuslaevis.Natunuissenschaften74: 192-194. Levine, R. L. 1980. An autoradiographic study of the retinal projection in Xenopus laevis with comparisons to Rana. J. Comp.Neurol. 189:1-29. Lewinson, D., M. Rosenberg, and M. R. Warburg. 1987. U1trastmctural and ultracytochemical studies of the giii epithelium in the larvae of Salamandrasaiamandra (Amphibia, Urodela). Zoommphoigy 107:17-25. Lewis, E. R. 1976. Surface morphology of the builfrog amphibian papia. Brain Behav. Evol. 13:196-215.

LITERATURE CITED Lewis, E. R., and R. E. Lombard. 1988. The arnphibian inner ear. pp. 93-123. In The evolutwn of the amphibian auitovy system, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W Waikowiak. New York: Wiley. Lewis, S., and C. Straznicky. 1979. The time of origin of the mesencephalic trigeminai neurons in Xenopus. J. Comp. Neurol. 183:633-646. Lewis, W M., Jr. 1970. Morphological adaptations of cyprindontoids for inhabiting oxygen deficient waters. Copeia 1970:319-326. Lewontin, R. C. 1955. The effects of popuiation density and composition on viability in Drosophila mehnogmter: Evolutwn 9:27-41. Leydig, F. 1853. Anatomisch-histologische Untersuchungen ubw Fische und Reptilien. Berlin. . 1857. Lehrbuch der Histologe des Menschen und der Thiere. Frankfurt: Meidinger Sohn & Co. . 1876. Ueber die Allgemeinen Bedeckungen der Amphibien. Arch. Mikros. Anat. 12:119-242. Li, J. C., and C. S. Lin. 1935. Studies of the "rain frog" Kaiuuh borealis 11. The food and feeding of the embryos and aduits. Peking Nat. Hist. Bull. 10:45-53. Licht, L. E. 1967. Growth inhibition in crowded tadpoles: Intraspecific and interspecific effects. Ecology 48:736-745. . 1968. Unpalatability and toxicity of toad eggs. Herpetob i c a 24:93-98. . 1969. Palatability of Rana and H y h eggs. American Midl. Nat. 88:296-299. . 1971. Breeding habits and embryonic thermai requirements of the frogs, Rana aurora aurora and Ranapretiosapetwsa, in the Pacific Northwest. Ecology 52: 116-124. . 1974. Survival of embryos, tadpoles and aduits ofRana aumra and Ranapretwsa sympatric in southwestern British Columbia. CanadianJ.Zool. 52:613-627. Lieberkund, I. 1937. Egkichende Studien uber die Mouphologe und Histgenese der Larvakn Haftmgane bei den Amphibien. Copenhagen: C. A. Reitzels Forlag. Liem, K. F. 1961. On the taxonomic status and the granular patches of the Javanese frog Rana chalconota Schlegel. Herpetologica 17:69-71. Liem, K. F., and D. B. Wake. 1985. Morphology: Current approaches and concepts. pp. 366-377. In Functional vertebrate mmphology, edited by M. Hildebrand, D. M. Brarnble, K. F. Liem, and D. B. Wake. Cambridge: Belknap Press. Ligname, A. H. 1964. Fixation method for arnphibian eggs. J. Roy. Micvoscop. Soc. 83:48 1 4 8 2 . Lillie, R. D. 1965. Histopathologic technic andpracticalhistochem3d ed. New York: McGraw-Hiil. Liilywhite, H . B. 1970. Behavioral temperature regulation in the bulifrog, Rana catesbeiana. Copeia 1970:158-168. Liilywhite, H. B., P. Licht, and P. Chelgren. 1973. The role of behavioral thermoregulation in the growth energetics of the toad, Buf boreas. Ecology 54:375-383. Limbaugh, B. A., and E. .E Volpe. 1957. Early development of ' the Gulf Coast toad, Bufo valliceps Wiegmann.AmericanMus. Nutitates (1842) :1-32. Lindquist, E. D., and T. E. Hetherington. 1998. Tadpoles and juveniles of the Panamanian golden frog,Ate&us zeteki (Bufonidae), with information on development of coloration and panerning. Herpetologica 54:370-376. Ling, R. W, J. P. Van Amberg, and J. K. Werner. 1986. Pond acidity and its relationship to larval development ofAmbystoma mmuhtum and Rana sylvatica in upper Michigan. J. Herpetol. 20:230-236.

387

w.

Linss, W., and G. Geyer. 1964. ber die elektronenmikroskopische Struktur der Nierentubui von Rana escuknta. Anat. Anz. 115:282-296. Lips, K. R., and J. M. Savage. 1996. Key to the known tadpoles (Amphibia: Anura) of Costa Rica. Stud. Neotrop. Fauna Environ. 31: 17-26. Littlejohn, M. J. 1963. The breeding biology of the Baw Baw frog Phiiuria f;ostzsh Spencer. Proc. Linnean Soc. New South Wales 88:273-276. Littlejohn, M. J., and A. A. Martin. 1964. The Crinia l m i complex (Anura: Leptodactylidae) in south-eastern Australia. Aus~valianJ. Zool. 12:70-83. . 1965. A new species of Crinia (Anura: Leptodactylidae) from South Australia. Copeia 1965:3 19-324. Liu, C.-C. 1940. Tadpoles of West China Salieni5a.J. W China Border Res. Soc. 12:7-62,4 plates. . 1950. Amphibians of western China. Fieldiana Zool. Mem. 2:1400, 10 plates. Liu, C.-C., and S. Q. Hu. 1961. TheAnura of China. Peking: Science Press. Liu, H., R. Wassersug, and K. Kawachi. 1996. A computational fluid dynamics study of tadpole swirnming. J. Exp. Zool. 199:1245- 1260. . 1997. The three-dimensional hydrodynamics of tadpole 1ocomotion.J. Exp. Bwl. 200:2807-2819. Lofts, B. 1974. Reproduction. pp. 107-218. In Physwlogy of the Amphibia, Vol. 2, edited by B. Lofts. New York: Academic Press. Lopez, K. 1989. Sex differentiation and early gonadal development in Bombina orientahs (Anura, Discoglossidae).J. Morphol. 199:299-311. Loschenkohl, A. 1986. Niche partitioning and competition in tadpoles. pp. 399402. In Stuies in herpetoiugy, edited by Z. RoEek Prague: Charles University Press. Lovtrup, S., and A. Pign. 1968. Obse~ations the giii develon opment in frogs. Ann. Embvyol. Morphol. 2:3-13. hvtrup, S., and B. Werdinius. 1957. Metabolic phases during arnphibian embryogenesis.J. Exp. Zool. 135:203-220. Lowe, D. A. 1986. Organization of lateral line and auditory areas in the midbrain of Xenopus laevis. J. Comp. Neurol. 245:498-5 13. . 1987. Single-unit study of lateral line ceiis in the optic tectum of Xenopw laevis: Evidence for bimodal lateral line/ optic units.J. Comp. Neurol. 257:396404. Lowe, D. A., and I. J. Russell. 1982. The central projections of the lateral line and cutaneous sensory fibres (VI1 and X) in Xenopus laevis. Proc. Roy. SOL.Lonhn 216B:279-297. Lucas, E. A., and W A. Reynolds. 1967. Temperature selection by yphibian larvae. Physwl. Zool. 40:159-171. Luckenbdl, L. M. 1964. Fine structure and development of horny jaws and teeth of the tadpole (Ranap$iens). Dissertation, Brown University . 1965. Morphogenesis of the horny jaws ofRanap$iens larvae. Deve&. Bwl. 11:2549. Ludwig, D., and L. Rowe. 1990. Life history strategies for energy gain and predator avoidance under time constraints. American Nat. 135:686-707. Lum, A. M., R. J. Wassersug, M. J. Potel, and S. A. Lerner. 1982. Schooling behavior of tadpoles: A potential indicator of ototoxicity. Phamml. Bwchem. Behav. 17:363-366. Luna, L. G., ed. 1968. Manual of histoiugic staining mthoak of the Amed Fmces Institute of Pathology. New York: McGrawHill Book. Luther, A. 1914. ber die vom N. trigeminus versorgte Musku-

388

LITERATURE CITED Maeda, N., and M. Matsui. 1989. Fmgs and to& of Japan. Tokyo: Bun-Ichi Sogo Shuppan Co., Ltd. Maetz, J. 1974. Aspects of adaptation to hypo-osrnotic and hyper-osrnotic environrnents. pp. 1-167. In BwchemiEal and bzaphysicalpenpectivesin marine bwqy, Vol. 1, edited by D.C. Malins and J. R. Sargent. New York: Academic Press. Magnin, E. 1959. Anatomie du ttard &Alpes obst&ns. Actes Soc. Linneenne Bordeam 98:33-79. Magnusson, W. E., and J.-M. Hero. 1991. Predation and the evolution of cornplex oviposition behaviour in Amazon rainforest frogs. Oecologa 86:3 10-3 18. Mahapatro, B. K., and M. C. Dash. 1987. Density dects on growth and rnetamorphosis of B u . stomaticus larvae. Alpes 6:88-98. Maher, S. W. 1962. Primary stmctures produced by tadpo1es.J. Sediment. P m l . 32:138-139. Main, A. R. 1957. Studies on Australian Amphibia. 1.The genus Crinia Tschudi in south-western Australia, and some species frorn south-eastern Australia. Australian J. Zool. 5:30-55. . 1968. Ecology, systernatics and evolution of Australian frogs. pp. 37-86. In Advances in ecologcal research, Vol. 5. Edited by J. B. Cragg. New York: Academic Press. Majecki, J., and K. Majecka. 1996. Predation by O l g o t d a s t r i ata caddis l a ~ a on amphibian eggs: Effeas of a high quality e food on growth rate. NetherlandrJ.Aquat. Ecol. 30:2 1-25. Malone, B. S. 1985. Mortality during the early life history stages of the Baw Baw frog, Philonafiosti (Anura: Myobatrachidae). pp. 1-5. In The bwlogyofAustralasianfigsand reptiles, edited by G. Grigg, R. Shine, and H. Ehmann. Chipping Norton, New South Wales: Surrey Beatty. Malvin, G. M. 1985a. Adrenoreceptor types in the respiratory vascuiature of the salamander gi. J. Comp. Physwl. 155B:591-596. . 1985b. Vascular resistance and vasoactivity of giiis and pulmonary artery of the saiamander, Ambystoma tz>rinum. J. Comp. Physwl. 155B:241-249. 1 9 8 8 . Gfl stmcture and function amphibian larvae. pp. 121-15 1. In Comparativepulmonaryphyswlogy: Current concepts, edited by S. C. Wood. New York: Marcel Dekker. Malvin, G. M., and R. G. Boutilier. 1985. Ventilationpersion relationships in Amphibia. pp. 114-124. In Circulatwn, respiration, and metabolism current comparative approaches, edited by R Giiles. Berlin: Springer-Verlag. Maivin, G. M., and M. P. Hlastala. 1986a. Effeas of lung volume and 0, and CO, content on cutaneous gas exchange in frogs. AmericanJ. Physwl. 20R941-946. . 1986b. Regulation of cutaneous gas exchange by environrnental 0, and CO, in the frog. Respir. Physwl. 65:99111. . 1988. Effeas of environmental 0, on blood flow and diffusing capacity in amphibian skin. Respil: Physwl. 76:229-242. Malvin, G. M., and C. Riedel. 1990. Autonomic regulation of cutaneous vascuiar resistance in the bullfrog Rana catesbeiana.J. Bwl. 152:425439. Manahan, D. T. 1990. Adaptations by invertebrate larvae for nutrient acquisition frorn seawater. Americun Zool. 30: 147-160. Mangia, F., G. Procicchiani, and H. Manelli. 1970. On the developrnent of the blood island in Xenupus luais ernbryos. ActaEmbryol. E+. 14:163-184. Mangold, 0.1936. Experimente zur Analyse der Zusamrnenarbeit der Keimblatter. Natum>senschafien 24:753-760.

I a m der Amphibien mit einern vergleichenden Ausblick uber den Adduaor rnandibulae der Gnathostornen, und einern Beitrag zum Vershdnis der Organisation der Anurenlarven. Acta Soc. Sci. Fennicm 44: 1-1 5 1, 1 plate. Lutz, B. 1944a. Obse~ations frogs without aquatic iarvae: on The hatching of Eleutherodmtylus p a m s Girard. Bol. Mus. Nac. Rio deJaneiro (15):5-7. . 1944b. Biologia e taxonomia de Zachaenuspamlus. Bol. M w . Nac. Rio deJaneiro (17):1-66. . 1948. Ontogenetic evolution in frogs. Evolutwn 2:29-39. Lyapkov, S. M. 1996. The influente of predators on brown frogs during and after rnetamorphosis. Adv. Amphib. Res. F m . Swiet Unwn 1:140-159. Lynch, J. D. 1971. Evolutionary relationships, osteology, and zoogeography of leptodactyloid fiogs. Misc. Publ. MW. Nat. H&., Univ. Kanm (53):1-238. . 1973. The transition frorn archaic to advanced frogs. pp. 133-182. In Evolutwnary b w k y of the anurans. Contemporary research on majorprobiems, edited by J. L. Vial. Colurnbia: University of Missouri Press. . 1986. Notes on the reproductive biology of Atefopus subomatus.J. Herpetol. 20: 126-1 29. Lynch, J. D., P. M. Ruiz, and J. V. Rueda. 1983. Notes on the distribution and reproductive biology of Centrolenegeckoideum Jimenez De la Espada in Colornbia and Ecuador (Amphibia: Centrolenidae). Stud. Neotrop. Fauna Environ. 18:239-244. Lynch, K. 1984. Growth and rnetarnorphosis of the r e m abdorninis rnusde in Ranaaipiens.J.Morphol. 182:317-337. Lynn, W. G. 1936. A study of the thyroid in ernbryos ofEieuthemdmtylus nubicola.Anat. Rec. 64:525-539. . 1940a. The developrnent of the skuli in the nonaquatic larva of the tree-toad Eleutherodmtylus nubicola. Bwl. Bull. 79:375. (Abstr.) . 1940b. The ernbryonic origin and developrnent of the pharyngeal derivatives in Eleutherodaclyus nubicola. Bwl. Bull. 79:376. (Abstr.) . 1942. The ernbryology of Eieutherodmtylus nubicola, an anuran which has no tadpole stage. Camegie Inst. Wmhzngcan Contrib.Embryol. 30:27-62. . 1948. The effeas of thiourea and phenylthiourea upon the developrnent of Eleutherodmtylus ricordii. Bwl. Bull. 94: 1-15. . 1961. Types of amphibian rnetamorphosis. American ZOO^. 1:151-161. Lynn, W. G., and A. Edelman. 1936. Crowding and rnetamorphosis in the tadpole. Ecology 17:104-109. Lynn, W. G., and B. Lutz. 1946a. The developrnent ofEZeutberodmtylusguentheri Stdnr. 1864 (Salientia).Bol. Mus. Nac. Rio deJaneiro (71):1 4 6 . . 1946b. The developrnent of Eleutherodmtylus n m t u s Lutz. Bol. Mus. Nac. Rio deJaneiro (79):1-30. Lynn, W. G., and A. M. Peadon. 1955. The role of the thyroid gland in direa developrnent in the anuran, Eleutherodmtylus martinhnsis. h w t h 19:263-286. Lynn, W. G., and Sister Alfred De Marie. 1946. The effea of thiouracil upon pigrnentation in the tadpole. Sczim 104:3 1 . Macgregor, H. C., and E. M. De1 Pino. 1982. Ribosornai gene ampiification in rnultinucleate oocytes of the egg brooding hylid frog Fiectonoturpygmaeus. Chromosomu 85 :475488. Mackay, W. C., and B. Schmidt-Nielsen. 1969. Osrnotic and diffusional water perrneabiiity in tadpoles and fiogs, Rana clamitans. Bull. Mount. Desert 1s. Bwl. Lab. 9:26-27.

w.

LITERATURE CITED hlaniatis, G. M., and V. M. Ingram. 1971a. Erythropoiesis during amphibian metamorphosis. I. Site of maturation of enh-ocytes in Rana catesbeiana. J. Cell Bwl. 49: 372-379. . 1971b. Erythropoiesis during amphibian metamorphosis. 1 .Immunochemical study of Iamal and adult hemoglo1 bins in Rana catesbeiana.J. Cell Bwl. 49:380-389. . 1971c. Erythropoiesis during amphibian metamorphosis. III. Immunochemical detection of tadpole and frog hemoglobins (Rana catesbeiana) in single erythrocytes.J. Cell Bwl. 49:390404. hanis, M. L., and D. L. Claussen. 1986. Environmental and genetic influentes on the thermal physiology of Rana sylaat&.J. T h m .Bwl. 11:31-36. hg -, M. J., and G. M. Ai Johari. 1985. Imrnunological memory and metamorphosis. pp. 420433. In MetammphoN edited by M. B d s and M. Bownes. Oxford: Clarendon , Press. >anning, M. J., and J. D. Horton. 1969. Histogenesis of hmphoid organs in larvae of the South African clawed toad, X i m p lmvis (Daudin). J. Embvyol. &p. Morphol. 22:265277. 1 9 8 2 . RES structure and function of the Amphibia. pp. 4 2 3 4 0 . In The reticuloendothelial system. A comprehensive &e, Vol. 3. Phylogeny and onto~eny, edited by N. Cohen and M. M. Sigel. New York: Plenum. Santeifel, Y. 1995. Chemicdy mediated avoidance of predators by Rana temporaria tadpoles.J. Herpetol. 29:461463. Shteifel, Y., and A. V. Lhushev. 1996. Avoidance of predator chemical cues by tadpoles of four East European anuran speaes (Bufi bufo, Rana antalis, R. Iessonae, and R. tempmaria). AdP. Amphib. Res. Fmm. SUaiet Unwn 1:161-180. Saateufel, G., O. Wess, and W Himstedt. 1977. Messungen am dioptrischen Apparat von Amphibienaugen und Berechnung der Sehrscharfe im Wasser und L&. Zool. Jahrb. 81:395406. S b , E. 1930. Zur Entwiddungsgeschichte des Vorderdarmes der Arnphibien. Zool Jahrb. Anat. 52:405486. hIarcus, H. 1908. Beitrage zur Kemtnis der Gymnophionen. I. ber das Schlundspaltengebiet. Arch. Mikros. Anat. E-.-Gesch. 77:695-774. lLuian, M. P., K. Sampath, A. R. C. Nirrnala, and T. J. Pandian. 1980. Behavioural response of Rana cyanuphylictZr tadpole exposed to changes in dissolved oxygen concentration. P w l . Behav. 25:35-38. hluinelli, M., S. Tei, D. Vagnetti, and M. P. Sensi. 1985. SEM observation on the development and the regression of the oral disc of Bufi vubaris l k a . Acta ~mbvyol Morphol. &p. 6:31-39. Siuinelli, M., and D. Vagnetti. 1988. Morphology of the oral disc of Bufi bufi (Salientia: Bufonidae) tadpoles.J Morphol. 195:71-81. S l u i q M. E, and R. M. Douglas. 1994. Shape quantization and the estimation of volume and surface area of reptile eggs. J. Hqetol. 28:281-291. bruiier, G. 1973. Limnology of the Congo and Amazon rivers. pp. 223-238. In Tropicalfirest ewsystems in A j k a and South L.inrma:A wmparative revim, edited by N. Cohen and A. M. Sigel. Washington, D. C.: Smithsonian Institution Press. h h h a i i , E., and G. C. Grigg. 1980a. Acclimation of CTM, LD,. and rapid loss of acclimation of thermal preferendum i t-adpoles of Limnodynastes peronii (Anura: Myobatrachiu &e). Austvalian Zool. 20:447455. . 1980b. Lack of metabolic acclimation to difierent ther-

389

mal histories by tadpoles of Limnodyn&s pmni (Anura: Leptodactylidae). Physwl. Zool. 53: 1-7. Marshd, G. A., R L. Amborski, and D. D. Cuiiey, Jr. 1980. Calcium and pH requirements in the culture of buiifrog (Rana catesbeiana) larvae. Proc. World Mricult. Soc. 11:445453. Martin, A. A, and A. K. Cooper. 1972. The ecology of terrestrial anuran eggs, genus Crinia (Leptodactylidae). Cupeia 1972:163-168. Martin, A. A., and M. J. Littlejohn. 1966. The breeding biology and larval development of Hyla jerpistewrj (Anura: Hylidae). Proc. Linnean Soc. New South Wales 9 1:47-57, 1 plate. Martin, W. 1991. Caliing phenology of savannah woodland frogs in northern Queensland, Australia. Honours Thesis, James Cook University, Townsville, Australia. Martinez, I., R. Aivarez, and I. Herraez. 1992. Skeletal malformations in hatchery reared Ranaperezi tadpoles. Anat. &c. 233:314-320. Martinez, I., R. Aivarez, J. M. Maiega, and M. P. Herraez. 1990. Estudio preliminar de 10s factores que afectan al desarroilo embrionarioi de Rana perezi en cautividad. Actas LU CongresoNm. Antuult. 621-629. Martinez, I., M. P. Herraez, and R. Aivarez. 1994. Response of hatchery-rearedRanaperezi larvae fed difierent diets. Aquaculture 128:235-244. Matesz, C. 1979. Central projection of the VIIIth cranial nerve in the frog. Neurosctena 4:2061-2071. Matesz, C., and G. Szkely. 1978. The motor column and sensory projections of the branchial cranial nerves in the frog.J. Comp.Neurol. 178: 157-176. Mathies, T., and R. M. Andrews. 1996. Extended egg retention and its influente on embryonic development and egg water balance: Implications for the evolution of viviparity. Physwl. Zool. 69: 1021-1035. Mathis, U., F. Schaeffel, and H. C. Howland. 1988. Visual optics in toad (BufO americanus). J Comp. Physwl. 163A:201-213. Matsurka, Y 1935. Uber die Entwicklung d a Vornierenkanal. chensystems der japanischen Kr6te (BufOjinmusus). Folia Anat. Japonica 13:417448. Matthews, L. H. 1957. Viviparity in Gastrotheca (Amphibia: Anura) and some considerations on the evolution of vivipirity. Bull. Soc. Zool. Fram 82:317-320. Maturana, H . R., J. Y. Lettvin, W. S. McCulloch, and W. H. Pitts. 1960. Anatomy and physiology of vision in the frog (Ranapipiens). Gen. Physwl. Suppl. 2: 129-175. J. Matz, G. 1975. Les grenouiues de Genre Mantella (Ranidae). Aqztarana 9:20-84. Maurer, F. 1888. Die Kiemen und ihre Gefasse bei anuren und urodelen Arnphibien und die Umbildungen der beiden ersten Arterienbogen bei Teleostiern. Gegenbaun Mmphol. Jahrb. 14:175-222. May, R. M. 1974. Stabiliq and comphiq in mode1 emsystems. 2d ed. Princeton: Princeton University Press. Mayhew, W W. 1965. Adaptations of the amphibian, Scaphwpus coucbi, to desert conditions. A m ' c a n Midl. Nat. 74:95109. Maynard Smith, J. 1964. Group selection and kin selection. Nature 201:1145-1147. McCarrison, R. 1921. Observations on the effects of fat excess Soc. on the growth and metamorphosis of tadpoles. Pmc. London 92B:295-303. McClanahan, L. L. 1975. Nitrogen excretion in arid-adapted amphibians. pp. 106-1 16. In EnGronmentalphyswlogy $de-

m.

390

LITERATURE CITED

sert mganisms, edited by N. F. Hadley. Stroudsburg, PA: Dowden, Hutchinson and Ross. McClendon, J. F. 1910. The development of isolated blastomeres of the frog's egg. AmmimnJ.Anat. 10:425-430. McCollurn, S. A. 1993. Ecologicai consequences of predatorinduced polyphenism in larvai hylid frogs. Dissertation, Duke University. McCollurn, S. A., and J. D. Leimberger. 1997. Predatorinduced morphologicai changes in an amphibian: Predation by dragonflies d e a s tadpole color, shape, and growth rate. Oecobgia 109:615-621. McCollum, S. A., and J. Van Buskirk. 1996. Costs and benefits of a predator-induced polyphenism in the gray treefrog Hyla cbrysoscelis. Edutwn 50: 583-593. McCormick, C. A. 1982. The organization of the octavolateraiis area in actinopterygian fishes: A new interpretation. J. Mmphol. 171:159-181. . 1988. Evolution of auditory pathways in the Amphibia. pp. 587-6 12. In Tbe evolution of tbe ampbibian auditqsystem, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W Walkowiak. New York: Wiley. McCranie, J. R., L. D. Wilson, and K. L. Williams. 1989. A new genus and species of toad (Anura: Bufonidae) with an extraordinary stream-adapted tadpole from northern Honduras. Occas. Pap. Mus. Nat. Hist., Univ.Kansas (129):1-18. McCutcheon, F. H. 1936. Hemoglobin function during the life history of the bullfrog. J. Cell. Comp. Pbyswl. 8:63-81. McDiarmid, R. W. 1978. Evolution of parenta1 care in frogs. pp. 127-147. In Tbe development of behavior: Comparativeand evobtwnary aspects, edited by R. M. Burkhardt and M. BekoE New York: STMP Press. McDiarmid, R. W., and R Altig. 1990. Description of a bufonid and two hylid tadpoles from western Ecuador. Alytes 8:5 1-60. McDiarmid, R. W., and S. Gorzula. 1989. Aspects of the reproductive ecology and behavior of the tepui toads, genus Oreopbrynella (Anura, Bufonidae). Copeia 1989:445-45 1. McDonald, D. G., J. L. Omg, and B. P. Simons. 1984. The influente of low pH environments on ion regulation in the larval stages of the anuran amphibian, Rana clamitans. Canadianf. 2001. 62:2171-2177. McEdward, L. R., and D. A. Janies. 1993. Life cycle evolution in asteroids: What is a larva?Bwl. Bull. 184:255-268. McIndoe, R., and D. G. Smith. 1984a. Functional anatomy of the internai gills of the tadpole of Litoria m'ngii (Anura, Hylidae). Zoomurpbob~y 104:280-291. . 1984b. Functional morphology of gills in larval amphibians. pp. 55-69. In Respiration and metabolism of embryonic vertebrates, edited by R. S. Seymour. Dordrecht: Dr. W.
McLachlan, A. 1981a. Interactions between insect larvae and tadpoles in tropical rain pools. Ecol. Entomol. 6: 175-182. . 1981b. Food sources and foraging tactics in tropical rain pools. Zoo1.J.Linnean Soc. 71:265-277. McLeman, I. S. 1988. Quantitative relationships between motoneuron and muscle development in&nopusla&~: Implications for motoneuron celi death and motor unit formation. J. Comp.Neurol. 271: 19-29. Medewar, P. B. 1953. Some immunological and endocrinologicai problems raised by the evolution of viviparity in vertebrates. Symp. Soc. Exp. Bwl. 7:328-338. Mhely, E. G. 1901. Beitrage zur Kemtniss der Engystomatiden von Neu-Guinea. T e m k z e i : Fuz. 24: 169-271. Mendelson, J. R., 1 1 1997. A new species of toad (Anura: Bu1.

J d .

fonidae) from the Pacific highlands of Guatemaia and southern Mexico, with comments on the status of Bufo valliqs macmm'status. Herpetobgicu 53 :14-30. Menke, M. E., and D. L. Claussen. 1982. Thermai acchation and hardening in tadpoles of the bullfrog, Rana cutesbeiana. J. Tberm. Biol. 7:215-219. Menzies, J. I. 1967. An ecologicai note on the frog Pseudhymenocbirus merlini Chabanaud in Sierra Le0ne.J. WestAfncanSci. Assoc. 12:23-28. Menzies, J. I., and R. G. Zweifel. 1974. Systematics of Litoria arfakiana of New Guinea and sibling species (Salientia, Hylidae).American Mus. Novitates (2558): 1-16. Merchant-Larios, H . 1978. Ovarian differentiation. pp. 47-81. In Tbe vertebrate ovay comparativebwlogy and evolution, edited by R. E. Jones. New York: Plenum. Merchant-Larios, H., and I. Villalpando. 1981. Ultrastructurai events during early gonadai development in Ranapipiens and &nopus laevik.Anat. Rec. 199:349-360. Mertens, R. 1938. HerpetologischeErgebnisse einer Reise nach Kamerun. Abb. Senckenberg. Naturf Ges. 443 :1-52. . 1960. Die Larven der Amphibien und ihre evolutive Bedeutung. 2001. Anz. 164337-358. Meseguer, J., A. Garca-Ayaia, and A. Lpez. 1996. Structure of the amphibian mesonephric tubule during ontogenesis in Rana ridibunda L. tadpoles: Early ontogenetic stages, renal corpuscle formation, neck segment, and peritoneal funnels. Anat. Embryol. 193:397-406. Meseguer, J., M. T. Lozano, and B. Aguilleiro. 1985. Ultrastructure of the renal granulopoietic tissue of the Rana ridibunda tadpole. J. Submicros. Cytol. 17:391-401. Metter, D. E. 1964. A morphologicai and ecological comparison of two populations of the tailed frog,Ascapbustnui Stejneger. Copeia 1964:181-195. . 1967. Variation in the ribbed frog, Ascapbus tnui Stejneger. Copeia 1967:634-649. Meyer, M., J. Frohlich, and I. Nagel. 1975a. Zur epidermaien Kugelzelie der Larve von Ascapbus truei Stejneger (Amphibia, Salientia, Ascaphidae): Ein biometrischer Beitrag. Zeit. Mikros. Anat. Foncb. 89: 1086-1098. Meyer, M., R. Guderjahn, and J. Frohlich. 1975b. ber Kugelzelien und die Stniktur der larvaien Epidermis von Froschlurchen. I. Ascapbus truei Stejneger. Zeit. Mikros. Anat. Fmsch. Leipzig 89:79-107. Michael, M. I., and M. A. Ai Adhami. 1974. The development of the thyroid glands in anuran amphibians of Iraq. J. 2001. 174:315-323. Michael, M. I., and A. Y Yacob. 1974. The development, . growth and degeneration of the pronephric system in anuran amphibians of Iraq. J. 2001.174:407417. Michael, S. F. 1997. Captive breeding of Eieutbero~lus antiliensis (Anura: Leptodactylidae)from Puerto Rico with notes on behavior in captivity. Herpetol. Ra7 28: 141-143. Michalowski, J. 1966. Studies on the relationship of Bombina bombina ( L i a e u s ) and Bombina variegata (Limaeus) II. Some taxonomic characters of tadpoles of both species and of tadpoles obtained from cross under laboratory conditions. Acta 2001.Cracov. 11:67-94. Mijares-Urrutia, A. 1990. The tadpole of Cen~oienella andina (Anura: Centrolenidae).J. Herpetol. 24:410-412. Millamena, O. M., E. J. Aujero, and I. G. Borlongan. 1990. Techniques for aigae harvesting and preservation for use in culture and as larvai food. Aquacult. Engineel: 9:295304. Millard, N. 1942. Abnormalities and variations in the v a s d x

LITERATURE CITED system of Xenupus h i s (Daudin). Trans.h y . Soc. Smth Af r i a 29:9-28. . 1945. The development of the arteriai system of Xenopus b i s , including experiments on the destruction of the larval aortic arches. Trans. Roy. Soc. South A f i m 30:2 17-234. . 1949. The development of the venous system of &nopus laepis. Trans.Roy. Soc. South A f i m 32: 55-96. Miranda, L. A., D. A. Paz, and A. Pisan. 1991. Efecto de Ia actividad tiroidea en Ia tolerancia al frio de larvas de Bufo arenarum. Rev.Espana Hep-tol. 6:55-60. ~Miranda,L. A., and A. Pisan. 1993. Efecto de la densidad poblacionai en larvas de Bufo arenarum producido a traves de seraies visuaies. Alytes 11:64-76. ~Mishra, K., and M. C. Dash. 1984. Larval growth and devell? opment of a tree frog, Rhaeophorus mmulatus Gray. T q . Ecol. 25:203-207. Mitchell, N., and R. Swain. 1996. Terrestrial development in the Tasmanian frog, Blyobatrachus nimbus (Anura: Myobatrachidae): Larval development and a field staging table. Pap. Pmc. Roy. Soc. Tasmania 130:75-80. Mitchell, S. L. 1983. Spatial distribution of eight species of anuran tadpoles. Thesis, Mississippi State University. Mitchell, S. L., and R. Altig. 1983. The feeding ecology of S o o oog lossusgardzneri.J.Hmpetol. 17:281-282. Miyamoto, M . M. 1985. Consensus cladograms and general classifications. Cladistics 1:186-189. Mobbs, I. G., V. A. King, and R. J. Wassersug. 1988. Prostaglandin E, does not inhibit metamorphosis of tadpole tails in tissue culture. Exp. Bwl. 47: 151-154. Mocquard, F. 1890. Recherches sur la faune herpetologique des Iles de Borno et de Paiawan. Nouv. Arch. Mus. Natl. H&. Nat. (3):115-168, plates 7-1 1. Mohanty, S. N., and M. C. Dash. 1986. Effects of diet and aeration on the growth and metamorphosis of Rana t&rina tadpoles. Aqwulture 51:89-96. Mohanty-Hejmadi, l?, S. K. Dutta, and S. C. Mallick. 1979. Life history of Indian frogs I1 The marbled baiioon frog Uperodonystoma Schneider.J. Zool. Soc. India 31:65-72. Mohawaid, A. I?, and S. Hennen. 1971. Ultrastructure of the germ plasm in eggs and in embryos of Ranapipiens. Dmelop. B w ~24:37-53. . Monayong Ako'o, M. 1981. L'evolution du tube digestif chez la larve de Phlynodun sandevsoni Parker (Amphibia, Anura, Phrynobatrachinae).Ann. Fm. Sci. Cameroun 28: 105-1 15. Montgomery, N. M. 1988. Projections of the vestibular and cerebellar nuclei in Rana pipiens. Brain Behav. Evol. 31:8295. Montgomery, N. M., and K. V. Fite. 1989. Retinotopic organization of central optic projections in Ranapipiens.J. Comp. Neurol. 283526-540. Montgomery, N. M., K. V. Fite, and A. M. Grigonis. 1985. The pretectal nucleus ient$omzis mesencephali of Rana pipiens. J. Comp.Neurol. 234:264-275. Montgomery, N., K. V. Fite, M. Taylor, and L. Bengston. 1982. Neural correlates of optokinetic nystagmus in the mesencephaion of Ranapipiens: A functional analysis. Brain Behav. Evol. 21:137-150. Mookerjee, H . K. 1931. On the development of the vertebral column of Anura. Philosoph. Trans. Roy. SOL. LoLondon 219B: 165-196. Mookerjee, H. K., and S. K. Das. 1939. Further investigation on the development of the vertebral column in Salientia (Anura).J. Morphol. 64: 167-209. Moore, J. A. 1949a. Pattems of evolution in the genus Rana.

391

pp. 315-338. In Genetzcs,paleontology,and molution, edited by G. L. Jepsen, E. Mayr, and G. G. Simpson. Princeton: Princeton University Press. . 1949b. Geographic variation of adaptive characters in Ranapipiens Schreber. Euolution 3: 1-24. . 1961. Frogs of eastern New South Waies. Bu11. Americanmus. Nat. H&. 121:149-386. . 1972. Heredzty and devekpment. 26 ed. New York: Oxford University Press. Moore, M. K., and S. G. Faragher. 1998. Tadpole jello: A technique for monitoring food consumption and manipulating the chemicai composition of larval foodstuffs. Helpetol. Rev. 29: 18-19. Moore, M. K., and V. R. Townsend, Jr. 1998. The interaction of temperature, dissolved oxgyen and predation pressure in an aquatic predator-prey system. Oikos 81:329-336. Moran, N. A. 1994. Adaptation and constraint in the complex life cycles of anitnals. Ann. Rev.Ecol. Syst. 25:573-600. Moreira, G., and A. I Lima. 1991. Seasonal patterns of juvenile ? recruiunent and reproduction in four species of leaf litter frogs in central Amazonia. Hqetokgiea 47:295-300. Morgan, B. E., N. I. Passmore, and B. C. Fabian. 1989. Metamorphosis in the frog Arthroieptella lightfooti (Anura: Ranidae) with emphasis on neuro-endocrine mechanisms. pp. 347-370. In Altematzue l$e-histoly styles ofanimals, edited by M. N. Burton. Dordrecht: Kluwer Acad. h b l . Morgan, M. J., J. M. O'Donnell, and R. F. Oliver. 1973. Development of the left-right asymmetry in the habenular nuclei of Rana temporaria.J. Comp.Neuml. 149:203-214. Morin, P. J. 1981. Predatory saiamanders reverse the outcome of competition among three species of anutan tadpoles. Science 212: 1284-1286. . 1983. Predation, competition, and the composition of larvai anuran guilds. Ewl. Monogr 53:119-138. . 1986. Interactions between intraspecific competition and predation in an amphibian predator-prey system. Ecology 67:713-720. . 1987. Predation, breeding asynchrony, and the outcome of competition among treefrog tadpoles. Ecology 68:675-683. . 1989. New directions in amphibian cornmunity ecology. Hqetologim 45 :124-128. Morin, l?J., and E. A. Johnson. 1988. Experimental studies of asyrnmetric competition among anurans. Oikos 53:398407. Morin, P. J., S. l? Lawler, and E. A. Johnson. 1988. Competition between aquatic insects and vertebrates: Interaction strength and higher-order interactions. Ecology 69:14011409. . 1990. Ecology and breeding phenology of larvai Hyh andersonii: The disadvantages of breeding late. Ecology 71: 1590-1598. Morris, J. L., and G. Campbeli. 1978. Renal vascular anatomy of the toad (Bufomarinus).Cell Tiss.Res. 189:501-5 14. Moss, B., and V. M. Ingram. 1968a. Hemoglobin synthesis during amphibian metamorphosis. I. Chemical studies on the hemoglobins from the larvai and addt stages of Rana catesbeiana.J. Mol. Bwl. 32:48 1-492. . 1968b. Hemoglobin synthesis during amphibian metamorphosis. 1 .Synthesis of adult hemoglobin following thy1 roxine administration.J. Mol. Biol. 32:493-504. Modton, J. M., A. Jurand, and H. Fox. 1968. A cytological study of Mauthner's cell in Xenupus l m i and R n temaa poraria during metamorphosis. J. Emblyol. lhp. Molphol. 19:415431.

392

LITERATURE CITED

Moury, J. D., and J. Hanken. 1995. Early cranial neural crest migration in the direct-developing frog, E l e u t h e r o ~ l u s wqui. ActaAnat. 153:243-253. Mudry, K. M., and R. R. Capranica. 1980. Evoked auditory activity within the telencephaion of the bullfrog (Rana catesbeiana). BrainRes. 182:303-311. . 1987. Correlation between auditory evoked responses in the thalamus and species-specific caii characteristics. I. Rana catesbeiana (Anura: Ranidae). J. Comp. Physwl. 160A:477489. Mudry, K. M., M. Constantine-Paton, and R. R. Capranica. 1977. Auditory sensitivity of the diencephdon of the leopard frog Ranapip'ens.J. Comp. Physwl. 114A:1-13. Mdaiiy, D. P. 1953. Observations on the ecoiogy of the toad Bufo canms. Copeiu 1953:182-183. Mer, F., and P. Sprumont. 1972. Entwicklung des Aditus laryngis bei Rana ridibundu. Licht-und elektronenoptische Befunde. Rev. Siusse 2001. 79: 1114-1 120. Munn, N. L. 1940a. Learning experiments with larval frogs. A preliminary rep0rt.J. Comp. Physwl. Psychol. 29:97-108. . 1940b. Learning experiments with larval fr0gs.J. Comp. Physwl. Psychol. 53:4434-45. Munro, A. F. 1953. The amrnonia and urea excretion of different species of Amphibia during their development and metamorphosis. Bwchem.J. 54:29-36. Muntz, L. 1975. Myogenesis in the trunk and leg during development of the tadpole of Xenopw W s (Daudin 1802).J. Embryol. Exp. Mmphol. 33 :757-774. Muntz, L., J. E. Homby, and M. R. A. Daiooi. 1989. A comparison of the distribution of muscle type in the tadpole tails of Xenopus h and Rana tempurana: An histological and ultrastructurai study. Tiss. Cell21:773-78 1. Muntz, W. R. A. 1962a. Microelectrode recordings from the diencephaion of the frog (Ranapipiens) and a blue sensitive system.J. Neuvophyswl. 25:699-711. . 1962b. Effectiveness of diferent colors of light in releasing positive phototactic behavior of frogs, and a possible function of the retinal projection to the diencephai0n.J.Neuvophyswl. 25: 712-720. Muntz, W R. A., and T. Reuter. 1966. Visual pigments and . spectral sensitivity in Rana t e m p k a and other European tadpoles. Viswn Res. 6:601-618. Murray, D. L. 1990. The effects of food and density on growth and metamorphosis in wood frogs (Rana ylvatica) from central Labrador. CanadianJ. Zool. 68: 1221-1226. Muto, Y 1971. Reduction in size or dosure of the spirade and . defective digital malformations in the toad larva reared in aerated water kept at a high temperature. Bull. Aichi Univ. Edu. 20:139-149. Myers, C. W., A. Paolillo O., and J. W. Daly. 1991. Discovery of a defensively malodorous and noaumai frog in the family Dendrobatidae: Phylogenetic significance of a new genus and species from the Venezuelan Andes.AmericanMus.Novitates (3002):1-33. Nachtigaii, W. 1974. Bwloghl mechanisms ofattmhment: The cmnparatite mmphokgy and bwengineeringo ogansfovlinkage, f s u h , and a b e h . New York: Springer-Verlag. . 1982. Biophysik der Fortbewegung. pp. 608-622. In Bwphysik, edited by W. Hoppe, H. M. Lohmann, and H . Ziegler. Berlin: Springer-Verlag. Nagai, Y., S. Nagai, and T. Nishikawa. 1971. The nutritional efficiency of cannibalism and an artificial feed for the growth of tadpoles of Japanese toad (Bufo atbarZc sp.). Agv. Bwl. Chem. 35:697-703.

Naito, I. 1984. The development of glomedar capiary tufts of the bulifrog kidney from a straight interstitial vessel to an anastomosed capiary network. A scanning electron microscopic study of vascular casts. Avch. Histol.Japonica 47441456. Naitoh, T., A. Miura, and H. Akiyoshi. 1990. Movements of the large intestine i the ,anuran larvae, Xenopw umi. Comp. n Biochem. Physwl. 97C:201-208. Naitoh, T., R. J. Wassersug, and R. A. Leslie. 1989. The physiology, morphology, and ontogeny of emetic behavior in anuran amphibians. Physwl. Zool. 62:819-843. Nakata, K., M. Sokabe, and R. Suzuki. 1982. A model for the crowding effect in the growth of tadpoles. Bwl. Cybem. 42: 169-1 76. Nanba, H. 1972. Ultrastructural and cytochemical studies on the thyroid gland of normal metamorphosing frogs (Rana japonica Guenther) I. Fine structural aspects.Avch. Himl. Japonica 34:277-291. Nathan, J. M., and V. G. James. 1972. The role of protozoan in the nutrition of tadpoles. Copeia 1972:669-679. Nation, J. L. 1983. A new method using hexamethyldisilazane for preparation of soft insea tissues for scanning electron microscopy. Stain Tech. 58:347-35 1. Naue, H. 1890. Ueber Bau und Entwicklung der Kiemen der Froschlarven. Zeit. Naturwiss. 63: 129-176. Neary, T. J. 1988. Forebrain auditory pathways in ranid frogs. pp: 233-252. In The evolution o the amphibian auditmy ystem, f edited by B. Fritzsch, M. J. Ryan, W. Wdczynski, T. E. Hetherington, and W Waikowiak. New York: Wiley. . Neary, T. J., and R G. Northcutt. 1983. Nuclear organization of the bulifrog diencephalon.J. Comp. Neuvol. 213 :262-278. Neary, T. J., and W. Wilczynski. 1977. Autoradiographic demonstration of hypothaiamic efferents in the bullfrog, Rana catesbeiana. Anat. Rec. 187:665. . 1986. Auditory pathways to the hypothaiamus in ranid frogs. Neuvosci. Lett. 71 :142-146. Nelson, C. E., and R Altig. 1972. Tadpoles of the microhylids Gnrtrophyne ele~ans G. &a. Hmpetokgica 28:381-383. and Nelson, C. E., and H. S. Cueiiar. 1968. Anatomical comparison of the tadpoles of the genera Hypopachus and Gnrtrophym. Copcia 1968:423424. Nelson, G. 1978. Ontogeny, phylogeny, paleontology, and the biogenetic law. Syst. Zool. 27:324-345. Newman, R. A. 1987. Effects of density and predation on Scaphwpus couchii tadpoles in desert ponds. Oewlogia 71:301-307. . 1988a. Adaptive plasticity in development of Scaphwpus corrchii tadpoles in desert ponds. Evolution 42:774-783. . 1988b. Genetic variation for larvai anuran (Scaphiopus couchii) development time in an uncertain environment. Evolutwn 42:763-783. . 1989. Developmental plasticity of Scaphiopw wuchii tadpoles in an unpredictable environment. Eco[ogy 70: 17751787. . 1994. Effects of changing density and food leve1 on metamorphosis of a desert amphibian, Scaphwpus wuchii. Ecokgy 75: 1085-1096. . 1995. Genetic variation for phenotypic plasticity in the larvai life history of spadefoot toads (Scaphwpus couchii). EVOlutwn 48: 1773-1785. Newman, R A., and A. E. Dunham. 1994. Size at metamorphosis and water loss in a desert anuran (Scaphioptrs corrchii). Copeia 1994:372-381. Nguenga, D., G. G. Teugels, M. Legendre, and F. Oiievier.

LITERATURE CITED 1997. First data on the experimentai evaluation of predation on Hetwobranchw longifiliz larvae (Siluroidei; Clariidae) by toad tadpoles (Buforegularis).Aqumult. Res. 28: 335-339. Nichols, H. W. 1975. Growth media-freshwater. pp. 7-24. In Handbook ofphywlogd methods. Culture methods andgrowth memurements, edited by J. R. Stein. Cambridge: Cambridge University Press. Nichols, R. J. 1937. Taxonomic studies on the mouth parts of larval Anura. Illinoiz Bwl. Monogl: 15(4):1-73. Nie, M., J. D. Crim, and G. R. Ultsch. 1999. Dissolved oxygen, temperature, and habitat selection by bulifrog (Ranacatesbeiana) tadpoles. Copeia 1999: 153-162. Nieuwkoop, P. D., and J. Faber. 1956. Normal tables ofxenopus laevis (Daudin).Amsterdam: North-Hoiiand. . 1967. Normal table of Xenopus laevis (Daudin):A systematical and chronologiml survey of the heiopmentfim thefertilized e til1 the end o metamrphosis. 2d ed. Arnsterdam: m f North-Holland. Nieuwkoop, E D., and L. A. Sutasurya. 1976. Embryological evidence for a possible polyphyletic origin of the recent amphibians. J. Embryol. Ekp. Morphol. 35: 159-167. Nieuwenhuys, R., and E Opdam. 1976. Structure of the brain stem. p. 811-855. In Frog neurobwlogy, a handbook, edited by R. Liins and W. Precht. Beriin: Springer-Verlag. Nikeryasova, E. N., and V. A. Golichenkov. 1988. Dynamics of pigment granule distribution in dermal melanophores of larval anurans. 2. Aggregation. S m & J. Deveiop. Bwl. 19:406412. Nikundiwe, A. M., and R Nieuwenhuys. 1983. The cell massa in the brainstem of the South African clawed frog Xenupus &s: A topographical and topological analysis. J. Comp. Neurol. 213: 199-219. Nina, H . L., and E. M. De1 Pino V. 1977. Estructura histologica de1 ovario de1 sapo Eleuthwodmtylus unimkatus y observaciones sobre e1 desarrollo embrionario. Rev. Univ. Catolim Ecuador 5:3141. Nishikawa, A., M. Kaiho, and K. Yoshizato. 1989. Cell death in the anuran tadpole tail: Thyroid hormone induces keratinization and tail-specific growth inhibition of epidermal cells. Deveiop. Bwl. 131:337-344. Nishikawa, K., and R. Wassersug. 1988. Morphology of the caudal spinal cord in Rana (Ranidae) and Xenupus (Pipidae) tadpoles. J Comp.Neurol. 269: 193-202. . 1989. Evolution of spinal nerve nurnber in anuran larvae. Brain Behav. Evol. 33: 15-24. Noble, G. K. 1917. The systematic status of some batrachians Mus. Nat. H&. from South America. BuU. Am&n 37:793-814. . 1925a. An outline of the relation of ontogeny to phylogeny within the Arnphibia, I, 11. AmericanMus. Nuvitates (165): 1-17, (166):1-10. . 1925b. The integumentary, pulrnonary and cardiac modifications correlated with increased cutaneous respiration in the Arnphibia: A solution to the "hairy frog" problem. J Morphol. 40:341416. . 1926a. The hatching process ofAlytes,E l e u t h e r o ~ l u s and other amphibians. AmericanMus.Nuvitates (229) :1-7. . 1926b. The importante of larval characteristics in the classification of South African Salientia.AmericanMus. Novitates (237):l-10. . 1927. The value of life history data in the study of the evolution of the Amphibia. Ann. New Tork Acad. Sci. 30:31128, plate 9. . 1929. The adaptive modifications of the arboreal tad-

393

poles of Hopiophyne and the torrent tadpoles of Stauroiz. BULI.AnzericanMus.Nat. H&. 58:291-334. . 1931. The bwlogy of the Amphibia. New York: McGraw-W. Noden, D. M., and A. De Lahunta. 1985. The embryology of hmestic animuls. Deveiopmental mechanizms and malformations. Baltimore: Williams and W h s . N o h n s k i , E., and R F. Inger. 1990. Uncoupling of related structural changes in metamorphosing torrent-dwelling tadpoles. Copeia 1990:1047-1054. Nodzenski, E., R. J. Wassersug, and R. F. Inger. 1989. Developmentai differences in visceral morphology of megophtyine pelobatid tadpoles in relation to their body form and mode of life. Bw1.J Linnean Soc. 38:369-388. Noland, R, and G. R. Ultsch. 1981. The roles of temperature and dissolved oxygen in microhabitat selection by the tadpoles of a frog (Ranapipiens)and a toad (Bufotewestris).Copeiu 1981:645-652. Noliert, A., and C. Nollert. 1992. DieAmphibien Europm. Besstimmung-Gefahrdung-Schua. Stuttgart: Franckh-Kosmos. Nomura, S., Y Shiba, Y. Muneoka, and Y Kanno. 1979a. A . . scanning and transmission electron microscope study of the premetamorphic papiiiae: Possible chemoreceptiveorgans in the oral cavity of an anuran tadpole (Ranajaponica). A&. H&ol. Japonica 42: 507-5 16. . 1979b. Developmend changes of premetamorphic and fungiform papiilae of the frog (Ranajaponica) during metamorphosis: A scanning electron microscopy. Hiroshima J Med. Sci. 28:79-86. Nordlander, R. H. 1984. Developing descending neurons of the early Xenupus taii spinal cord in the caudal spinal cord of early Xenupus. J Comp.Neurol. 228: 117-128. . 1986. Motoneurons of the ti of young Xenupus tadai p01es.J. Comp.Neurol. 253:403413. . 1987. Axonal growth cones in the developing amphibian spinal c0rd.J Comp.Neurol. 263:485496. Nordlander, R. H., S. T. Baden, and T. Ryba. 1985. Development of early brainstem projections to the ti spinal cord of ai Xenupus. J Comp. Neurol. 23 1:5 19-529. Nordlander, R H., J. W. Gazzerro, and H. Cook. 1991. Growth cones and axon trajectories of a sensory pathway in the amphibian spinal c0rd.J Comp.Neurol. 307:301-250. Nordlander, R. H., J. F. Singer, and M. Singer. 1981. An ultrastructural examination of early ventral root formation in ArnphibiaJ Comp.Neurol. 199:535-551. Nordlander, R. H., and M. Singer. 1982a. Spaces precede axons in Xempus embryonic spinal cord. Exp. Neurol. 75 :221-228. . 1982b. Morphology and position of growth cones in the developing Xenupus spinal cord. Deveiop. Brain. Res. 4: 181-193. Northcutt, R. G. 1974. Some histochemicd observations on the telencephalon of the bullfiog, Rana catesbeiana Shaw.J Comp.Neurol. 157:379-390. . 1980. Central auditory pathways in anamniotic vertebrates. pp. 79-1 18. In Proceedings in li$ sciences. Comparative studies ofhearing in vertebrates, edited by A. N. Popper and R. R. Fay. New York: Springer-Verlag. . 1981. Evolution of the telencephalon in nonrnarnrnals. Ann. Rev. Neurosci. 4:301-350. . 1986. Lungish neural characters and their bearing on sarcopterygian phylogeny.J Morphol. Suppl. 1:277-297. . 1989. The phylogenetic distribution and innervation of craniate mechanoreceptive lateral iines. pp. 17-78. In Tbe mechanosensory lateral line. Neurobwlogr and evolutwn, edited

394

LITERATURE CITED Omerza, F. F., and K. E. Aey. 1992. Redeployment of trigeminal motor neurons during metamorphosis. J. Cornp.Neurol. 325:124-134. Opatrny, E. 1973. Clay ccnests"of toad tadpoles. BritishJ. Herpetol. 4:338-339. Opdam, l?, M. Kemali, and R. Nieuwenhuys. 1976. Topological analysis of the brain stem of the frogs Rana esculenta and Rana catesbeiana.J. Cornp. Neurol. 165 :307-3 32. Orlando, K., and A. W. Pinder. 1995. Larval cardiorespiratory ontogeny and dometry in Xenupus aevis. Physwl. Zool. 68:63-75. Orr, P. R. 1955. Heat death. 11. Differential response of entire animal (Rana pipiens) and several organ systems. Physwl. Zool. 28:294-302. Orton, G. L. 1943. The tadpole of Rhinophrynus hrsalis. Oam. Pap. Mus. Zool., Unis.Michigan (472):1-7, 1 plate. , 1944. Studies on the systematic and phylogenetic significance of certain larval characters in the Amphibia Salientia. Dissertation, University of Michigan. 1 9 4 6 . The unknown tadpole. TuvtoxNavs 24:131-132. . 1949. Larval development of Neetophrynoides tomieri (Roux), with comments on direct development in frogs. Ann. Camegie Mus. 31:257-276. . 1952. Key to the genera of tadpoles in the United States and Canada. Arnerican Midl. Nat. 47:382-395. . 1953. The systematics of vertebrate larvae. Syst. Zool. 2:63-75. . 1954. Dirnorphism in larval mouthparts in spadefoot of the Scaphiopus harnmondi group. Copeia 1954:97-100. . 1957. The bearing of larval evolution on some problems in frog classification. Syst. Zool. 6:79-86. Osborne, P. L., and A. J. McLachlan. 1985. The effect of tadpoles on algal growth in temporary, rain-Med rock pools. Freshw. B d . 15 :77-87. Ott, M. E., N. Heisler, and G. R Ultsch. 1980. A reevaluation of the relationship between temperature and the critical oxygen tension in freshwater fishes. Cornp. Bwchern. Physiol. 67A:337-340. Ottoson, D. 1976. Morphology and physiology of muscle spindles. pp. 643-675. In Frogneurobwlogy, a handbook, edited by R. Llins and W Precht. Berlin: Springer-Verlag. Ovalle, W. K. 1979. Neurite complexes with Merkel ceiis in larval tentacles ofXenqpus b v t s . Cell Tiss.Res. 204:233-341. Ovaska, K. 1991. Reproductive phenology, popdation structure, and habitat use of the frog Eleuthemda@us johnstonei i Barbados, West 1ndies.J. Herpetol. 25:424430. n Overton, J . 1959. Studies on the mode of outgrowth of the amphibian pronephric duct. J. Ernbryol. -Exp. Morphol. 7:86-93. Oxner, W. M., J. Quinn, and M. E. DeMont. 1993. A mathematical model of body kinematics in swimming tadpoles. CanadianJ.Zool. 71 :407-413. Padhye, A. D., and H. V Ghate. 1988. Effect of altered p H on embryos and tadpoles of the frog Microhyh m a t a . Herpetol. J. 1:276-279. Paicheler, J.-C., F. De Broin, J. Gaudant, C. Mourer-Chauvire, J.-C. Rage, and C. Vergnaud-Grazzini. 1978. Le bassin lacustre Miocne de Bes-Konak (Anatolie-Turquie): Gologie et introduction Ia palontologie des vertbrs. Gobws (11):43-65. Pancharama, M., and S. K. Saidapur. 1985. Ovarian cycle in the frog Rana cyanuphlyctis: A quantitative study of foilicdar kinetics in relation to body mass, oviduct, and fat body cycles. J.Morphol. 186:135-147.

by S. Coombs, l? Gorner, and H. Munz. New York: Springer-Verlag. Northcutt, R. G., and G. J. Royce. 1975. Olfactory bulb projections in the bulifrog Rana mtesbeiana. J. Morphol. 145:251-268. Nowogrodzka-Zagrska, M. 1974. The organization of extraocular muscles in Anura. Acta Anat. 87:2244. Nyman, S. 1986. Mass mortality in larval Ranasylvatica attributable to the bacterium, Aeromunac hydrophih. J. Herpetol. 20:196-201. Odendaal, F. J., and C. M. B d . 1983. Water movements, tadpole competition, and lirnits to the distribution of the frogs Ranidelh riparia and R. signij2ra. Oecologia 57:361-367. Odendaal, F. J., C. M. Bull, and R. C. Nias. 1982. Habitat selection in tadpoles of Ranidelh signij2ra and R. riparia (Anura: Leptodaaylidae). Oecologia 52:411-414. Ogielska, M . , and E. Wagner. 1990. Oogenesis and development of the ovary in European green frog, fina ridibunda (Pdas). I. Tadpole stages until metamorphosis. Zool.Jahrb. Abt. Anat. Ontog. Tiere 120:211-221. O'Hara, R. K. 1974. Effects of developmental stage and prior experiente on habitat selection in three species of anuran larvae. Thesis, Michigan State University. . 1981. Habitat selection behavior in three species of anuran larvae: Environmental cues, ontogeny, and adaptive significance. Dissertation, Oregon State University. O'Hara, R. K., and A. R Blaustein. 1981. An investigation of sibling recognition in Rana u d tadpoles. Anirn. Behav. 29:1121-1126. . 1982. Kin preference behavior in Bufo borem tadpoles. Behav. Ewl. Socwbwl. 11: 4 3 4 9 . . 1985. Rana c a s d tadpoles aggregate with siblings: An experimental field study. Oewhgia 67:44-5 1. . 1988. Hyh regilh and Ranapehsa tadpoles fail to display kin recognition behaviour. Anirn. Behas. 36:946-948. Ohtani, O., and I. Naito. 1980. Renal microcircdation of the bulifrog, Rana catesbeiana. A scanning electron microscope study of vascular casts. Arch. Histol.Japonica 43 :319-330. Ohzu, E., N. Abe, and I. Nakatani. 1987. Scanning electron microscopic observations of the development of hatching gland cells in amphibian embryos. Bull. Yarnagata Uniu., Nat. Sci. 11:431438. Oka, Y., R. Ohtani, M. Satou, and K. Ueda. 1989. Location of forelimb motoneurons in the Japanese toad (Bufojaponicus): A horseradish peroxidase study. J. Cornp. Neurol. 286:376-383. Okada, Y. 1931. The tailless batrachians of the Japanese Empire. Irnp. &K Exp. Sta. 1-207, 29 plates. Oku, U., K. Katakura, J. Nagatsuka, and M. Kayima. 1979. Possible role of the tadpole of Rana chensinensis as an intermediate host of Angwstrongylus mntonensts parasitic to rodents. JapaneseJ. V&.Res. 27:1 4 . Okutomi, K. 1937. Die Entwicklung des Chondrocraniums von Polypedates buegeri schlegelii. Zeit. Anat. Entwickl. 107:28-64. Oldham, R. S. 1977. Terrestrial locomotion in two species of amphibian 1arvae.J. Zool. 181:285-295. Olson, D. H. 1988. The ecological and behavioral dynamics of breeding in three sympatric anuran amphibians. Dissertation, Oregon State University. . Olson, D. H., W l? konard, and R. B. Bury, eds. 1997. Sampling amphibians in lentic habitats: Methods and approaches for the Paciic Northwest. Nurthwest Fauna ( ) Soc. Nurthw. 4, Evt. Bwl. Olympia, WA.
-

LITERATURE CITED Pandian, T. J., and M. P. Marian. 1985a. Predicting anuran metamorphosis and energetics. Physwl. ,2201. 58:538-552. . 1985b. Time and energy costs of metamorphosis in the Indian buirog Rana tgrina. Copeia 1985:653-662. . 1985c. Physiological correlates of surfacing behaviour- effect of aquarium depth on surfacing, growth, and metamorphosis in Rana tgrina. Physwl. Behav. 35:867-872. Papernal, I., and R. Lainson. 1995.Allogu~ea bufinis nov. gen., nov. sp. (Microsporea: Glugeidae), a microsporidian ofBufo marinus tadpoles and metamorphosing toads (Amphibia: .bura) from Amazonian Brasil. Dzi Aguat. Org. 23:7-11. PLnchy, D. M., and R. H. Kaplan. 1992a. Developmental consequences of tail injury on larvae of the Oriental fire-beilied toad, Bombina mientalis. Copeia 1992:129-137. . 1992b. Maternal effects on offspring growth and development depend on environmental quality in the frog Bombina mientaltj. Oecologia 91: 579-586. . 1995. Maternal investment and developmental plasticity: Functional consequences for locomotor performance of hatchling frog larvae. Funct. Ecol. 9:606-6 11. F'arker, G. E. 1967. The iduence of temperature and thyroxine on oxygen consumption in Rana pipiens tadpoles. Copeia 1967:610-616. F'arker, H. W. 1934. A manograph o the Microhylidae. London: f Jarrold and Sons, London (67 figs). . 1936. A coilection of reptiles and amphibians from the mountains of British New Guinea.Ann. MaJ. Nat. Hist. Ser. 10. 17:66-93. Parker, W. K. 1871. On the structure and development of the s k d of the common frog (Rana tempmaha L.). Philosoph. Tmm. Roy. Soc. London 161B:137-2 11. . 1876. On the stmcture and development of the skull in the Batrachia. Part 11. P m . Zool. Soc. London 1876:601669. . 1881. On the structure and development of the skull in the Batrachia. Part 1 1 Philosoph. Trans. Roy. Soc. London 1. 172: 1-266. Purridge, B. L. 1982. The structure and tiinction of fish schools. Sci. Amehmn 246: 114-123. Purridge, B. L., and T. J. Pitcher. 1980. The sensory basis of fish schools: Relative roles of lateral line and visi0n.J Comp. P w l . 135A:315-325. Paterson, N. F. 1939a. The head of Xenopus l m i . Qwrt.J. X ~ S . 81:161-234. Sci. . 1939b. The olfactory organ and tentades of Xelzc~pus InepLs. South A f i a n J . Sci. 36: 390404. Parrerson, J. W., and A. J. McLachlan. 1989. Larval habitat duration and size at metamorphosis in frogs. Hydrobwlogia 171:121-126. P r i ,M. J., ed. 1956. Ovid's Metamwphoses. 2d ed. Gronarn ingen: J. B. Wolters. Pa\ignano, I. 1989. Method employed to study the diet of anuran amphibians Iarvae. Amphibia-Reptilia 10:453-456. . 1990. Niche overlap in tadpole populations of Pelobates f i a s insubricus and Hyla arbmea at a pond in northwestern Iraly. Boll. Zool. 57: 83-87. Peacor, S. D., and E. E. Werner. 1997. Trait-mediated indirect interactions in a simple aquatic food web. Ecology 78: 11461156. Pearman, P. B. 1995. Effects of pond size and consequent predatory density on two species of tadpoles. Oewloga 102:1-8. h a n , P. B., and H. M. Wilbur. 1990. Changes in population dynamics resulting from oviposition in a subdivided habitat. A&an Nat. 135:708-723.

395

Pearse, A. G. 1968. Histochemistry: Theoreticaland applied. Vols. I and 11. Boston: Little, Brown and Co. Pehek, E. L. 1995. Competition, pH, and the ecology of larval Hyla andmoni. E w k y 76: 1786-1793. Peixoto, O. L. 1981. Notas sobre o girino de Crossodactylodes pintoi Cochran (Amphibia, Anura, Leptodaaylidae). Rev. Brmil. Bwl. 41: 339-341. . 1982. Observaes sobre a larva de Pleurodema diplalistri (Peters 1870)(Amphibia,Anura, Leptodaaylidae). Rev. Brasil. Bwl. 42:631-633. Peixoto, O. L., and C. A. G. Da Cruz. 1983. Girinos de espcies de Hya do grupo 'Xlbomarginata" do Sueste Brasileiro (Amphibia, Anura, Hylidae).Arg. Unzv. Fed. Rural& deJaneiro 6: 155-163. Pelaz, M. P. 1987. On the identity of the so-called "algae like ceh" in tadpole cultures of European green frogs (Ranaridibunda).Alpes 6:23-26. Pelaz, M. P., and C. Rougier. 1990. A comparative study of buccal pump volumes and the cartilagenous branchial basket skeleton of Rana ridibunda and Rana dalamatina tadpoles. Copeia 1990:658-665. Prez-Rivera, R. A., and E. Nadal. 1993. Dos nidos de coques (Eleutherodactylus) en nidos de Aves. Caribbean J Sci. 29: 128. Perret, J.-L. 1966. Les amphibiens du Cameroun. Zool. Jahrb. Syst. 93:289464. Persson, L. 1988. Asymmetries in competitive and predatory interactions in fish populations. pp. 203-218. In Si=structuredpopuhtwns: Ewlogy and evolutwn, edited by B. Ebenmann and L. Persson. Berlin: Springer-Verlag. Petersen, H. 1922. Zur Konstruktionanalyse des Kieferapparates der Froschlarven [V. Beitrag zur Mechanik des Tierkorpers.]. B e i Zeit. Ges. Anat. 64:22-28. ~ Peterson, J. A., and A. R. Blaustein. 1991. Unpalatability in anuran larvae as a defense against natural salamander predators. Ethol. Ecol. Evol. 3:63-72. Petranka, J. W. 1985. Does age-specific mortality decrease with age in amphibian larvae?Copeia 1985:1080-1083. . 1989. Chemical interferente competition in tadpoles: Does it occur outside laboratory aquaria? Copeia 1989:921-930. . 1995. Interferente competition in tadpoles: Are multiple agents involved?Hepto1.J. 5:206-207. Petranka, J. W., M. E. Hopey, B. T. Jennings, S. D. Baird, and S. J. Boone. 1994. Breeding habitat segregation of wood frogs and American toads: The role of interspecific tadpole predation and adult choice. Copeia 1994:691-697. Petranka, J. W., and L. Hayes. 1998. Chemically mediated avoidance of a predatory odonate (Anmjunius) by American toad (Bufoamericanus) and wood frog (Rana sylvaticu) tadpoles. Behav. Ewl. Socwbwl. 42:263-271. Petranka, J. W., J. J. Just, and E. C. Crawford. 1983. Hatching of amphibian embryos; the physiological trigger. Science 217:257-259. Petranka, J. W., J. J. Just, E. C. Crawford, L. B. Kats, and A. Sih. 1987. Predator-prey interactions among fish and larval amphibians: Use of chemical cues to detect predatory fish. Anim. Behav. 35 :420425. Petranka, J. W., A. W. Rushlow, and M. E. Hopey. 1998. Predation by tadpoles of Rana sylvatica on embryos o f A m b y s t m mmulatum: Implications of ecological role reversals by Rana (predato;) and ~ m b ~ s t o m(prey). ~ e r p e t o l o g h d 54:l-13. Petranka, J. W., and D. A. G. Thomas. 1995. Explosive breed-

396

LITERATURE CITED Piiper, J. 1982. Respiratory gas exchange at lungs, gi, and tissues: Mechanisms and adjustments.J. Exp. Bwl. 100:5-22. Piiper, J., and P. Scheid. 1977. Comparative physiology of respiration: Functional analysis of gas exchange organs in vertebrates. pp. 219-253. In Intl. Rev. Physwl., Reqil: Physiol. 11, Vol. 14, edited by J. G. Widdicombe. Baltimore: University Park Press. Piai, R S. 1978. On NyctibatraEhus major Boul. (Ranidae) with a descnption of its tadpole. Bull. Zool. Suw. India 1:135-140. Piiai, R. S., and G. M. Yazdani. 1973. Bufmdes, a new genus for the rock-toad,Ansonia meghaiuyana Yazdani and Chanda, with notes on its ecology and breeding habits. J. Zool. Soc. India 25:65-70. Pinder, A. W., and W W. Burggren. 1983. Respiration during chronic hypoxia and hyperoxia in larvai and adult bulifrogs (Rana catesbeiana).J.Exp. Bwl. 105:205-213. Pinder, A. W., and M. E. Feder. 1990. Effect of boundary layers on cutaneous gas exchange.J. Exp. Bwl. 143:67-80. Pinder, A. W., B. Pelster, and R G. Boutilier. 1990. Regulation of blood flow in isolated amphibian skin. Amezcan Zool. 30:136 (Abstr.). Pinder, A. W, K. B. Storey, and G. R. Ultsch. 1992. Estivation and hibemation. pp. 250-274. In Environmentdphysiology of the amphibians, edited by M. E. Feder and W. W. Burggren. Chicago: University of Chicago Press. Pintar, T. 1987. Zur Kenntnis des Tomatenfrosches Dyscophw antongili (Grandidier 1877)(Anura: Microhylidae). Salamandra 23: 106-121. Pitcher, T. J. 1986. The behavior of teleostfishes. Baitimore: John Hopkins University Press. Plasota, K. 1974. The development of the chondrocranium (neurocranium and the mandibular and hyoid arches) in Rana temporaria L. and Pelobatesfiscus (Laur.).Zool. Poloniae 24:99-168. Plassman, W. 1980. Central neuronai pathways in the lateral line system ofXenopus&S. J. J m p . Physwl. 136:203-213. Plytycz, B., and J. Bigaj. 1983. Amphibian lymphoid organs: A review. Folia Bwl. (Krakw) 3 1:225-240. Plytycz, B., J. Dulak, and A. Pecio. 1984. Genetic control of length of larvai period in Rana temporaria. Folia Bwl. (Krah) 32:155-166. Plytycz, B., and M. Klimek. 1992. Effects of diet on transplantation immunity of tadpoles of the European cornmon frog, Rana temporaria.J. Nu& Immunol. 1:19-30. Poinar, G. O., Jr. 1988. Infeaion of frog tadpoles (Amphibia) with insect parasitic nematodes. Expmientia 44528-53 1. Polis, G. A. 1984. Age structure component of niche width and intraspecific resource partitioning: Can age groups funaion as ecological species?American Nat. 123541-564. . 1988. Exploitation cornpetition and the evolution of interferente, cannibalism, and intraguild predation in age/ size-structured populations. pp. 185-202. In S i m - ~ u r e d populatwns: Ecology and evolution, edited by B. Ebenmann and L. Persson. Berlin: Springer-Verlag. Polis. G. A.. and C. A. Mvers. 1985. A survev of intras~ecific piedatio; among rephes and amphibik. J. ~&petol. 19:99-107. Poole, T. J., and M. S. Steinberg. 1977. SEM-aided analysis of morphogenetic movements: Development of the amphibian pronephric duct. pp. 43-52. In Scanning electron micro~copyl 1977. Vol. 2, edited by O. Johari. Chicago: ITT Res. Inst. Pope, C. H. 1931. Notes on the amphibians from Fukein, Hai-

ing reduces egg and tadpole cannibalism in the wood frog, Rana ylvatica. Anim. Behav. 50:73 1-739. Petrini, S., and F. Zaccanti. 1998. The effects of aromatase and 5 alpha-reductase inhibitors, antiandrogen, and sex steroids on Bidder's organ development and gonadai differentiation in Bufo bufi tadpo1es.J. Exp. Zool. 280:245-259. Pemgrew, A. G. 1981. Brainstem afferents to the t o m sernicird a r i s of the Queensland cane toad (Bufomarinw).J. Comp. Neurol. 202:59-68. PfeiTer, W. 1966. Die Verbreitung der Schreckreaktion bei Kaulquappen und die Herkunft des Schreckstoffes. Zeit. E&. Physwl. 52:79-98. . 1974. Pheromones in fishs and Amphibia. pp. 269f 296. In Pheromone: Fmmtim o bwlogy. Vol 32, edited by M. C. Birch. New York: North Holland. Pfennig, D. W. 1990a. The adaptive significance of an environmentaiiy cued developmentai switch in an anuran tadpoles. Oecologa 85: 101-107. . 1990b. "Kin selection" among spadefoot toad tadpoles: A side-effect of habitat seleaion?Evolutwn 44:785-798. . 1992a. Polyphenism in spadefoot toad tadpoles as a locally adjusted evolutionarily stable strategy. Evolution 46: 1408-1420. . 1992b. Proximate and funaional causes of polyphenism in an anuran tadpole. Funct. Ecol. 6: 167-174. Pfenning, D. W., and W. A. Frankino. 1997. Kin-mediated morphogenesis in fadtatively cannibalistic tadpoles. Evolutwn 51: 1993-1999. Pfenning, D. W, W. A. Frankino, A. Mabry, and D. Orange. 1991. Environmentai causes of correlations between age and size at metamorphosis in Scaphwpus multiplicatus. Ewlogy 72:2240-2248. Pfennig, D. W, H. K. Reeve, and P. K. Sherman. 1993. Kin recognition and cannibalism in spadefoot toad tadpoles. Anim. Behav. 46:87-94. Pflugfelder, O., and G. Schubert. 1965. Elektronenmikroskopische Untersuchungen an der Haut von Larven-und Metamorphosestadien von Xenupus luevis nach Kaliumperahlanatbehandlung. Zeit. ZellfDvsch.Mikros. Anat. 67:96-112. Piatka, B. A., and C. W Gibley, Jr. 1967. Alkaline phosphatase activity in the developing pronephros of the frog (Ranapipiens). TexasRept.Bwl. Med. 25:83-89. Piavaux, A. 1972. B-polysaccharidases du tube digestifde Rana tempuraria L. pendant Ia croissance et Ia mtamorphose. Arch. Intl. Physwl. Bwcbim. 80:153-160. Pick, J. 1970. The autonomic newous system. Philadelphia: J.B. Lippincott. Picker, M. D. 1985. Hybridization and habitat selection inXenopusgilli and Xenopw & in south-western Cape Province. s Copeia 1985:574-580. Picker, M. D., C. J. McKenzie, and P. Fielding. 1993. Embryonic tolerance of Xenopus (Anura) to acidic blackwater. Copeia 1993:1072-1081. Pickersgill, M. 1984. Three new Afrucdus (Anura: Hyperoliidae) frorn south-eastem Africa. Durban Mus. Nuvitate5 13:207-220. Pierce, B. A. 1985. Acid tolerance in amphibians. BwSciem 35:239-243. Pierce, B. A., and 7. M. H a ~ e y 1987. Geographic variation in . acid tolerance of Conneaicut wood frogs. Cupeia 1987:94103. Pierce, B. A., J. B. Hoskins, and E. Epstein. 1984. Acid tolerante in Connecticut wobd frogs (Ranasylvatica).J. Herpetol. 18:159-167.

LITERATURE CITED nan, and other parts of China. Bull. Amrican Mus. Nat. H&. 61:397-611. . 1940. China's animalfruntim New York: V i g . Porges, R., L. A. Van Den Berg, and D. G. Banger. 1986. Reassessing an old problem - acid mine drainage. Pmc. A d can Soc. Civil E n g ~(San. Engr. Div.) 92(SA-1)369-83. Poska-Teiss, L. 1930. ber die larvale Amphibienepidermis. Zeit. Zellforsch. Mikms. Anat. 11:445483. Postek, M. T., K. Howard, A. H. Johnson, and K. L. McMichael. 1980. Scanning electron microscopy:A student's handbook. Buriington, VT: Ladd Res. Ind., Inc. Potel, M. J., and R. J. Wassersug. 1981. Computer tools for the analysis of schoohg. Envimn. Bwl. Fkh. 6: 15-19. Potter, H. D. 1965. Mesencephalic auditory region of the b d frog. J. Neurophyswl. 28: 1132-1 154. Potthoff, T. L., and J. D. Lynch. 1986. Interpopuiation variability in mouthparts in Scaphiupus bombzj+ons in Nebraska (Amphibia: Pelobatidae). Prairie Nat. 18:15-22. Pough, F. H., and S. Kamel. 1984. Post-metamorphic change in activity metaboiism of anurans in relation to Me history. Oewloga 65:138-144. Pouyet, J. C., and A. Beaumont. 1975. Ultrastructure du pancras larvaire d'un Amphibien Anoure, Alpes obrtetricans L. en culture organotypique. C.R. Soc. Bwl. Paris 169:846-850. Powell, G. V. N. 1974. Experimental analysis of the social value of Aocking by stariings (Stumus aubans) in relation to predation and foraging. Anim. Behav. 22: 501-1505. Poweii, T. L., and J. J. Just. 1985. Development of the urinary bladder in Rana catesbiana [sic] tadpoles. A m ' c a n 2001. 25:99. (Abstc) . 1987. Ultrastructure and function of the developing urinary bladder in Rana catesbeiana. American Zool. 27: 3. (Abstr.) Power, J. H. 1927a. Some tadpoles from Griqualand West. Trans.Roy. Soc. South Ajbca 14:249-25 1, 14 plates. . 1927b. On the herpetological fama of the LobatsiLinokana area. Trans Roy. Soc. Smth AfriEa 14:405422. Power, J. H., and W Rose. 1929. Notes on the habits and Mehistories of some Cape Peninsula Anura. Trans. Roy. Soc. SouthAfTcca 17:109-115. Prakash, R. 1954. The heart and its conducting system in the tadpoles of the frog, Rana tgrina Daudin. Pmc. Zool. Soc. Caicutta 7:28-36. Prasadmurthy, Y. S., and S. K. Saidapur. 1987. Role of fat bodies in oocyte growth and recruitrnent in the frog Rana cyanophlyctis Schn.J. Exp. Zml. 243: 153-162. Precht, W 1976. Physiology of the peripheral and central vestibular systems. pp. 481-512. In F r g neurobwlogy, a handbook, edited by R. Llins and W Precht. Beriin: SpringerVerlag. Prestige, M. C. 1967. The control of ceii number in the lumbar ventral horns during the development of Xenupw Ia&> tadp01es.J. Embryol. Exp. Mmphol. 18:359-387. . 1973. Gradients in time of origin of tadpole motoneurons. Brain h 59:400404. . Preston, F. W 1968. The shape of bird eggs: Mathematical aspects. Auk 85:454463. Pretty, R., T. Naitoh, and R. J. Wassersug. 1995. Metamorphic shortening of the alirnentary tract in anuran larvae (Rana catesbeiana).Anat. Rec. 242:4 17423. Price, G. C. 1897. Development of the excretory organs of a myxinoid Bdellortoma stouti Lockington. Zool. Jahrb. Abh. Anat. Ontog. Time. 10:205-226. Prigioni, C. M., and D. Arrieta. 1992. Descripcion de Ia larva

397

de Melaophynkcus sanmartini Klappenbach, 1968 (Amphibia: Anura: Bufonidae). Bol. Soc. Zool. Uwguay 7157-58. Prigioni, C. M., and J. A. Langone. 1990. Descripcion de Ia larva de Melanophryniscus mejasmirandai Prigioni & Langone, 1986 (Amphibia, Anura, Bufonidae). Comun. 2001. Mus. H&. Nat. Monteadeo 12:1-9. Pronych, S., and R. Wassersug. 1994. Lung use and development in Xenopus W s tadpoles. Canadian J. Zool. 72:738-743. Prosser, C. L. 1973. Comparativeanimalphyswlogy. Philadelphia: W. B. Saunders. Prosser, C. L., and F. A. Brown, Jr. 1961. Comparative animal pbsology. 2d ed. Philadelphia: W. B. Saunders. Pgener, L. A., and A. M. Magiia. 1997. Osteology and skeletal development of Discoglossus sardus (Anura: Discoglossidae). J. Mmphol. 233:267-286. Puiiiam, H. R. 1973. On the advantages of Aocking. J. Thew Bwl. 38:419422. Puiiiam, H. R., and T. Caraco. 1984. Living in groups: 1s there an optimal group size? pp. 122-147. In Behavhural ecology: An evolutionay apoach, edited by J. R. Krebs and N. B. Davis. Sunderland, MA: Sinauer Assoc. Punzo, F. 1976. The effects of early experiente on habitat selection in tadpoles of the Malayan painted frog, Kaloulapulchra (Anura: Microhylidae). J. Bornbay Nat. H&. Soc. 73:270277. . 1991. Group learning in tadpoles of Rana heckrcheri (Anua: Ranidae).J. Herpetol. 25 :214-217. Pusey, H. K. 1938. Structural changes in the anuran mandibular arch during metamorphosis, with referente to Rana temp m a k . Quart.J.Micms. Sci. 80:479-552. . 1943. On the head of the iiopelmid frog,Rrcaphm t s e i . I. The chondrocraniurn, jaws, arches, and muscles of a partly-grown larva. Quart.J.Micros. Sci. 84: 105-185. Pyburn, W F. 1967. Breeding and larval development of the hyiid frog Phynohyas spilmnma in southern Veracniz, Mexico. He'petohgica 23 :184-1 94. . 1975. A new species of microhylid frog of the genus Synapturanw from southeastern Colombia. Herpetologica 31:439-443. . 1980. An unusual anuran larvae from the Vaups region of southeastern Colombia. Pap. Aaul. Zool. 33:23 1 238. Pyles, R. A. 1987. Morphology and rnechanics of the jaws of anuran amphibians. Dissertation, University of Kansas. Quaiis, C. l?, and R. Shine. 1995. Maternal body-volume as a constraint on reproductive output in lizards: Evidence from the evolution of viviparity. Oecologa 103:73-78. . 1996. Reconstructing ancestral reaction norms: An example using the evolution of reptilian viviparity. Funct. Ecol. 10:688-692. Quinn, D., and W Burggren. 1983. Laaate production, tissue distribution, and elimination foiiowing exhaustive exercise in larval and aduit bullfrogs Rana catesbeiana. Physiol. Zool. 56:597-613. Rabb, G. B., and M. S. Rabb. 1960. On the mating and egglaying behavior of the Surinam toad, Pipa pipa. Copeia 1960:271-276. Race, J., Jr., C. Robinson, and R. J. Terry. 1966. The influente of thyroxine on the normal development of the pancreas in Ranaptpkns 1arvae.J.Exp. 2001. 162:181-192. Rada de Martinez, D. 1990. Contribucion a1 conocirniento de Ia larvas de anfibios de Venezuela. Mem. Soc. Cknc. Nat. La Salle 49/50:391403.

398

LITERATURE CITED

Radakov, D. V. 1973. Schooling in the ecology offih. Israel Programmefor Sctentzfic Transhtwns. New York: Wiley. R d , R. A. 1987. Constraint, flexibility, and phylogenetic history in the evolution of direct development in sea urchins. Deveup. Bwl. 119:6-19. Raff, R. A., J. A. Amstrom, J. E. Chin, K. G. Field, M. T. Ghiselin, D. J. Lane, G. J. Olsen, N. R. Pace, A. L. Parks, and E. C. R d . 1987. Molecular and developmentai correlates of macroevolution. pp. 109-138. In Development as an evolutwnaryprocess.IMBL Lect. Bwl. 8. New York: Alan R. Liss. Raff, R. A., and T. C. Kaufmann. 1983. Embryos,denes, and evolutwn. The deveupmental-genetic basis of evolutzonary change. New York: M a c d a n . Raines, D. E., S. E. Korten, and A. G. Hd. 1993. Anesthetic cutoff in cycloaikanemethanols: A test of current theories. Anestheswlogy 78 :918-927. Ramaswami, L. S. 1938. Connections of the pterygoquadrate in the tadpoles of Phtlautus vanabilts (Anura). Nature 192:577. . 1940. Some aspects of the chondrocranium of the South Indian frogs.J. Mysore Univ. 1:15-41. . 1941. Some aspects of the head of Xenupus M s . Proc. Indian Sci. Congr 28:183-184. . 1943. An account of the chondrocranium of Rana af ghana and Mgophrys, with a description of the masticatory musculature of some tadpoles. Pmc. Natl. Inst. Sci. Indza 9:43-58. . 1944. The chondrocranium of two torrent-dwelling anuran tadpoles. J.Morphol. 74:347-374. Rao, C. R. N. 1923. Notes on Batrachia. Rec. Indian Mus. 28:439447. Rappaport, R., Jr. 1955. The initiation of pronephric function in Ranapipiens.J. Exp. 2001. 128:481-487. Rastogi, R. K., L. Iela, M. Di Meglio, L. Di Matteo, S. Minucci, and I. Izzo-Vitiello. 1983. Initiation and kinetic proiles of spermatogenesis in the frog, Rana esculenta (Amphibia).J. Zool. 201:515-525. Rathke, H . 1832.Anatomisch-phiiusophZrche Untersuchungenuber den Kiemenapparat und d Zungenbein der Wirbeltiere. Riga Dorpat, E. Frantzens Buchhaundlung. Ratiba, R., C. Girard, and J.-D. Gipouloux. 1981. Mise en evidente de catcholamines dans Ia corde dorsaie de embryons d'an~phibiens anoures. Relations possibles aves l'activit adnylate-cyclasiquede cet organe et la migration de ceilules germinales primordiales. Arch. Anat. Micms. Molphol. Exp. 70: 149-160. Rau, R. E. 1978. The development of Xenopusgzlli Rose & Hewitt. Ann. South Ajkcan Mw. 76:247-263. Rauthner, M. 1967. Pharynx and epitheliaie Organe der Pharynxwand. 1.Kiemen der Anamnier-Kiemendarmderivate der Cyclostomen und Fische. pp. 21 1-278. In Handbuch der verglezchenden Anatomie der Wirbeltiere. Band 3, edited by L. Bolk, E. Goppert, E. Kallius, and W. Lubosch. Amsterdam: Asher. Rautio, S. A., E. A. Bura, K. A. Berven, and G. J. Gamboa. 1991. Kin recognition in wood frog tadpoles (Rana ylvatica)-factors affecting spatiai proximity to siblings. CanadianJ. Zool. 69:2569-2571. Razarihelisoa, M. 1969. Observations sur ies Batraciens aquatiques du centre des hauts plateaux de Madagascar. Verh.Intl. Vrein.Limnol. 17:949-955. . 1973a. Contribution a l'tude des Batraciens de Madagascar. Ecologie et dveloppement larvaire de Gephyromantis

rnethueni, Angel, Batracien biotope vgtai sur les Pandunus. Bull. Acad. Mabache 5 1:11 3-128. . 1973b. Contribution a I'tude des Batraciens de Madagascar. Ecologie et dveloppement larvaire de Mantidactylus brevipalmatus Ahl., Batracien de eaux courantes. Bull. Acad. Malgache 51:113-128. . 1979. Contribution l'tude biologique de quelgues Batraciens de Madagascar. Dissertation, University of Paris. Read, A. W. 1994. Treaunent of redleg in tadpoles of Pleurodeles. British Herpetol. Soc. Bu11. (49) :7. Readiig, C. J. 1990. Paimate newt predation on common frog, Rana temporaria, and common toad, Bufo bufo, tadpoles. Helpeto1.J. 1:462-465. Redshaw, M. R. 1972. The hormonai control of the amphibian ovary. American Zool. 12:289-306. Redshaw, M. R., and T. J. Nichoh. 1971. Oestrogen biosynthesis by ovarian tissue of the South African clawed toad, Xenupus laevii Daudin. Gen. Comp. Endocrinol. 16:85-96. Reed, S. C., and H. P. Stanley. 1972. Fine structure of spermatogenesis in the South African clawed toad X+ es laevii Daudin.J. Ultrastruct.Res. 41 :277-295. Regos, J., and A. Schluter. 1984. Erste Ergebnisse zur Fortpflanzungsbiologie von Lithodytes lineatus (Schneider 1799). Sahmandra 20:252-261. Reh, T. A., E. C. Pitts, and M. Constantine-Paton. 1983. The organization of the fibers in the optic nerve of normal and tectum-less Rana pipiens. J. Comp.Neuml. 218:282-296. Reiliy, S. M., and G. V. Lauder. 1990. Metamorphosis of craniai design in tiger salamanders (Ambystoma tgnnum): A morphometric anaiysis of ontogenetic change. J. Morphol. 204:121-137. Reinbach, W. 1939. Untersuchungen uber die Entwicklung des Kopfskeletts von Calyptocephalusgayi (mit einem Anhand uber das Os supratemporale der anuren Arnphibien). Zeit. Natumirs. (Jena)72:211-362. Reiss, J. 0. 1997. Early development of chondrocranium in the tailed frog Rrcaphus truei (Amphibia: Anura): Implications for anuran paiatoquadrate homologies. J. Morphol. 231:63-100. Rengel, D., and A. Pisan. 1981. I1 diierenziamento sessuaie della gonade di Phylhdusa sauvagii. Rend. Accad. Sci. Fis. Mat. 48:329-335. Reques, R., and M. Tejedo. 1995. Negative correlation between length of larvai period and metanlorphic size in natural populations of natterjack toads (Bufo calamita). J. Hevpetol. 29:311-3 14. Resetarits, W. J., Jr. 1998. Differential vulnerability of Hyla chrysoscelis eggs and hatchlings to larvai insect predation. J. Herpetol. 32:440443. Resetarits, W. J., Jr., and H. M. Wilbur. 1989. Choice of oviposition site by Hyla chrysoscelir: Role of predators and competitors. Ecology 70:220-228. Retallick, R. W., and J.-M. Hero. 1998. The tadpoles of Taudactylus eungellensisand ?: liemi and a key to the stream-dwelling tdpoles of the Eungella Rainforest in east-central Queensland, Austraiia.J. Helpetol. 32:304-309. Reynolds, W. W., and M. E. Casterlin. 1979. Behaviorai thermoregulation and the "final preferendum" paradigm. AmencanZool. 19:211-224. Rhodii, J. A. G., and A. Lametschwandmer. 1993. Circulatory pattern and structure in the tail and tail fins of Xenopus laevii tadpoles.J. Submimos. Cytol. Pathol. 25 :297-3 18. Rice, T. M., S. E. Waiker, B. J. Blackstone, and D. H. Taylor.

LITERATURE CITED 1998. A new method for marking individual anuran larvae. Herpetol. Rev. 29:92-93. Richards, C. M. 1958. The inhibition of growth in crowded Ranupipiens tadpoles. Physiol. 2001. 138-15 1. 31: Richards, S. J. 1992. The tadpole of the Austraiian frog Litmia nyakalensis (Anura, Hylidae), and a key to the torrent tadpoles of northern Queensland.Alpes 10:99-103. Richards, S. J., and C. M. B d . 1990a. Nonvisual detection of anuran tadpoles by odonate larvae.J.Herpetol. 24: 311-3 13. . 1990b. Size-limitedpredation on tadpoles of three Australian frogs. Copeia 1990:1041-1 046. Richmond, J. E. 1968. Changes in semm proteins during the development of bullfrogs (Rana catesbeiana) from tadpoles. Comp. Bwchem. Physiol. 24:991-996. Richmond, N. D. 1947. Life history of Scaphwpus holbrookii holbrookii (Harlan). Part I: Larval development and behavior. Ecology 28:53-67. Richter, S. 1989. The amphibian pronephros in interaction with the arterial and venous system during ontogenesis. pp. 8892. In Trenak in vertebrate mmphology.Progress in Zoology. Vol. 35, edited by H . Splechtna and H. Hilgers. Stuttgart: Gustav Fischer-Verlag. . 1992. Ontogenese des Exkretionssystems bei Rana esculenta (Ranidae,Anura). Histo- und morphogenetische Analyse der Anurenniere. Dissertation, University of Wien. . 1995. The opisthonephros of Rana esculenta (Anura). I. Nephron development.J. Morphol. 226: 173-187. Richter, S., and H. Splechtna. 1990. Morphogenesis of the Malphighian corpuscle in the anuran opisthonephros (Rana esculenta). p. 469. In Verhandlungen der Deutschen Zoologischen Gesellschafl, 83. Jahresvenammlung 1990, Frankfite am Main, edited by H . D. Pfannenstiel. Stuttgart: Gustav Fischer-Verlag. Ridewood, W. G. 1897a. On the stniaure and development of the hyobranchial skeleton and larynx in Xempus and Pipa; with remarks on the aKnities of the Aglossa. J. Linnean Soc. London 26:53-128. . 1897b. On the structure and development of the hyobranchial skeleton of the parsley-frog (Pelodytespunctatus). Soc. Proc. 2001. London 1897:577-595. . 1898a. On the development of the hyobranchial skeleton of the midwife-toad (Alytesobste~rt'cans). Proc. 2001. Soc. London 1898:4-12. . 1898b. On the larval hyobranchial skeleton of the anurous batrachians with special referente to the axial parts.J. Linnean Soc. London 26:474-487. Rieger, R. M. 1994. The biphasic life cycle- a central theme of metazoan evolution.American Zool. 34:484491. Riggs, A. 1951.The metatnorphosisof hemoglobin in the buiifr0g.J. Gen. Physwl. 35:2340. Riis, N. 1991. A field study of survival, growth, biomass, and temperature dependence of Rana dulmatinu and Rana tempmaria larvae.Amphibia-Reptilia 12:229-243. Rist, L., R. D. Semlitsch, H. Hotz, and H. U. Reyer. 1997. Feeding behaviour, food consumption, and growth efficiency of hemiclonal and parental tadpoles of the Rana esculenta complex. Funct. Ecol. 11:735-742. Ritke, M. E., and R. L. Mumme. 1993. Choice of calling sites and oviposition sites by gray treefrogs (Hyla chysoscelis)- a cornrnent. Ecology 74:623-626. Rivero, J. A. 1976 (1978). Notas sobre 10s anfibios de Venezuela. 11. Sobre 10s Colostethus de Los Andes venezolanos. Mem. Soc. Cien.Nat. Lu Salle 35 :327-344.

399

Roberts, A. 1976. Neuronal growth cones in an amphibian embryo. Brain Res. 118:526-530. Roberts, A., and S. T. Aiford. 1986. Descending projections and excitation during fictive swimming in X e q u s embryos: Neuroanatomy and lesion expenments. J. Comp. Neurol. 250:253-261. Roberts, A., and J. D. W Clarke. 1982. The neuroanatomy of an amphibian embryo spinal cord. Philosoph. Trans. Roy. Soc. London 296B:195-212. Roberts, A., and B. P. Hayes. 1977. The anatomy and function of the free nerve endings in an amphibian skin sensory system. Proc. Roy. Soc. London 196B:415429. Roberts, A., and D. T. Patton. 1985. Growth cones and the formation of central and peripheral neurites by sensory neurones in amphibian embryos.J. Neurosci. Res. 13:23-28. Roberts, A., S. R. Soffe, J. D. W. Clarke, and N. Dale. 1984. Initiation and control of swimming in amphibian embryos. pp. 261-284. In Neural onkin ofrhythmic muvements. SEB Symp. (37).New York: Cambridge University Press. Roberts, A., and J. S. H. Taylor. 1983. A snidy of the growth cones of developing embryonic sensory neurites.J. Embyol. Exp. Morphol. 75:3147. Roberts, J. D. 1981. Terrestrial breeding in the Austraiian leptodactylid frog Myobatrachusgouldii (Gray).Australian Wildl. Res. 8:451462. . 1984. Terrestrial egg deposition and direct development in Aremphlyne rotunda Tyler: A myobatrachid frog from coastal sand dunes at Shark Bay, Western Austraiia. Australia Wildl.Res. 11:191-200. Roberts, J. D., P. Honvitz, G. Wardell-Johnson, L. R. Maxson, and M. J. Mahony. 1997. Taxonomy, relationships, and conservation of a new genus and species of myobatrachid frog from the high rainfaii region of southwestern Austraiia. Copeia 1997:373-381. Robinson, D.C. 1961. The identie of the tadpole of Anutheca coronata. Copeia 1961:495. Rocek, Z. 1981 (1980). Cranial anatomy of frogs of the family Pelobatidae Stannius, 1856, with outlines of their phylogeny and systematics. Acta Univ. Carolinae B d . 1980: 1-164. . 1989. Developmental patterns of the ethmoidal region of the anuran skull. pp. 412420. In Trenak in vertebrate m r phology, edited by H. Splechma and H. Helgers. Stuttgart: Gustav Fischer-Verlag. . 1990. Ethmoidal endocranial structures in primitive tetrapods: Their bearing on the search for anuran ancestry. Zool.J. Linnean Soc. 99: 389407. Rocek, Z., and L .Veself. 1989. Development of the ethmoidal M structures of the endocranium in the anuran Pipa pipa. J. Morphol. 200:300-319. Rodel, M.-O. 1996. Amphibkn der westafnkanischen Savanne. Frankfurt: Edition Chimaira. . 1998. A reproductive mode so far unknown in African ranids: Phlynobatrahusguineensis Guib & Lamotte, 1961 breeds in tree holes (Anura: Ranidae). Hqetozoa 11: 19-26. Rodel, M.-O., and K. E. Linsenmair. 1997. Predator-induced swarms in the tadpoles of an African savanna frog, Phynomantis microps. Ethology 103:902-914. Root, R. B. 1967. The niche exploitation pattern of the Bluegray Gnatcatcher. Ecol. Monog~37:3 17-350. Rosel von Rosenhof, A. J. 1753-1758. Histma naturalis Ranurum nustratium. Niirenberg. Rose, M. J. 1982. Antagonistic pleiotropy, dominance, and genetic variation. Heredity 48:63-78.

400

LITERATURE CITED tions in &na pipiem, &na catesbeiana, and Xenupus mulhri with a note on lateral line and VI11 nerve projection wnes. Brain Behav. Evol. 14:368-380. Rubinson, K., and M. P. Skiles. 1975. Efferent projections of the superior olivary nucleus in the fiog, &na catesbeiana. Bvain Behav. Evol. 12:151-160. Ruby, J. R., R. F. Dyer, R. G. Skalko, and E. P. Volpe. 1970. Intercellular bridges between germ ceiis in the developing ovary of the tadpole, Ranapipiens. Anat. Rec. 167:l-10. Rugh, R. 1934. The space factor in the growth rate of tadpoles. Ecology 15:407411. . 1951.Thefrog.Its r e p v o d h n and dmelopment. Philadelphia: Blakiston Co. Ruibai, R. 1959. The ecology of a brackish water population of Ranapipiens. Copeia 1959:315-322. . 1962. Osmoregulation in amphibians from heterosaline habitas. Physwl. Zool. 35:133-147. Ruibai, R., and E. Thomas. 1988. The obligate carnivorous larvae of the frog, Lepidobatrachw iumi (Leptodactylidae). Copeia 1988:591-604. Ruiz Garcia, F. N. 1980. La larva de BufO longinasw (Stejneger) (Amphibia:Bufonidae) en e1laboratorio. Poeyana (207):1-8. Rusconi, M. 1826. Dheloppment de hlagrenouilh commune. Paris. Russeii, A. P., and A. M. Bauer. 1993. The amphibians and reptihs ofAlberta. Caigary, Alberta: University of Caigary Press and University of Alberta Press. Russeii, I. J. 1976. Amphibian lateral line receptors. pp. 513550. In F r y neurobwlogy, a handbook, edited by R. Llins and W Precht. Beriin: Springer-Verlag. Ruthven, A. G. 1915. Observations on the habits, e&s, and young of Hyh&hmtanni Peracca. O-. Pap. Mw. Zool., Univ.Michigan (14):14. Ruthven, A. G.,, and H. T. Gaige. 1915. The breeding habits of Prostherapis subpunctatus Cope. Occas. Pap. Mus. Zool., Univ. Michigan (10):1-17. Ryan, M. J. 1986. Synchronized caing in a treefrog (Smilisca sih).Brain Behav. Evol. 29: 196-206. . 1988. Constraints and pattems in the evolution of anuran acoustic communications. pp. 637-677. In The mlutMn of the amphibian auditory +em, edited by B. Fritzsch, M. J. Ryan, W Wilczynski, T. E. Hetherington, and W Waikowiak. New York: Wiley. Ryke, P. A. J. 1953. The ontogenetic development of the somatic musculature of the trunk of the aglossai anuran Xenopw laepis (Daudin).Acta Zool. 34: 1-70. Sabater-Pi, J. 1985. Contribution to the biology of the giant frog (Cmraua goliath, Boulenger). Amphibia-Reptilia 6: 143-153. Saber, P. A., and W. A. Dunson. 1978. Toxicity of bog water to embryonic and larval anuran amphibians. J. &p. Bwl. 204:3342. Sabnis, J. H., and K. S. M. Kuthe. i980. Observations on food and growth ofBufOmelanostzstz~s tadpo1e.J. Bombay Nat. H&. SOL. 77:21-25. Sahu, A. K., and M. K. Khare. 1988. Food and feeding habits of Rana alticoh Boulenger (Anura: Ranidae) during different stages of metamorphosis.J. Adv. Zool. 9:97-104. Saidapur, S. K. 1983. Patterns of testicular aaivity in Indian amphibians. I n d i a n h . Life Sci. 3: 157-184. 1 9 8 6 . Patterns of ovarian aaivity in Indian amphibians. Indian Rev. Life Sci. 6:231-256. Saidapur, S. K., and V B. Nadkarni. 1974. Steroid-synthesizing celular sites in amphibian ovary. A histochemical study. Gen. Comp.Endocrinol. 22:459-462.

Rose, S. M. 1959. Failure of survival of slowly growing members of a population. Science 129:1026. . 1960. A feedback mechanism of growth control in tadpoles. Ecology 41: 188-199. Rosen, D. E., P. L. Forey, B. G. Gardiner, and C. Patterson. 1981. Lungfishes, tetrapods, paieontology, and plesiomorphology. Bull.AmevKan Mw. Nat. Hist. 167:163-275. Rosenberg, E. A., and B. A. Pierce. 1995. Effect of initiai mass on growth and mortality at low pH in tadpoles of PseudanW chrkii and Bufo valliceps.J. Herpetol. 29: 181-185. Rosenthal, B. M., and W L. R. Cruce. 1985. The dendritic extent of motoneurons in frog brachiai spinai cord: A computer reconstruction of HRP-iiied ceh. Brain Behav. Evol. 27: 106-114. Rossi, A. 1959. Tavole cronologiche deiio sviluppo embrionale e larvai de1BufO bufo (L.). Monit. Zwl. Italiano 66: 1-11. Rostand, J. 1934. Toadr and toa life. Translated by J. Fletcher. London: Methuen. Roth, A. H., and J. F. Jackson. 1987. The effect of pool size on recruiunent of predatory insects and on mortality in a larvai anuran. Herpetologica 43:224-232. Roth, G. 1987. Visual behavior in sahmanders. Berlin: Springer-Verlag. Rouf, M. A. 1969. Hematology of the leopard frog, Ranapipiem. Copeia 1969:682-687. Roux, W 1888. Conmbutions to the developmental mechanisms of the embryo. On the artificial production of halfembryos by destruction of one of the first two blastomeres, and the later development (postgeneration of the missing half of the body). Vimhows Arch. Path. Anat. Physwl. Klin. Med. 114:113-153. [English translation in B. H. Willier and J. M. Oppenheimer. 1974. Foundatwns ofexpm'wntal embryology. 2d ed. New York: Hafner Press.] Rovira, J., A. C. Viaro, and M. E. Bodegas. 1993. Structural study of the fiog Rana temporaria larvai stomach. Tiss. Cell 25:695-707. Rowe, C. L., and W A. Dunson. 1995. Impacts of hydroperiod on growth and survivai of larvai amphibians in temporary ponds of central Pennsylvania. Oecologia 102:397403. Rowe, C. L., O. M. Kinney, A. P. Fiori, and J. D. Congdon. 1996. Oral deforrnities in tadpoles (Ranacatesbeiana) associated with coai ash deposition: Effects of grazing ability and growth. Freshw. Bwl. 36:723-730. . 1998a. Oral deformities in tadpoles of the bulifrog (Rana catesbeiana) caused by conditions in a poluted habitat. Cupeia 1998:244-246. Rowe, C. L., O. M. Kinney, R. D. Nagle, and J. D. Congdon. 1998b. Elevated maintenance costs in an anuran (Ranacatesbeiana) exposed to a rnixture of trace elements during the embryonic and early larval periods. Physwl. Zool. 71:27-35. Rowe, C. L., W. J. Sadinski, and W A. Dunson. 1994. Predation on larvai and embryonic amphibians by acid-tolerant caddisfly larvae (Ptilostomispostica).J. Herpetol. 28:357-364. Rowe, L., and D. Ludwig. 1991. Size and timing of metamorphosis in complex life cycles: Time constraints and variation. Ecology 72:413427. Rowedder, W 1937. Die Entwicklung des Geruchsorgans bei Alpes obstemiam und Bufo vubaris. Zeit. Anat. Enhvickl. 107:91-123. Rubin, D. I., and L. M. Mendell. 1980. Location of motoneurons supplying individual muscles in norrnai and grafied supemumerary limbs of Xenopus i m. J. Comp. Neurol. u i 192:703-715. Rubinson, K., and B. Friedman. 1977. Vagal derent projec-

LITERATURE CITED Saint-Ange, M. 1831. Recherches anatomiques et physiologiques sur les organes transitoires et la metamorphose des batraciens.Ann. Sci. Nat. 34. Salthe, S. N., and W E. D u e h a n . 1973. Quantitative constraints associated with reproductive rnodes in anurans. pp. 229-249. In Evolutwnaty bwlagy of the anurans. Contemporaty reseavch on majm p b l e m , edited by J. L. Vial. Columbia: University of Missouri Press. Salthe, S. N., and J. S. Mecham. 1974. Reproductive and courtship patterns. pp. 309-35 1. In Physwlogy of the Amphibia. Vol. 2, edited by B. Lofts. New York: Academic Press. Salvador, A. 1985. a i a de campo de hs anpwsy reptiiesde h Peninsula Iberica, Ish Baleares y Canarais. Madrid: Unigraf. Salvatorelli, G. 1970. Osservazioni citornetriche sulgi eritrociti di Bufo bufo durante il periodo larvale e adulto. Ann. Univ. Fewara (SmAnat. Comp., Sezwne 13) 3: 17-27. Sarnpson, L. i? 1900. Unusual rnodes of breeding and developrnent among Anura. Amrican Nat. 34:687-715. . 1904. A contribution to the ernbryology of Hyhdes martiniumsiS.A m ' c a n J Anat. 3:473-504. Sarikas, S. N. 1977. Histochemistry of egg membrana and thickness of egg rnernbranes related to sperrn dimensions in two species of salarnander of the genus Ambystomu. Thesis, Southem Illinois University. Sarnat, H. B., and M. G. Netsky. 1981. Evolution of the nmous *em. 2d ed. Oxford: Oxford University Press. Saruwatari, T., J. A. Lbpez, and T. W. Pietsch. 1997. Cyanine Blue: A versatile and harrnless stain for specimen observation. Copeia 1997:840-841. Sasaki, F., and T. Kinoshita. 1994. Adult precursor cells in the tail epiderrnis of Xenupus tadpoles. Histochemistry 101:391-396. Sasaki, H., and H. Mannen. 1981. Morphological analysis of astrocytes in the bullfrog (Ranacatesbeiana) spinal cord with special reference to the site of attachment of their processes. J. Comp.Neurol. 198:13-35. Satel, S. L., and R. J. Wassersug. 1981. On the relative sizes of buccal floor depressor and elevator rnusculature in tadpoles. Copeia 1981:129-137. Sater, A. K., and A. G. Jacobson. 1989. The specification of heart rnesodem occurs during gastrulation in Xenupus laet.is. Devehpment 105321-830. 1 9 9 0 . The role of the dorsal iip in the induction of heart rnesoderrn i Xenupus W . n Devehpment 108:461470. Sato, K . 1924. ber die Metamorphose von Bufo vubaris. Zeit. Anat. EntwickL. 71:41-184. Savage, J. M. 1960. Geographic variation in the tadpole of the toad Bufo marinus. Gpeia 1960:233-235. . 1968. The dendrobatid frogs of Central America. Copeia 1968:745-776. . 1980a. A synopsis of the larvae of Costa Rican frogs and toads. Bull. Southem CalzfmniaAd. Sci. 79:45-54. . 1980b. The tadpole of the Costa Rican fringe-limbed tree-frog, Hyh jmbrimembra. Prac. Bwl. Soc. Washington 93:1177-1183. Savage, R M. 1935. The ecology of young tadpoles, with specific reference to some adaptations to the habit of rnass spawning in Rana temparia tempmaria L. Proc. 2001. Soc. London 1935:605-610. . 1952. Ecological, physiological, and anatomical observations of some species of anuran tadpoles. Proc. Zool. Soc. London 122:467-514. . 1955. The ingestive, digestive and respiratory systerns of the microhylid tadpole, Hypopachus aquae. Copeia 1955:120-127.

40 1

. 1962. The ecology and life histoty ofthe w m m nfrog. New York: Hafner Publ. Co. Saxn, L. 1987. Organugenesis ofthe kidney. Cambridge: Carnbridge University Press. Sazima, I., and W C. A. Bokerrnann. 1978. Cinco novas espc. ies de Leptoaktylus do centro e sudeste Brasileiro (Arnphibia, Anura, Leptodactylidae). Rerz Brasil. Bwl. 38:899-912. . 1982. An6bios da Serra do Cip, Minas Gerais, Brasil 5: Hyiodes otavioi sp. n. (Anura, Leptodactyiidae).Rev. Brasil. Bwl. 42: 767-771. Scadding, S. R. 1993. Viaoria Blue B staining for cartilage. p. 20. In Handbook of methodr. Intemutwnd workrhop on the mlecular bwiogy of axoiotIs and other urodeles. Indianapolis, IN. Scalia, F. 1976. Structure of the olfactory and accessory olfactory systerns. pp. 213-233. In Frog neurobwlogy, a handbook, edited by R. Llins and W Precht. Berlin: Springer-Verlag. . Scalia, F.,, G. Gallousis, and S. Roca. 1991a. A note on the organization of the amphibian olfactory bulb. J Cmp. Neurol. 305:435442. . 1991b. Differential projections of the rnah and accessory olfactory bulb in the frog. J. Comp. Neuml. 305:443461. Scalia, F. M. Halpem, H. Knapp, and W Riss. 1968. The efferent comexions of the olfactory bulb in the frog: A study of degenerating unmyeiinated fibers.J Anat. 103:245-262. Schechtman, A. M. 1934. Unipolar ingression in Tritums tmosu, a hitherto undescribed rnovernent in the pregastrular stages of a urodele. Univ. Califomia Publ. Zool. 39:303-310. Scheel, J. J. 1970. Notes on the biology of the African tree-toad, Nectq%yne afia Buchholz & Peters, 1875 (Bufonidae, Anura) frorn Fernando Poo. Rev. Zool. Bot. Aficaines 81:225-236. Schijfsrna, K. 1932. Notes on some tadpoles, toads, and frogs frorn Java. Tmebia 14:43-72. Schi~tz, 1963. The amphibians of Nigeria. Vidensk.Medd. A. Dansk Naturh. Foren. 125:1-125,4 plates. . 1967. The treSrygs (Rhaaphmz'dae)of westANa. Copenhagen: Bianco Lunos Bogtrykkeri. . 1975. The treefigs of edctemAfica. Copenhagen: Steenstrupia. Schlaepfer, M. A. 1998. Use of a flourescent rnarking technique on srnall terrestrial anurans. Herpetol. Rev. 29:25-26. Schley, L., R. A. Griffiths, and A. Rornn. 1998. Activitfpattems and microhabitat selection of Mallorcan midwife toad (Alytes muletensis) tadpoles in natural torrent pools. Amphibia-Reptilia 19:143-151. Schiosser, G., and G. Roth. 1995. Distribution of cranial and rostral spinal nerves in tadpoles of the frog Discglossuspictus (Discoglossidae).J Morphol. 226: 189-212. . 1997a. Evolution of nerve developrnentin frogs. I. The developrnent of the peripheral nervous system in Discoglosus p i a s (Discoglossidae).Brain Behav. Evol. 50:61-93. . 199%. Evolution of nerve developrnent in frogs. 11. Modified developrnent of the peripheral nervous systern in the direct-developing fiog Elactherodactylus coqui (Leptodactyiidae).Brain Behav. Evol. 50:94-128. Schluga, A. 1974. Beitrag zur Entwicklung des Exkretionssysterns bei Arnphibien. Dissertation, University of Wien. Schlter, A. 1990. Reproduction and tadpole of Edalorhinaperezi (Amphibia, Leptodactylidae). Stud. Neotrq. Fauna Environ. 25:49-56. Schlter, A., and A. W Salas. 1991. Reproduction, tadpoles, . and ecological aspects of three syntopic rnicrohylid species

402

LITERATURE CITED

from Peru (Amphibia: Microhylidae). Stutt~arter Beitr: Naanuran larvae: Feeding regulation inXenupus and Rana. Bwl. Bull. 162:214-231. turk. (458):l-17. Schmalhausen, I. I. 1968. The origin of terrestrial vertebrates. . 1987. Amphibia. pp. 467-552. In Animal energetics. New York: Academic Press. Vol. 2, edited by T. J. Pandian and E. J. Vernberg. New York: Academic Press. Schmidt, B. R. 1993. Are hybridogenetic frogs cyciical parthogens? Trenak Eml. Evol. 8:271-272. Seale, D. B., and N. Beckvar. 1980. The comparative abiiity of Schmidt-Nielsen, B., and W. C. Mackay. 1970. Osmotic and anuran larvae (genera: Hyiu, Bu., and Rana) to ingest susdiffisional water permeability in metamorphosing Rana pended blue-green algae. Copeia 1980:495-503. clamitans tadpoles. Bull. MountDesertIs. Bwl. Lab. 10:71-75. Seale, D. B., K. HoK and R. Wassersug. 1982. Xenopus laeYis Schmidt-Nielsen, K. 1983.Animalphyswkyy:Aahptath and enlarvae (Amphibia, Anura) as model suspension feeders. Hydrobwigia 87:161-169. vironment. London: Cambridge University Press. Schmuck, R., W. Geise, and K. E. Linsenmair. 1994. Life cycle Seale, D. B., and R. J. Wassersug. 1979. Suspension feeding dynamics of anuran larvae related to their functional morpholstrategies and physiological adjustments of reedfrog tadpoles ogy. Oeco[ogia 39:259-272. (Amphibia,Anura, Hyperoliidae) in relation to environmental conditions. Copeia 1994:996-1007. Sedra, S. N. 1950. The metamorphosis of the jaws and their muscles in the toad, B u . re~uiuris Reuss, correlated with Schmutzer, W. 1977. HistopathologischeVeranderungen durch Nephrotrema twncatum (Leuckart 1942)(Digenea: TrogloSoc. changes in the animal's feeding habits. Proc. 2001. Lontrematidae) in der Schwanzregion von Froshlawen (Rana don 120:405449. . 1951. On the morphology of the suprarostral system temporaria L.). Zool.Anz. 198:355-368. Schneider, L. 1957. Beidussung des Wachstums durch im and the mandibular arch of B u . ient&inusus. P m . Egyptian Acad. Scz. 7: 128. Thymus enthaltene Stoffe. I. Hersteliung fraktionierter Thymusextrakte und ihre Prfung im Kaulquappenversuch.Wil- Sedra, S. N., and M. I. Michael. 1956a. The development of the skeleton and muscles of the head from stage 55 t i l the helm Roux' Arch. Entmz'ckiungsmech. Org. 149:644-683. Schoonbee, H. J., J. F. Prinsloo, and J. G. Nxiweni. 1992. Obend of metamorphosis. pp. 55-68. In Nomzal table of Xenosewations on the feeding habits of larvae, juvenile, and adult pus laevis Daudin, edited by E D. Nieuwkoop and J. Faber. stages of the Afncan clawed frog, Xenopus iuevis, in imAmsterdam: North-Holiand. poundments in Transkei. WaterSouth Afica 18:227-236. . 1956b. The stnicture of the hyobranchial apparatus of Schorr, M. S., R. Altig, and W. J. Diehl. 1991. Populational the W y developed tadpole lama of B u . regulan5 Reuss. changes of the enteric protozoans Opalina spp. and 2VvctothProc. Egyptian Acad. Sci. 12:3 8 4 6 . ents mrdifomzis during the ontogeny of anuran tadpoles. J. . 1957. The development of the skull, viscera aches, larProtowol. 37:479-481. ynx, and visceral muscles in the South African clawed toad Schulze, F. E. 1888. ber die inneren kiemen der BatrachierXenupw laevis (Daudin) during the process of metamorpholawen. I. Mittheilung. ber das epithel der Lippen, der sis (from stage 55 to stage 66). E&. Knink. NederlandAkad. Mund-, Rachen- und Kiemenhohle erwachsener larven von Wetensch.Natuurk. 5 1:1-80. Peiobatesfiscus. Abh. Preuss. Akad. Wts. Berlin 1888:l-59, 1 9 5 8 . The metamorphosis and growth of the hyobran4 plates. chial apparatus of the Egyptian toad, B u . reguaris Reuss.1 . 1892. ber die inneren Kiemen der Batrachierlarven. Molphol. 103:l-30. 11. Mitteilung. Skelet, Musculatur, Blutgefasse, Filterap. 1959. The ontogenesis of the sound conducting appaparat, respiratorische Anhiinge und Athmungsbewegungen ratus of the Egyptian toad, Buf regularis Reuss, with a reerwachsener Lawen von Peiubatesfiscus. Abh. Preuss. Akad. view of this apparatus in Salientia.J. Morphol. 104:359-375. Wzss. 1892:1-66. . 1961. Normal table of the Egyptian toad, B u . regularis Schtte, F., and A. Ehrl. 1987. Zur Haltung und Zucht der Reuss, with an addendum on the standardization of the gro3en sdamerikanischen Wdenkrote Pipapipa (Linnaeus stages considered in previous publications. Ceskoshenska 1758). Salamandra 23:256-268. Moflol. 9:333-351. Schtte, M., and S. G. Hoskins. 1993. Ipsilaterally projecting Sedra, S. N., and A. A. Moursi. 1958. The ontogenesis of the retinal gangiion celis in Xennpus 15 -: An HRP study. J. vertebral column of B u . reguiuZ( Reuss. Mo@io~ie 7-32. 6: Comp. Neurol. 331:482-494. Seigel, R. A. 1983. Natural suwival of eggs and tadpoles of the wood frog, Rana sylvatica. Copeia 1983:1096-1098. Schwartz, A., and R. W. Henderson. 1991.Amphibians and reptiles of the West Indies. Descriptwns, di&butwns, and natzwal Sekar, A. G. 1992. A study of the food habits of six anuran tadhistovy. Gainesville: University of Florida Press. poles. J.Bombay Nat. Hist. SOL.89:9-16. Schwartz, J. J., and K. D. Weiis. 1983. An experimental study Selenka, E. 1882. Der embryonale excretionsapparat des kiemof acoustic interferente between two species of neotropical losen Hyiodes martinicensis. Sztzungsber: Kon. Preuss. Aka. treefrogs. Anim. Behav. 31: 181-190. Wiss.Berlrlin 117-124. Scott, N.J., Jr., and R. D. Jennings. 1985. The tadpoles of five Semiitsch, R. D. 1990. Effects of body size, sibship, and tail species of New Mexican leopard frogs. Occas. Pap. Mus. injury on the susceptibility of tadpoles to dragonfly predaSouthw. Bwl. (3):1-21. tion. CanudianJ.2001. 1027-1030. 68: Scott-Birabn, M. T., and K. Fernndez-Marcinowski. 1921. . 1993. Asymmetric competition in mixed populations Variaciones locales de caracteres especficos en larvas de anof tadpoles of the hybridogenetic Rana esculenta complex. fibios.Ann. Soc. Cient.Argentina 92: 129-144. Evoluth 47:510-519. Seale, D. B. 1980. Influente of amphibian lawae on primar- Semlitsch, R. D., and J. E Caldweli. 1982. Effects of density production, nutrient flux, and competition in a pond ecosyson growth, metamorphosis, and suwivorship in tadpoles of tem. Ecolqy 61:1531-1552. Scaphiqus holbrooki. Ecoigy 63:905-911. . 1982. Obligate and facultative suspension feeding in Semlitsch, R. D. and S. Gavasso. 1992. Behavioural responses

LITERATURE CITED of Bufo bufo and Bufo caiumita tadpoles to chemical cues of vertebrate and invertebrate predators. Ethol. Ecol. Evol. 4: 165-173. Semiitsch, R. D., and J. W. Gibbons. 1988. Fish predation in size-structured popuiations of treefrog tadpoles. Oewlogia 75:321-326. Semlitsch, R. D. and H.-U. Reyer. 1992a. Modification of antipredator behaviour in tadpoles by environmental conditioning.J Anim. Ecol. 61:353-360. . 1992b. Performance of tadpoles from the hybridogenetic Rana escuenta complex: Interactions with pond drying and interspeciic competition. EvolutMn 46:665-676. Semiitsch, R. D., and S. Schmiedehausen. 1994. Parenta1 contribution to variation in hatchling size and its relationship to growth and metamorphosis in tadpoles of Rana bssonae and Rana escubnta. Copeia 1994:406-412. Semiitsch, R. D., D. E. Scott, and J. H . K. Pechmann. 1988. Time and size at metamorphosis related to aduit fitness in Ambystoma talpoideum. Ecology 69: 184-192. Semlitsch, R. D., D. E. Scott, J. H. K. Pechmann, and J. W. Gibbons. 1993. Phenotypic variation in the arrivai time of breeding salamanders: Individual repeatability and environmental influentes. J Anim. Ecol. 62:334-340. . 1996. Structure and dynamics of an amphibian community. Evidence from a 16-yearstudy of a natural pond. pp. 217-248 h Long-temz studks of vertebrate communities. New York: Academic Press. Semiitsch, R. D., and H. M. Wiibur. 1989. Artificial selection for paedomorphosis in the salamander Ambystm talpoideum. Evolutwn 43: 105-1 12. Sem, D. G. 1972. Development of tegmental and rhombencephalic suuctures in a frog (Rana temporaria L.). Acta Anut. 82:525-548. Serra, J. A., and R. M. Albuquerque. 1963. Anibios de Portugal. Rev. Portuguesa Zool. Bwl. Geral 4:75-227. Sesama, I?, J. Rovira, A. C. Viaro, M. E. Bodegas, and E. Valverde. 1995. Metamorphic changes in the stomach of the frog Rana temporaria tadpoles. Tiss. Cell27:13-22. Sever, D. M. 1994. Comparative anatomy and phylogeny of the cloacae of salamanders (Amphibia: Caudata). VII. Plethodontidae. Herpetol. Monogr (8):276-33 7. Severtsov, A. S. 1969a. Food-seizing mechanism in anuran larvae. Dokl. Akad. Nauk SSSR 187:211-214. . 1969b. Origin of the basal elements of the hyobranchial skeleton of larvae of amphibians. Dokl. Aka. Nauk SSSR 187:677-680. Sexton, O. J., and C. Phillips. 1986. A qualitative study of fishamphibian interactions in 3 Missouri ponds. Trans.Missou~ Acad. Sci. 20:25-35. Seymour, R. S. 1973. Gas exchange in spadefoot toads beneath the ground. Copeia 1973:452460. Seymour, R. S., and D. F. Bradford. 1995. Respiration of amphibian eggs. Physwl. Zool. 68: 1-25. Seymour, R. S., F. Geiser, and D. F. Bradford. 1991a. Gas conductance of the jeliy capsuies of terrestriai frog eggs correlates with embryonic stage, not metaboiic demand or ambient PO,, Physwl. Zool. 64:673-687. . 1991b. Metabolic costs of development in terrestrial frog eggs (Pseuduphlynebibronii).Physwl. Zool. 64:688-690. Seyrnour, R. S., M. J. Mahony, and R Knowles. 1995. Respiration of embryos and larvae of the terrestriaiiy breeding frog Kyawanus heridgei. Herpetologica 5 1:369-3 79. Seymour, R. S., and J. D. Roberts. 1991. Embryonic respira-

403

tion and oxygen distribution in foamy and nonfoamy egg masses of the frog Limnodpastes tasmizniensis. Physwl. Zool. 64:1322-1340. Shafier, H. B. 1984. Evolution in a paedomorphic iineage. 11. Aometry and form in the Mexican ambystomatid salamander. Evolutwn 38: 1207-1218. Shaffer, H. B., R. A. Aiford, B. D. Woodward, S. J. Richards, R. G. Altig, and C. Gascon. 1994. Quantitative sampiing of amphibian larvae. pp. 130-141. In Measu~ng munitorand ing biologcal diversity. Standard mthoh for amphibians. Biological Diversity Handbook Series, edited by W. R. Heyer, M. A. Domeliy, R. W. ~McDiarmid, L.-A. C. Hayek, and M. S. Foster. Washington, D.C.: Smithsonian Institution Press. Shafier, L. R., and D. R. Formanowicz, Jr. 1966. A cost of viviparity and parental care in scorpions: Reduced sprint speed and behavioural compensation.Anim. Behav. 5 1:1017-1024. Shaw, E. 1978. Schooiing fishes. American Sci. 66: 166-175. Shaw, J. P. 1979. The time scale of tooth development and replacement inxenupus l d s (Daudin).J.Anat. 129:323-342. Shelton, P. M. J. 1970. The lateral line system at metamorphosis in Xenopus iuevis (Daudin). J. Embryol. Exp. Morphol. 24:511-524. . 1971. The structure and fimction of the lateral line system in larval Xenopus W 5 . J .Exp. Zool. 178:211-232. Sheratt, T. N., and I. F. Harvey 1989. Predation by larvae of Pantala flavescens (Odonata) on tadpoles of Phyllomdusa trinitatis and Physalaemuspustulosus: The influente of absolute and relative density of prey on predator choice. Oikos 56:170-176. Sherman, E. 1980. Ontogenetic change in thermal tolerante of the toad Bufo woodhousiz f m b ~ .Comp. Bwchem. Physiol. 65A:227-230. Shield, J. W., and l? J. Bentley. 1973. Respiration of some urodele and anuran Amphibia. I. In water, role of the skin and giiis. Comp. Bwchem. Physiol. 46A: 17-28. Shine, R. 1995. A new hypothesis for the evolution of viviparity in reptiles. American Nat. 145:809-823. Shirane, T. 1970. On the formation of germ celis of the frog's larvae from eggs irradiated with ultraviolet rays. Mem. Fm. Gen. Edw. Hiroshima Univ.1 1 4124-37. 1. 1 9 7 2 . On the formation of primordial celis of the Japanese pond frog, Rana brevipodu Ito. Mem. Fm. Gen. Edw. Hiroshima Unzv.1 1 5: 17-27. 1. . 1982. Gonadal differentiation in frogs, Rana japonica and Rana brevipodu, raised from UV irradiated eggs. J. %. Zool. 223: 165-173. . 1986. A new, early, morphological indication of sex differentiation in Anura, Rana japonica, and Rana nlgromacu1ata.J. Exp. Zool. 240: 113-1 18. Shiriaev, B. I., and 0 . V Shupiiakov. 1986. Synaptic organization of dorsal root projections to lumbar motoneurons in the clawed toad (Xenopuslaevis).&p. BBrainRes. 63: 135-142. Shoemaker, V H., and L. L. McClanahan. 1973. Nitrogen excretion in the larvae of a land-nesting frog (Leptodactylusbufonius). Comp.Bwchem. Physiol. 44A: 1149-1 156. . 1982. Enzymatic correlates and ontogeny of uricotelism in tree frogs of the genus Phyllumedusa. J. Exp. Zool. 220: 163-169. Shofner, W. l?, and A. S. Feng. 1984. Quantitauve light and scanning clectron microscopic study of the developing auditory organs in the bulfrog: Implications on their fimctional characteristics.J. Comp.Neurol. 224: 141-154.

404

LITERATURE CITED morphology and retarded development in Bombina mientalis (Anura, Discoglossidae). 2001. Anz. 225 :324-332. . 1992. The influente of variation in lamal period on adult cranial diversity in Pelobatesfiscus (Anura: Pelobatidae). J. 2001. 226:601-612. . 1994. Postmaturation skuil development in Xenopus laeois (Anura, Pipidae): Late-appearing bones and their bearing on the pipid ancestral morphology. RmsianJ. Herpetol. 1:21-29. . 1995. Extra bones in the Pelobates skull as evidence of the paedomorpic origin of the anurans. Zhul: 06s. Bwl. 56:317-328. . 1997. Additionai dermal ossifications in the anuran skuil: Morphological novelties or archaic elements?hsstan J.Hqetol. 4: 17-27. Smith, C. L., and E. Frank. 1988a. Peripheral specification of sensocy connections in the spinal cord. Brain Behav. Evol. 31:227-242. . 1988b. Specificity of sensocy projeaions to the spinal cord during development in bullfrogs. J. Comp. Neurol. 269:96-108. Smith, D.C. 1983. Factors controlling tadpole populations of the c h o m frog (Pseuhmiswismiata) on Isle Royale, Michigan. Ecology 64:501-510. . 1986. Enhanced growth in fuli sib populations of c h o m frog tadpoles. American 2001. 26:s. (Abstr.) . 1987. Addt recruitment in c h o m frogs: Effects of size and date at metamorphosis. Ecology 68:344350. 1 9 9 0 . Popdation stmcture and competition amang kin in the c h o m frog (Pseu&s triseriata).Evolutwn 44: 15291541. Smith, D.C., and J. Van Buskirk. 1995. Phenotypic design, plasticity, and ecological performance in two tadpole species. American Nat. 145:211-233. Smith, L. D. 1965. Transplantation of the nuclei of primordial germ celis into enucleated eggs of Rana pipiens. Proc. Natl. Acad. Sci. United States ofAmerica 54: 101-107. . 1966. The role of a germinal plasm in the formation of primordial germ celis in Rana pipiens. Develup. Bwl. 14:330-347. Smith, L. D., and M. Williams. 1979. Germinai plasm and germ celi determinants in anuran amphibians. pp. 167-197. In Proc. 4th Symp. British Soc. Develup. Bwl., edited by D. R Newth and M. Bails. Cambridge: Cambridge University Press. Smith, M. A. 1916a. Descriptions of five tadpoles from Siam. J.Nat. H h . SOL.Siam 2:3743. 1 916b. On the frogs of the genu Oxy~losssis. Nat. H&. J SOL.Szam 2: 172-1 76. . 1917. On tadpoles from S i m . J. Nat. Hist. Soc. Siam. 2:261-275. . 1930. The Reptilia and Amphibia of the Malay Peninsda. Bull. Ra@s Mus. (3):1-149. Smith, R. J. F. 1977. Chemical communication as an adaptation: Alarm substances of fish. pp. 303-320. In Chemicals~nals in vertebrates, edited by D. Muller-Schwarze and M. M. Mozeii. New York: Plenum. Smith, S. C., M. J. Lannoo, and J. B. Armstrong. 1988. Lateraliine neuromast development in Ambyrtoma rnexicanum and a companson with Rana ppiens. J. Morphol. 198:367-379. . 1990. Development of the mechanoreceptive lateraliine system in the axolotl: Placode specification, guidance of migration, and the origin of neuromast polarity. Anat. Embryol. 182:171-180.

Shumway, W 1940. Stages in the normal development of Rana pipiens. I . Externalfomz. Anat. Rec. 78: 139-147. Shvarts, S. S., and O. A. Pyastolova. 1970. Regulators of growth and development of amphibian larvae. I. Specificity of effects. Ekologiya 1:58-62. Siegfried, W. R., and L. G. Underhill. 1975. Flocking as an antipredator strategy in doves. Anim. Behav. 23:504-508. Siekrnann, J. M. 1949. A survey of the tadpoles of Louisiana. Dissertation, Tulane University. Sih, A., P. Crowley, M. McPeek, J. Petranka, and K. Strohmeier. 1985. Predation, competition, and prey communities: Areview of field experiments. Ann. Rev. Ecol. Syst. 16:269-311. Siiver, M. L. 1942. The motoneurons of the spinal cord of the fr0g.J. C~mp. Neurol. 77: 1 4 0 . Silverstone, P. A. 1975. A revision of the poison-arrowfrogs of the genus Dendrobates Wagler. Sci. Bull., Nat. Hist. Mus. Los An~eles 2 1:1-5 5. Co. Sirnnett, J. D., and D. P. Chopra. 1969. Organ specific inhibitor of mitosis in the amphibian kidney. Nature 222: 1189-1 190. Simon, M. P. 1983. The ecology of parenta1 care in a terrestrial breeding fcog from New Guinea. Behav. Ecol. Socwbwl. 14:61-67. Simpson, H. B., M. L. Tobias, and D. B. Kelley. 1986. Origin and identification of fibers in the cranial nerve E - X complex of &nopus W :Lucifer yeiiow backfis in vitro. J. Comp. Neurol. 244:430+. Sin, G., and L. Gavrila. 1977. Study of the diet of Rana ridibunda tadpoles. Stud. Cercet.Bwl. 29:93-98. Sinsch, U. 1990. Froschiurche (Anura) der z e n d pemanischen Anden: Artdiagnose, Taxonomie, Habitate, Verhaitensokologie. Salamandra 26: 177-214. . 1997. Effeas of larval histocy and microtags on growth and survival of nattejack (Bujhcalamita) metamorphs. Hmpeto1.J. 7: 163-168. Sive, H., and L. Bradley. 1996. A sticky problem: TheXenupus cement gland as a paradigm for anteroposterior patterning. Develup. Bwl. 205:265-280. Siwila, J. C., M. C. Mix, and D. S. McKenzie. 1972. Oxygen consumption of Bujh bmeas bmeas tadpoles dunng variou developmental stages of metamorphosis. Heptologca 28:309-313. Skeliy, D. K. 1990. Behavioral and life-historical responses of larval American toads to an odonate predator. Ewlogy 71:2313-2322. . 1991. Field evidence for a cost of behavioral antipredator response in a larvai amphibian. Ewlogy 73:704-708. . 1995a. A behavioral trade-off and its consequences for the distribution of Pseu&s treefrog lamae. Ecology 76:150-164. . 1995b. Competition and the distribution of spring peeper lamae. Oecolo~ia 103:203-207. . 1996. Pond drying, predators, and the distribution of Pseuducris tadpoles. Copeia 1996599-605. Skeliy, D. K., and E. E. Werner. 1990. Behavioral and lifehistoncal responses of Iamal American toads to an odonate predator. Ecology 71:2313-2322. Slade, N. A., and R. J. Wassersug. 1975. On the evolution of complex life cydes. Evolution 29:568-571. Slater, D. W., and E. J. Dornfeld. 1939. A triple stain for amphibian embcyos. Stain Tech. 14:103-104. Slater, P., and A. R. Main. 1963. Notes on the biology of Notaan nichollsi Parker (Anura, Leptodactylidae).WestemAustralianNat. 8:163-166. Smirnov, S. V. 1990. Evidence of neoteny: A paedomorphic

LITERATURE CITED Smithberg, M. 1954. The origin and development of the t d in the frog, Ranapipiens.J. Exp. sol. 127:397425. Smith-Gili, S. J., and K. A. Berven. 1979. Predicting amphibian metamorphosis. AmericanNat. 113:563-585. Smith-Gill, S. J., and D. E. G d . 1978. Curvilinearities in the competition equations: An experiment with ranid tadpoles. AmeriEanNat. 112:557-570. Snetkova, E., N. Chelnaya, L. Serova, S. Saveliev, E. Cherdanzova, S. Pronych, and R. Wassersug. 1995. The effects of space fight on Xevwpus Ia&> larval development. J. Exp. Zwl. 273:21-32. Sobotka, J. M., and R. G. Rahwan. 1994. Lethai effect of latex gloves on Xevwpus luevis tadpoles.J. Phamurcol. Toxicol.Meth. 32:59. Sokal, R. R., and I. Huber. 1963. Competition among genotypes in Tribolium castaneum at varying densities and gene frequencies (the sooty locus). American Nat. 97: 169-184. Sokal, R. R., and I. Karten. 1964. Competition among genotypes in Tribolium custaneum at varying densities and gene frequencies (the black locus). Genetics 49: 195-211. Sokol, A. 1984. Plasticity in the fine timing of metamorphosis in tadpoles of the hylid frog, Litmia ewin~i. Cupeia 1984:868-873. Sokol, O. M. 1959. Studien an pipiden Froschen. I. Die Kauiquappe von Hpnoehirus curtipes Noble. Anat. Anz. 162:154-160. . 1962. The tadpole of Hymenochirus boett~eri.Cupeia 1962:272-284. . 1975. The phylogeny of anuran larvae: A new look. Cupeia 1975:l-23. . 1977a. The free swimrning Pipa larvae, with a review of pipid larvae and pipid phylogeny (Anura: Pipidae).J. Mmphol. 154:357426. . 1977b. A subordinal classification of frogs (Amphibia: Anura).J. 2001. 182:505-508. . 1981a. The tilter apparatus of larval Pelodyespunctatus (Amphibia: Anura). AmphibifiReptilia 2: 195-208. . 1981b. The larval chondrocranium of Pelodpespnctatus, with a review of tadpole chondrocrania. J Mmphol. 169:161-183. Song, J., and L. R. Parenti. 1995. Clearing and staining whole fish specimens for simuitaneous demonstration of bone, cartilage and nerves. Copeia 1995:114-118. Sotelo, C. 1976. Morphology of the cerebeliar cortex. pp. 864892. In Frog neurobwlo~y, handbwk, edited by R. Liins and a W. Precht. Beriin: Springer-Verlag. Spaiianzani, L. 1785. Exprienmspour servir a l'histoire de lalagfratwn des animam et des plantes. Genve: Chez Barthelemi Chirol, Libraire. Spannhof, I., and L. Spannhof. 1971. Beobachtungen zur Brutbiologie und Larvenennvickiungvon G&othecu marsupiata. Wiss.Zeit. Uniu.Rostock Math. Naturwiss. Reihe 20: 97-104. Spannhof, L. 1956. Zur Morphologie und Histologie muzinhaltiger Zellen in den Nierentubuli des Kraiienfrosches.2001. Anz. Suppl. 19:291-296. Spannhof, L., and S. Dittrich. 1967. Histophysiologische Untersuchungen an den Flaschenzelien der Urniere vonXenupus hmi Daudin unter experimentelien Bedingungen. Zeit. Zellfmch. Mikros.Anat. 8 1: 4 0 7 4 15. Spannhof, L., and L. Jonas. 1969. Elektronenmikroskopische Untersuchungen zur Genese und Sekretbiidung in den Flaschenzellen der Urniere vom Kraiienfrosch. Zeit. Zellfwsch. Mikros. Anat. 95: 134-142. Spemann, H. 1898. ber die Entwickiung der tuba Eustachi

405

und des Kopfskeletts von Rana tempmaria. Zool. Jahrb. 11:389416. Speny, D. C. 1987. Relationship benveen natural variations in motoneuron number and body size in Xevwpus luevis: A test for size matching.J. Comp.Neurol. 264250-267. Sperry, D. C., and P. Grobstein. 1983. Postmetamorphic changes in the lumbar lateral motor column in relation to muscle growth in the toad, Bufo ameknus. J. Cmnp. Neurol. 216: 104114. . 1985. Regulation of neuron numbers in Xenupusluevis: Effects of horrnonal manipuiation aitering size at metamorph0sis.J. Comp.Neurol. 232:287-298. Sperry, R. W. 1963. Chemoaffinity in the orderly growth of nerve fiber patterns and connections. Proc. Natl. A d . Sci. United States ofAmenEa 50: 703-710. Spinar, Z. V. 1972. Tertialy@~sfiommn~alEurope. Prague: Academia. . 1980. Fossiie Raniden aus dem oberen Pliozan vonwiliershausen (Niedersachsen). Stutt~arterBeitr Naturk. 53B:l-53. Spimr, J. L., and N. C. Spitzer. 1975. Time of origin of RohonBeard neurons in the spinal cord of Xenopus luevis. Ameriuzn 2001. :781. (Abstr.) 15 Spray, D.C. 1976. Pain and temperature receptors of anurans. pp. 607-628. In Frog neurobwlo~y, handbook, edited by R. a Llins and W. Precht. New York: Springer-Verlag. Starr, R. C., and J. A. Zeikus. 1993. UTEX-The cuiture collections of algae at the University of Texas at Austin. J. Phywl. S u ~ l29: 1-106. . Starrett, P. H . 1960. Descriptions of tadpoles of Middle American frogs.Misc. Publ. Mus. Zool., Uniu.MMichigan (110):1-38, 1plate. . 1967. Observations on the life history of frogs of the family Atelopodidae. Herpetolo~iEa 23:195-204. . 1968. The phylogenetic signicance of the jaw musculature in anuran amphibians. Dissertation, University of Michigan. . 1973. Evolutionary pattems in larval morphology. pp. 25 1-271. In Evolutionay bwlogy o the anurans. Contemf researeh on mujmprobhms, edited by J. L. Vial. Columbia: University of Missouri Press. Stader, H. P., and R. D. Semlitsch. 1993. Effects of visual, chemical and tactile cues of fish on the behavioural responses of tadpoles. Anzm. Behav. 46:355-364. Stebbins, R. C. 1985.A@i~u& to westem qtiksand amphibians. Field marks of all species in westemNmthAmerica, including Baja Caltfmia. Boston: Houghton Mitflin. Stebbins, R. C., and J. R. Hendrickson. 1959. Field studies of amphibians in Colombia, South America. Uniu. Calfomia Publ. 2001. 56:497-540. Steele, C. W., S. Strickler-Shaw, andD. H . Taylor. 1991. Fdure of Bufo americanus tadpoles to avoid lead-enriched water. J. Herpetol. 25 :241-243. Stefaneh, A. 1951. The Mauthnerian apparatus in the Ichthyopsida: Its nature and function and correlated problems of neurohistogenesis. Quart. Rm.Bwl. 26: 17-34. Stehouwer, D. J. 1987. Effect of tectotomy and decerebration on spontaneous and elicited behavior of tadpoles and juved e frogs. Behav. Neurosci. 101:378-384. . 1988. Metamorphosis of behavior in the bullfrog (Ranacatesbeiana).Develop. Psyebobwl. 21 :383-395. Stehouwer, D. J., and P. B. Farel. 1980. Central and peripheral controls of swimming in anuran larvae. Brain. Res. 195:323-335.

406

LITERATURE CITED nial ganglia and the lateral line sense organs in Amblystoma punctatum. J. Exp. Zool. 35 :421496. . 1933. The development of the lateral-line sense organs in amphibians observed in living and vital-stained preparati0ns.J. Cump.Neurol. 57:507-540. Storm, R. M. 1960. Notes on the breeding biology of the red-legged frog (Rana auma aurora).Herpetologca 1 6 2 51259. Straka, H. and N. Dieringer. 1991. Internuclear neurons in the ocular motor system of fr0gs.J. Cump.Neurol. 3 12:537-548. Strathmann, R. R. 1978. The evolution and loss of feeding larvai stages of marine invertebrates. Evolutwn 32394-906. . 1993. Hypotheses on the origins of marine larvae.Ann. Rev. Ecol. Syst. 24:89-117. Strathmann, R. R., L. Fenaux, and M. F. Strathrnan. 1992. Heterochronic developmental plasticity in larval sea urchins and its implication for evolution of nonfeeding larvae. Evolution 46972-986. Strauss, R. E., and R. Altig. 1992. Ontogenetic body form changes in three ecological morphotypes of anuran tadpoles. Gmwth Develop.Aging 56:3-16. Straw, R. M. 1958. Experimental notes on the Deep Springs toad Bufo msul. Ecology 39:552-553. Strawinski, S. 1956. Vascularization of respiratory surfaces in ontogeny of the edible frog, fina esculenta L. 2001. Poloniae 7:327-365. Straznicky, C., and D. Tay. 1982. Retinoteaal map formation in d u d y innervated te-: A regeneration study in Xenopus with one compound eye following bilateral optic nerve section. J. Comp.Neurol. 206: 119-1 30. Streb, M. 1967. Experimenteiie Untersuchungen uber die Beziehung zwischen Schilddruse und Hypophyse wahrend der Larvalentwickiung und Metamorphose von Xenopus herk Daudin. Zeit. Zellfonch. 82,407433. Strickler-Shaw,S., and D. H. Taylor. 1990. Sublethal exposure to lead inhibits acquisition and retention of discriminate avoidance learning in green frog ( f i n a cumitans) tadpoles. Environ. Toxuol. Chem. 9:47-52. . 1991. Lead inhibits acquisition and retention learning in bullfrog tadpoles. Neurotoxicol. Teratol. 13:167-1 73. Strijbosch, H. 1979. Habitat selection of amphibians during their aquatic phase. Oikos 33:363-372. Strong, D. R., Jr. 1983. Natural variability and the manifold mechanisms of ecological communities. A m . c a n Nat. 122:636-660. Stuart, L. C. 1941. Another new Hypopachus from Guatemala. Proc. Bwl. Soc. Washington 54: 125-128. . 1948. The amphibians and reptiles of Alta Verapaz, Guatemala. MZtc. Publ. Mus. Zwl., Uniu. Michhan (69):l109. . 1961. Some observations on the natural history of tadpoles of Rhinuphrynw dorsalis Dumeril and Bibron. Heqetologua 17:73-79. Stuesse, S. L., W. L. R. Cruce, and K. S. Poweii. 1983. Merent and efferent components of the hypoglossal nerve in the grass frog, finapipiens.J. Comp.Neurol. 217:432439. . 1984. Organization within the cranial IX-X complex in ranid frogs: A horseradish peroxidase transport study. J. Comp.Neurol. 222:358-365. Sdivan, B. 1974. Reptilian hemoglobins. pp. 77-122. In Chemical wologl Vol. 9. Amphibia and Rqtilia. Edited by M. Florkin and B. Scheer. New York: Academic Press. Surnrners, K., and W Amos. 1997. Behavioral, ecological, and molecular genetic analyses of reproductive strategies in the

. 1981. Sensory interactions with a central motor program in anuran larvae. Brain Res. 218: 131-140. Stein, J. R., ed. 1975. Handbook ofphycologcal mmetak. Culture methoak and growth mmuremenh. Cambridge: Cambridge University Press. Steinke, J. H., and D. G. Benson, Jr. 1970. The structure and polysaccharide chemistry of the jeiiy envelopes of the egg of the frog, finapipiens. J.Morphol. 130:57-66. Steinman, R. M. 1968. An electron microscopic study of ciiiogenesis in developing epidermis and trachea in the embryo ofXenopuslaePis.AmericanJ Anat. 122:19-56. Steinwascher, K. 1978a. The effect of coprophagy on the growth of fina catesbeiana tadpoles. Cupeia 1978:130134. . 1978b. Interference and exploitation competition among tadpoles of fina utricularia.Ewlogy 59: 1039-1046. . 1979a. Competitive interactions arnong tadpoles: Responses to resource level. Ecology 60: 1172-1 183. . 1979b. Host-parasite interaction as a potential population-regulating mechanism. Ecology 60:884890. . 1981. Competition for two resources. Oecologia 49:415-418. Steinwascher, K., and J. Travis. 1983. Influente of food quality and quantity on early larval growth of two anurans. Copeia 1983:238-242. Steneck, R. S., and L. Watling. 1982. Feeding capabiiities and h t a t i o n of herbivorous mollusc: A functional group approach. Mar Bwl. 68:299-319. Stensaas, L. J., and S. S. Stensaas. 1971. Light and electron microscopy of motoneurons and neuropiie in the amphibian spinal cord. Brain Res. 31:67-84. Stephenson, E. M. T. 1951. The anatomy of the head of the Soc. London New Zealand frog, Lekpelma. Trans. 2001. 27:255-305. Stephenson, N. G. 1951a. Observations on the development of the amphicoelous frogs, Leiopelma and Ascaphus. J. Linnean Soc. (Zool.) 42: 18-28. . 1951b. On the development of the chondrocranium Soc. and visceral arches of Lewpelma anheyi. Trans.2001. Lundon 27:203-252. . 1955. On the development of the frog, Lewpelma hochstetteri Fitzinger. Proc. Zool. Soc. Lonun 124:785-795. Sterba, G. 1950. ber die morphologischen und histogenetischen Thymusprobleme bei Xenopus les Daudin nebst einigen Bemerkungen uber die Morphologie der Kaulquappen. Abh. Slrchs. AAkad. Wiss.Leipz& 44:1-55. Stevenson, H. M. 1976. Vertebrates ofFlorida. IdentzfiEatwn and distn'bution.Gainesville: University Presses of Florida. Stiffler, D. F. 1996. Exchanges of calcium with the environment and between body compartments in Amphibia. Physwl. Zool. 69:418434. Stiiiing, H . 1898. Zur Anatomie der Nebenniere. Anh. Mikros. Anat. EntwicklzLngsmech. 52: 176-195. Stirling, R. R, and E. G. Merrill. 1987. Functional morphology of frog retinal ganglion ceiis and their central projections: The dimming detect0rs.J. Comp.Neurol. 258:477495. Stohr, P. 1882. Zur Entwickiungsgeschichte des Anurenschadels. Zeit. WZs.Zool. 36:68-103. Stokely, I? S., and J. C. List. 1954. The progress of ossification in the skull of the cricketfrog P s e u h nigrita triseriata. Copeia 1954:211-217. . 1955. Observations on the development of the anuran urostyle. Trans.Amoican Micros. Soc. 74: 112-1 15. Stone, L. S. 1922. Experiments on the development of the cra-

LITERATURE CITED

407

Ammnian dart-poison frog, Dendrobates ventrttrtmacuhtus. Takisawa, A., Y Ohara, and K. Kan. 1952a. Die Kaumuskula. Behav. Ecol. 8:260-267. tur der Anuren (BufO vuharisjaponicus) wahrend der Metarnorphose. OkajimmFolia Anat. Japonica 24: 1-28. Summey, M. R., and A. Mathis. 1998. Alarrn responses to . chemicai stimuli frorn damanged conspecifics by larval an- Takisawa, A., Y. Ohara, and Y Sunaga. 1952b. Uber die Urngesurans: Tests of three neotropical species. Herpetologica taltung der Mrn. interrnandibularesder Anuren wahrend der Metamorphose. O&imas Folia Anat. Japonica 24:2 17-241. 54:402408. Sssbier, W 1936. Anlage und Metamorphose der Driisen der Takisawa, A., and Y. Sunaga. 1951. Uber die Entwickiung des M. depressor rnandibulae bei Anuren irn Laufe der MetaDarrnwandung bei den Anuren. Zeit. Wks. Zool. 148:309-349. rnorphose. OkajimmFoliaAnat.Japonica 23:273-293. Sutherland, R. M., and R. F. Nunnernacher. 1981. Fibers in Tanaka, S., and M. Nishinara. 1987. Foam nest as a potential spinal nerves of treefrogs: Eleutherodactylus and Hyh. J. food source for anuran larvae: A preiiminary experirnent. J. Ethol. 5:86-88. Comp.Neurol. 202:415-420. Swammerdam, J. 1737-1738. Bibliu Naturae. Vol. 11. Leyden: Tang, Y.-Y., and C. M. Rovainen. 1996. Cardiac output inXenuI. Severinus. pus &mis tadpoles during developrnent and in response to an Swanepoel, J. H. 1970. The ontogenesis of the chondrocraadenosine agonist. AmericanJ. Physwl. 270R997-1004. nium and of the nasal sac of the rnicrohylid frog Breviceps Tanirnura, A., and H. Iwasawa. 1986. Developrnent of gonad adcpenwpentheri Werner.Ann. Univ.Stellenbosch 45A: 1-1 19. and Bidder's organ in Bufo japonicus$msus: Histological obsemation. Sci. Rept. Nigata Unia. (23)D:11-21. Swart, C. C., and R. O. De S. 1999. The chondrocranium of . 1987. Germ cell kinetics in gonadal developrnent in the the Mexican burrowing toad, Rhznophrynus doralis (Anura: toad Bufo japonicusfOmsus. Zool. Sci. 4:657-664. Rhinophrynidae).J Helpetol. 33 :23-28. Sweet, S. S. 1980. Aiiometric inference in rnorphology. Ameri. 1988. Ultrastructurd observations on the origin and differentiation of sornatic cells during gonadal developrnent can 2001. 20:643-652. in the frog Rana nipomauhta. Develop. Gvoiah Dzffm Swofford, D. L., and G. J. Olsen. 1990. Phylogeny reconstruction. pp. 41 1-501. In Mobcuhr Systematics, edited by D. M. 30:681-691. Hiilis and C. Moritz. Sunderland, MA: Sinauer Associates. . 1989. Origin of somatic cells and histogenesis in the Syuzyamova, L. M., and N. L. Iranova. 1988. Morphological primordial gonad of the Japanese tree frog Rhaophorrrs arbcharacteristics and energy rnetaboiisrn in the tadpoles of meus. Anat. Embryol. 180:165-173. Rana awalis bread [sic] at different diets. EkologZya . 1991. Proliferative activity of sornatic cells during go1988:18-24. nadal development in the Japanese pond frog, Rana n@oSzabo, T, J.-P. Denzot, S. Blahser, M. Veron-Ravaiiie, and D. mauhta.J. Exp. 2001. 259:365-370. Rouilly. 1990. The olfactoretinalis systern does not corre- Tanim, J. B. 1975. Egg volume. Auk 92:576-580. spond to the "terminal nerve" as ascribed to F. Pinkus. Soc. Taugner, R., A. Schiller, and S. Ntokalou-Knittel. 1982. Cells Neurosci.A b a , p. 129. and intercellular contacts in glorneruii and tubules of the Szarski, H . 1957. The origin of the larva and rnetamorphosis in frog kidney: A freeze-fracture and thin section study Cell Amphibia. American Nat. 91 :283-301. Tiss. Res. 226: 589-608. Szkely, G. 1976. The rnorphology of rnotoneurons and dorsal Taxi, J. 1976. Morphology of the autonomic nervous systern. root fibers i the frog's spinal cord. BrainRes. 103:275-290. n pp. 93-1 50. In Frog neurobwlogy,a handbook, edited R Llins Szkely, G., N. Antal, and T. Gorcs. 1980. Direct dorsal root and W. Precht. New York: Springer-Verlag. projection onto the cerebellum in the frog. Neurosci. Lett. Taylor, A. C., and J. J. Kollros. 1946. Stages in the normal development of Ranapipiens larvae. Anat. Rec. 94:7-24. 19:161-165. Szkely, G., and G. Czh. 1976. Organization of locomotion. Taylor, C. L., and R. Altig. 1995. Oral disc kinernatics of four rheophilous anuran tadpoles. Helpetol.Nat. Hist. 3: 101-106. pp. 765-792. In Frog neurobwlogy, a handbook, edited by R. Llins and W Precht. New York: Springer-Verlag. Taylor, C. L., R. Altig, and C. R. Boyle. 1995. Can anuran tadTachibana, T. 1978. The Merkel cell in the labial ridge epidermis poles choose among foods that vary in quaiity? Alytes of anuran tadpole. I. Fine smcture, distribution, and cyto13:81-86. . 1996. Oral disc kinernatics of four lentic anuran tadchemical studies.Anat. Rec. 191:487-502. Tachihama, H., F. Sasaki, T. Tachibana, H. Iseki, T. Horiguchi, poles. Herpetol. Nat. Hist. 4~49-56. T. Kinoshita, and K. Watanabe. 1987. Fine structure and de- Taylor, D. H., and D. E. Ferguson. 1970. Extraoptic celestial generation of the horny teeth and the epidermis of the labial orientation in the southem cricket frog Acrisgryllus. Science 168:390-392. ridge of the anuran tadpole. Eurumi Univ. Dental J. 13:49-63. Taylor, D. H., C. W. Steele, and S. Strickier-Shaw. 1990. Responses of green frog (Rana chmitans) tadpoles to leadTahin, Q. S., N. C. Meirelles, and S. B. Gargoggini. 1979. Amrnoneoteiisrn and ureoteiisrn of terrestrial and aquatic ampolluted water. Environ. Toxicol. Chem. 9: 87-93. Taylor, E. H. 1942. Tadpoles of Mexican Anura. Univ. Knsas phibians. Cienc. Cult. (So Paulo) 31:771-774. Taigen, T. L., and F. H. Pough. 1985. Metaboiic correlates of Sci. Bull. 28: 37-55. anuran behavior. American Zool. 25:987-997. . 1954. Frog-egg-eating tadpoles of Anotheca cmonata Taigen, T. L., F. H. Pough, and M. M. Stewart. 1984. Water (Stejneger)(Saiientia, Hyiidae). Univ. Knsas Sci. Bull. balance of terrestrial anuran Ebutherodactylus cogui eggs: Im36:589-596. portance of parenta1 care. Ecology 65:248-255. Taylor, J. D., and J. T. Bagnara. 1972. Dermal chromatophores. Takahashi, H. 1971. Sex differentiation and development of the American Zool. 12:43-62. gonad i Rana rugosa. 2001. n Mag. (Tokyo) 80:15-21. Taylor, J. S. H., and A. Roberts. 1983. The early development Takasu, T., and H. Iwasawa. 1983. Ultrasmctural studies on of the primary sensory neurones in an amphibian embryo: A the testicular differentiation in the frog Rha~ophorus arboreus. scanning electron microscope study J. Embryol. lhp. Mor2001. Mag. (Tokyo) 92:501. (Abstr.) phol. 5:49-66.

408

LITERATURE CITED Thrail, J. H. 1972. Food, feeding, and digestive physiology of the larval bullfrog, Rana catesbeiana Shaw. Dissertation, IKnois State University. Tihen, J. A. 1965. Evolutionary uends in frogs. American Zool. 5:309-318. Tibergen, N. 1968. Curkus naturalists. Garden City, NY: Natural history libras; Doubleday & Co. Toal, K. R., 1 1 and J. R. Mendelson, 1 1 1995. A new species 1, 1. of H y h (Anura: Hylidae) from cloud forest in Oaxaca, Mexico, with comments on the status of the H y h bistincta group. Occas. Pap. Mus. Nat. H&., Unzv. Kansas (174):1-20. Toews, D. P ,and R. G. Boutilier. 1986. Acid-base reguiation in . the Amphibia. pp. 265-308. In Acid-base reguation in anid edited by N. Heisler. Amsterdam: Elsevier. , Toloza, E. M., and J. M. Diamond. 1990a. Ontogenetic development of nuuient uansporters in bullfrog intestine. A m ' canJ. Pbysiol. 2586 :760-769. . 1990b. Ontogenetic development of transporter reguiation in bullfiog intestine. American J. Physwl. 2586 :770-773. Toner, P. G., and K. E. Carr. 1968. Cell strwture.A n introduction to bwiugicai eiectron microswpy, Edinburgh: E. and S. Livingstone Ltd. Tosney, K. W. 1978. The early migration of neural crest ceiis in the trunk region of the avim embryo: An electron microscopic study. Develop. Bwl. 62:317-333. Townsend, D. S. 1986. The costs of male parentai care and its evolution in a neotropical frog. Behav. Ecol. Socwbiol. 19:187-195. . 1996. Pattems of parental care in frogs of the genus Eleutberodactylus. pp. 229-239. In Cont7~'butwns West Into dian berpetolagy: A tribute to Albert Schwartz, edited by R. Poweii and R. W. Henderson. Contrib. Herpetol., vol. 12. Ithaca, NY: Society for the Study of Amphibians and Reptiies. Townsend, D. S., and W. H. Moger. 1987. Plasma androgen levels during maie parental care in a tropical frog (Eieutberodactylus).Hom. Behav. 2 1:93-99. Townsend, D. S., and M. M. Stewart. 1985. Direct development in Eleutherodactylus coqui (Anura: Leptodactylidae): A staging table. Copeia 1985:423436. . 1986. The effect of tem~erature direct deveio~ment on in a terrestrial-breeding, neotropical frog. Copeia 4986: 520-523. Townsend, D. S., M. M. Stewart, and F. H. Pough. 1984. Male parental care and its adaptive significance in a neotropical frog. Anim. Behav. 32:42143 1. Townsend, D. S., M. M. Stewart, F. H. Pough, and P. F. Brussard. 1981. Internal ferrilization in an oviparous frog. Science 212:469471. Tracy, C. R. 1976. A model of the dynamic exchanges of water and energy between a terrestrial amphibian and its environment. Ecol. Momgr. 46:293-326. Trainor, F. R., S. Wailett, and J. Grochowski. 1991. A usem algal growth medium.f. Phycol. 27:460-461. Travis, J. 1980a. Genetic variation for larvai specific growth rate in the frog Hyhgratwsa. Growth 44: 167-18 1. . 198C)b.Phenotypic variation and the outcome of interspecific competition in hylid tadpoles. Evolutwn 34:40-50. . 198la. A key to the tadpoles of North Carolina. Brimieyana 6:119-127. . 1981b. Control of larval growth variation in a population o f PseudanZr mieriata (Anura: Hylidae). Evolutwn 5:423432.

Taylor, R E., Jr., and S. B. Barker. 1965. Transepidermai potential dXerence: Development in anuran larvae. Science 148:1612-1613. Taylor, R. J. 1976. Value of clumping to prey and the evolutionary response of ambush predators. American Nat. 110: 13-29. . 1979. The value of clumping to prey when detectability increases with group size. AmericanNat. 113:299-301. Taylor, W. R. 1977. Observations on specimen fkation. Proc. Biol. Soc. Washington 90: 753-763. Tejedo, M. 1991. Effects of predation by two species of sympatric tadpoles on embryo survival in natterjack toads (Bufocahmita).Herpetologca 47: 322-327. . 1992. Absence of the uade-off between the size and number of offspring in the natterjack toad (Bufo cahmita). Oecologa 90:294296. Tejedo, M., and R. Reques. 1992. Effects of egg size and density on metamorphic traits in tadpoles of the natterjack toad (Bufocalamita).J. Herpetol. 26: 146-152. . 1994a. Does larvai growth history determine timing of metamorphosis in anurans? A field experiment. Herpetologica 50:113-118. . 1994b. Plasticity in metamorphic traits of natterjack tadpoles: The interactive effeas of density and pond duration. Oikos 71:295-304. Ten Donkelaar, H. J., and R. De Boer-Van Huizen. 1982. Observations on the development of descending pathways from the brain stem to the spinal cord in the clawed toad Xenupus W s . Anat. Embyol. 163:461473. Teran, H. R., and A. M. De Cerasuolo. 1988. Estudio histomorfolgico de1 tracto digestivo larval de Gastmthecagracilis Laurent (Anura Hylidae). Neotrpica 34: 115-123. Terent'ev, P. V., and S. A. Chernov. 1965. Kq to tbe amphibians and reptiies. 3d enlarged ed. [English Translation]. Israel Program for Scientiic Translation, Jerusalem. Terhivuo, J. 1988. Phenology of spawning for the common frog (Rana temporaria L.) in Finland from 1846 to 1986. Ann. Zool. Fennici 25:165-175. Test, F. H., and R G. McCann. 1976. Foraging behavior of Bufo a m e h n w tadpoles in response to high densities of micro-organisms. Copeia 1976:576-578. Tevis, L., Jr. 1966. Unsuccessful breeding by desert toads (Bufo punctatus) at the limit of their ecological tolerante. Ecohgy 47: 766-775. Theil, E. C. 1970. Red blood ceii replacement during the transition from embryonic (tadpole) hemoglobin to adult (frogj hemoglobin in Rana catesbeiana. Comp. Biochem. Physwl. 33:717-720. Thibaudeau, D. G., and R. Altig. 1988. Sequence of ontogenetic development and atrophy of the oral apparatus of six anuran tadpo1es.J. Morphol. 197:63-69. Thibier-Fouchet,C., O. Mulner, and R. Ozon. 1976. Progesterone biosynthesis and metabolism by ovarian foilicles and isolated oocytes ofXenupush v k . Bwl. Reprod. 14:3 17-326. Thiele, J. 1888. Der Hafiapparat der Batrachierlarven. Zeit. Wiss.Zool. 46:67-79. Thomas, E. I. 1972. Structure of the heart in the tadpole larvae of Rana tzkrinu Daud. Ann. Zool.,Agra 8:41-50. Thompson, R L. 1996. Larval habitat, ecology, and parental invesunent of Osteopilus h n n e u s (Hylidae). pp. 259-269. In Contributwns to West Indian hqetolagy: A tribute to Albert S h a r t z , edited by R. Poweii and R. W. Henderson. Contrib. Herpetol., vol. 12. Ithaca, NY: Society for the Study of Amphibians and Reptiies.

LITERATURE CITED

409

. 1983a. Variation in development patterns of larval anurans in temporary ponds. I. Persistent variation within a Hylagratwsa population. Evolutwn 37:496-5 12. . 1983b. Variation in growth and survival of Hyhgratiosa larvae in experimental enclosures. Copeia 1983:232237. . 1984. Anuran size at metamorphosis: Experimental test of a model based on intraspecific competition. Ecology 65:1155-1160. Travis, J., S. B. Emerson, and M. Blouin. 1987. A quantitativegenetic analysis of larval Me-history traits in Hyh m q e 1 : Evolutian 41:145-156. Travis, J., W. H. Keen, and J. Juilianna. 1985a. The effects of muitiple factors on viability selection in Hyhgratiasa tadpoles. Evolution 39: 1087-1099. . 1985b. The role of relative body size in a predator-prey relationship between dragonfly naiads and larval anurans. Oihs 45:59-65. Travis, J., and J. C. Trexler. 1986. Interactions among factors affecting growth, development, and survival in experimental popuiations of Bufo tewestris (Anura: Bufonidae). Oecalogia 69: 110-116. Treisman, M. 1975. Predation and the evolution of gregariousness. I. Models for concealment and evasion. Anim. Behav. 23: 779-800. Trenerry, M. P. 1988. The ecology of tadpoles in a tropical rainforest stream. Honours thesis, James Cook University, Townsviile, Queensland, Australia. Trewavas, E. 1933. The hyoid and larynx of the Anura. Trans. Roy. Phiiosqh. Soc. Loruion 222:401-527. Tmeb, L. 1973. Bones, frogs, and evolution. pp. 65-132. In Evolutwnary bwlogy of the anurans. Contempmary research on majorpmblems, edited by J. L. Vial. Columbia: University of Missouri Press. . 1979. Leptodactylid frogs of the genus Telmatobim in Eqador with the description of a new species. Copeia 1979:714733. . 1985. A summary of osteocranial development in anurans with notes on the sequence of cranial ossification in Rhinophynus dmalis (Anura: Pipoidea: Rhinophrymdae). South AficanJ. Sci. 81:181-1 85. . 1993. Patterns of cranial diversity among the Lissamphibia. pp. 255-343. In The vertebrateskull, edited by J. Hanken and B. Haii. Chicago: University of Chicago Press. Tmeb, L., and D.C. Cannatelia. 1982. The cranial osteology and hyolaryngeal apparatus of Rhinophrynw domalZr (Anura: Rhinophrynidae) with comparisons to Recent pipid frogs. J. Motphol. 171:1140. . 1986. Systematics, morphology, and phylogeny of genus Pipa (Anura: Pipidae). Herpetolo~ica 42:412449. Tmeb, L., and W. E. Dueilman. 1970. The systematic status and Me history of Hyh vmcigeva Werner. Copeia 1970:601-610. Tmeb, L., and J. Hanken. 1993. Skeletal development in Xempus uevzj (Anura: Pipidae).J. Morphol. 214: 1 4 1 . Tschugunova, T J. 1981. Interfiontalia in Bombina orientalis (Blgt) and Bombina bombina (L.). pp. 117-121. In Herpeto@cal investgatwnsin Sibwut and the Far East, edited by L. J. Borkin. Leningrad: Akad. Nauk SSSR. Tsuneki, K., H. Kobayashi, R. Gallardo, and P K. T. Pang. . 1984. Electron microscopic study of the inne~ation the of renal tubuies and urinary bladder epithelium in Rana catesbeiana and Nectums muiosus. J. Morphol. 181:143-154. Tubbs, L. O. E., R. Stevens, M. Weils, and R. Altig. 1993. On-

togeny of the oral apparatus of the tadpole of Bufo a m . canus. Amphibia-Reptilia 14:333-340. Tung, T. C. 1935. On the time of determination of the dorsoventral axis of the pronephros in Discoglossw. Peking Nat. Hist. Bull. l0:115. Tung, T. C., and S. H. KY. 1944. Experimental studies on the development of the pronephric duct in anuran embryos. J. Anat. 78: 52-57. Turner, R. J. 1969. The functional development of the reticuloendothelial system in the toad, Xenopm uevzj (Daudi). J. Exp. Zool. 170:467480. Turner, R. J., and M. J. Manning. 1972. Response of the toad, Xenopus laevis, to circuiating antigens. I. Cellular changes in the spleen.f . Exp. Zool. 183:21-34. . 1974. Thymic dependente of amphibian antibody responses. EuropeanJ Immunol. 4:343-346. Turpen, J. B., and C. M. Knudson. 1982. Ontogeny of hematopoietic ceils in Ranapipiens: Precursor celi migration during embryogenesis.Deveiup. B d . 89: 138-15 1. Turpen, J. B., C. M. Knudson, and P S. Hoefen. 1981a. Em. byonic origin of hematopoietic precursor cels from the region of the presumptive mesonephros in Ranapipiens. Anat. Rec. 199:261. (Abstr.) . 1981b. The early ontogeny of hematopoietic cells studied by grafting cytogenetically labeled tissue aniagen: Localization of a prospective stem ceii compartment. Deveiup. Bwl. 85:99-112. Turpen, J. B., C. J. Turpen, and M. Flajnik. 1979. Experimental analysis of hematopoietic celi development in the liver of larvai Ranapipns. Develup. Bwl. 69:466479. Turnipseed, G., and R. Altig. 1975. Popuiation density and age structure of three species of hylid tadpoles. J. Herpetol. 9:287-291. Twitty, i? C. 1955. Eye. pp. 402414. InAnalyk ofdeaeiupment, edited by B. H. Willier, P. A. Weiss, and i? Hamburge~ Philadelphia: W B. Saunders. . 1966. Of scientZrts and salamandem. San Francisco: W. H. Freeman. Twombly, S. 1996. Timing of metamorphosis in a freshwater crustacean: Comparison with anuran models. Ecology 77: 1855-1866. Tyler, M. J. 1962. On the presemation of anuran tadpoles. AustralianJ Sci. 25:222. . 1963. A taxonomic study of amphibians and reptiles of the Central Highlands of New Guinea, with notes on their ecology and biology. 2. Anura: Ranidae and Hylidae. Trans. Roy. SOC. Australia 86: 105-130. South . 1968. Papuan hylid frogs of the genus Hyh. Zool. Vmh. (96):1-203,4 plates. . 1971. The phylogenetic significance of vocal sac structure in hylid frogs. Univ. Knsas Publ. Mus. Nat. Hist. 19:319-360. . 1985. Reproductive modes in Australian Arnphibia. pp. 265-267. In Biology ofAustrahn'anF70gsand reptiles, edited by G. Grigg, R. Shine, and H. Ehmann. Chipping Norton, New South Wales: Surrey Beatty. . 1989. A~alianJTogs. Penguin Books Australia, Ringwood, Victoria. Tyler, M. J., and M. Anstis. 1975. Taxonomy and biology of frogs of the Litoria citropa mmplex (Anura: Hylidae). Rec. South Australia Mus. 17:41-50. Tyler, M. J., M. Anstis, G. A. Crook, and M. Davies. 1983. Reproductive biology of the frogs of the Magela Creek system, Northern Territory. Rec. South Australia Mw. 18:415440.

410

LITERATURE CITED

Tyler, M. J., and M. Davies. 1978. Species-groups within the Austraiopapuan hyiid frog genus LitmUt Tschudi. Australian J. Zool. Suppl. (63):1-17. . 1983. Larvai development. pp. 44-57. In Theg& broodingfrog, edited by M. J. Tyler. London: Croom Helm. Tyler, M. J., A. A. Martin, and M. Davies. 1979. Biology and systematics of a new limnodynastine genus (Anura: Leptodactyiidae) from north-western Australia. AustralianJ. Zool. 27:135-150. Uchiyama, H., T. A. Reh, and W. K. Steii. 1988. Immunocytochemicai and morphologicai evidence for a retinopectai projection in anuran amphibians.J. Comp.Neurol. 274:48-59. Uchiyama, M., S.-T. Iwasaki, and T. Murakami. 1991. Surface and subsurface structures of neuromasts in tadpoles of the crab-eating frog, Rana cancrivora.J. Morphol. 207: 157-164. Uchiyama, M., T. Murakami, and H. Yoshizawa. 1990. Notes on the development of the crab-eatingfrog, Rana cancrivma. Zool. Sci. 7:73-78. Uchiyama, M., and H. Yoshizawa. 1992. Salinity tolerance and structure of externai and internai gills in tadpoles of the crabeating frog, Rana cancrivma. Cell Tiss.Res. 267:3544. Udin, S. B., and M. D. Fisher. 1985. The development of the nucleus isthmi in Xenupus M s . I. Ceii genesis and the formation of connections with the tectum. J. Comp. Neurol. 232:25-35. Udin, S. B., M. D. Fisher, and J. J. Norden. 1992. Isthmotectal axons make ectopic synapses i the monocular regions of the n tectum in developing Xenopus luevis frogs. J. Comp. Neurol. 322:461470. Ueck, M. 1967. Der manicotto glandulare ("Drsenmagen") der Anurenlarve in Bau, Funktion und Beziehung zur Gesamtlbge des Darmes. Eine mikroskopisch-anatomische, histochemische, und elecmonenoptische Studie an der omnivoren und mikrophagen larve von Xenqpus laevts und der carnivoren und makrophagen larve von Hymenochimts boettgeri (Anura, Pipidae). Zeit. Wiss.Zool. 176:173-270. Ueda, H. 1986. Reproduction of Chirixalusezfingm' (Boettger). Sci. Rept. Lab. Amphib. Bwl., Hiroshimu Univ. 8: 109-116. Uhlenhuth, E. 1919. Relation between thyroid gland, metamorphosis, and gr0wth.J. &p. Zool. 1:473482. . 1921. The internai secretions in growth and development of amphibians.American Nat. 55: 193-221. Ultsch, G. R. 1973. The effects of water hyacinths (Eichhornia mmsipes) on the rnicroenvironment of aquatic cornmunities. Arch. Hydrobwl. 72:460-473. . 1976a. Ecophysiologicai studies of some metabolic and respiratory adaptations of sirenid salamanders. pp. 287-312. In Respiratwn of amphibww vertebrates, edited by G. M. Hughes. New York: Academic Press. . 1976b. Gas exchange and metabolism in the Sirenidae (Amphibia: Caudata).I. Oxygen consumption of submerged sirenids as a function of body size and respiratory surface area. Comp.Bwchem. Physwl. 47A:485498. . 1987. The potentiai role of hypercarbia in the transition from water-breathing to air-breathing in vertebrates. Evolutwn 41 :442445. . 1989. Ecology and physiology of hibernation and overwintering among freshwater fishes, turties, and snakes. Bwl. Rea. 64:435-516. . 1996. Gas exchange, hypercarbia and acid-base balance, paleocology, and the evolutionary transition from waterbreathing to air-breathing among vertebrates. Palaeogeograph. Palaeoclimatol.Palaeoecol. 123:1-27. Ultsch, G. R., and J. T. Duke. 1990. Gas exchange and habitat

selection in the aquatic salamanders Nectums mactrlosus and Ctyptobranchusalkganiensis. Oecologia 83:250-25 8. Ultsch, G. R., R. W. Hanley, and T. R. Baurnan. 1985. Responses to anoxia during simulated hibernation in northern and southern painted turtles. Ecology 66:388-395. Ultsch, G. R., and D.C. Jackson. 1996. pH and temperature in ectothermic vertebrates. Bull. Alabama Mus. Nat. H h . 18:l-41. Ultsch, G. R., and D.C. Jackson, and R. Moalli. 1981. Metaboiic oxygen conformity among lower vertebrates: The toadfish revisited.J Comp.Physwl. 142:439443. Ultsch, G. R., M. E. Ott, and N. Heisler. 1980. Standard metaboiic rate, critical oxygen tension, and aerobic scope for spontaneous activity of trout (Salmogairdneri)and carp (Cypinus curpw) in acidified water. Comp. Bwchem. Physwl. 67k329-335. Underhill, D. K. 1966. An incidente of spontaneous caudal scoliosis in tadpoles of Rana ptpiens Schreber. Copeia 1966:582-583. . 1967. Spontaneous silvery color and pituitary abnormaiity in tadpoles from a laboratory cross of Ranapipiens. Copeia 1967:673-674. Uray, N. J. 1985. Early stages in the formation of the cerebeiiurn in the fr0g.J. Comp.Neurol. 232: 129-142. Uray, N.J., and A. G. Gona. 1979. Golgi studies on Purkinje ceii development in the frog during spontaneous metamorphosis. 11. Details of dendritic development. J. Comp. Neurol. 185:237-252. . 1982. Golgi studies on Purkinje ceii development in the frog during spontaneous metamorphosis. 111. Axonai deve1opment.J. Comp.Neurol. 212:202-207. Uray, N.J., A. G. Gona, and K. F. Hauser. 1987. Autoradiographic studies of cerebeiiar histogenesis in the premetamorphic bullfrog tadpole I. Generation of the externai granular 1ayer.J. Comp.Neurol. 266:232-246. . 1988. Autoradiographic studies of cerebeliar histogenesis in the premetamorphic buiifrog tadpole. 1 . Formation 1 of the interauricular granular band. J. Comp. Neurol. 269:118129. Utsunomiya, Y., and T. Utsunomiya. 1983. On the eggs and larvae of the frog species occurring in the Ryukyu Archipelag0.J. Fm. Appl. Bwl. Sci., Hiroshima Univ. 22:255-270. Utsunomiya, Y., T. Utsunomiya, S. Katsuren, and M. Toyama. 1983. Habitat segregation observed in the breeding of five frog species dwelling in a mountain stream of Okinawa Island. Annot. Zool.Japonenses 56: 149-1 53. Vaientine, J. W., and D. Jablonski. 1983. Larvai adaptations and patterns of brachiopod diversity in space and time. Evolutwn 37: 1052-1061. Vaiett, B. B., and D. L. Jameson. 1961. The embryology of Eleutherod~@ylus augusti latrans. Copeia 1961:103-109. Van Bambeke, C. 1863. Recherches sur Ia structure de Ia bouche chez les ttards des batraciens anoures. Bull. Acad. Roy. Belgique Ser. 2 16:339-354. Van Bergeijk, W. A. 1959. Hydrostatic baiancing mechanism of Xenqpus 1arvae.J. Acoustu. Soc. America 31 :1340-1347. Van Beurden, E. K. 1980. Energy metabolism of dormant Australian water-holding frogs (Cycloranaplatycephalus). Copeia 1980:787-799. Van Buskirk, J. 1988. Interactive effects of dragonfly predation in experimental pond communities. Ecology 69:857-867. Van Buskirk, J., S. A. McCollurn, and E. E. Werner. 1997. Natural selection for environmentally induced phenotypes in tadpoles. Evolution 5 1:1983-1992.

LITERATURE CITED Van Buskirk, J., and K. L. Yurewicz. 1998. Effects of predators on prey growth rate: Relative contributions of thinning and reduced aaivity. Oikos 82:20-28. Vance, W H., G. L. Clifton, M. L. Applebaum, W D. Wiis, Jr., and R. E. Coggeshail. 1975. Unmyelinated preganglionic fibers in frog ventral roots. J Comp. Neurol. 164:117-126. Van de Karner, J. C. 1965. Histologicai structure and cytology of the pineai cornplex in fishes, amphibians and reptiles. Proo~ Brain Res. 10:3 0 4 8 . Van Den Broek, A. J. P., G. J. Van Oordt, and G. C. Hirsch. 1938. Das Urogenitalsystem. pp. 683-854. In Handbuch der vergleischenden Anatomie dere Wirbeltime, edited by L. Bolk, E. Goeppen, E. Kallius, and W Lubosch. Berlin: Urban & Schwarzenberg. Van Der Horst, C. J. 1934. The lateral line nerves of Xenupus. Psychiat. Neurol. B1. Amsterdum 3:426-435. Van Der Linden, J. A. M., and H. J. Ten Donkelaar. 1987. Observations on the developrnent of cerebellar af'ierents inXenopus laevis. Anat. Embyol. 176:431-439. Van Der Linden, J. A. M., H . J. Ten Donkelaar, and R. De Boer-van Huizen. 1988. Development of spinocerebellar af'ierents in the clawed toad, Xenopus iuevis.J Comp. Neurol. 277:41-52. Van Der Westhuizen, C. M. 1961. The development of the chondrocranium of Heleophynepumlli Sclater, with special reference to the palatoquadrate and the sound-conducting apparatus. Acta 2001. 1-72. 42: Van Dijk, E. D. 1959. On the cloacal region of Anura in particular of larvai Ascaphus. Ann. Univ.Stellenbosch 35: 169-249. . 1966. Systematic and field keys to the families, genera, and described species of southern African anuran tadpoles. Ann. Natal Mus. 18:231-286. . 1972. The behavior of southern African anuran tadpoles with particular reference to their ecology and related external morphologicai features. 2001. Ajkmna 7:49-55. . 1981. Material data other than preserved specimens. Monit. Zool. Italiano (N.S.)Suppl. 15:393-400. . 1985. Hemisus mamratum adults reported to carry tadpoles. SouthAfTzcanJ Sci. 8 1:209-210. Van Eeden, J. A. 1951. The development of the chondrocranium ofrircaphustruei Stejneger with special reference to the relations of the palatoquadrate to the neurocranium. Acta 2001. 32:41-176. Van Gansen, P. 1986. Ovognese des amphibiens. pp. 21-55. In Trait de Zoologie,Anatomie, Systematique, Biologie. Vol. 14, Fasc. lB, edited by P. P. Grass and M. Delsol. Paris: - Masson. Van Geertruyden, J. 1946. Recherches exprimentales s u la formation du msonphros chez les amphibiens anoures. Arch. Bwl. 57:145-181. . 1948. Des premiers stades de dveloppement du msonphros chez les amphibiens anoures. Acta Morphol. Neerbndo Scandinavica 6:1-17. Van Kampen, P. N. 1923. The Amphibia of the Indo-Australian archz$eboo. Leiden: E. J . Bri, Ltd. (29 figs). Van Mier, P., and H. J. Ten Donkelaar. 1988. The development of primary af'ierents to the lumbar spinal cord in Xenupus l m i . Neurosci. Lett. 84:35 4 0 . Van Mier, P., R. Van Rheden, and H. J. Ten Donkelaar. 1985. The development of the dendritic organization of primary and motoneuons in the spinal cord of &nupus 1ami. An HRP study.Anat. Embyol. 172:311-324. Vannini, E., and P. P. Giorgi. 1969. Organogenesi de1 I'apparato
'

41 1

urogenitale degli anfibi: Agenesi ed internizione de1 dotto di WoM in embrioni di Bufo bufo. Arch. Italiano Anat. Embnol. 74: 111-143. Vannini, E., and A. Sabbadin. 1954. The relation of the interrenai blastema to the origin of the somatic tissue of the gonad in frog tadpo1es.J Embyol. Exp. Morphol. 2:275-289. Van Oordt, P. G. W. J. 1974. Cytology of the adenohypophysis. pp. 53-106. In Physwlogy of the Amphibia. Vol. 2, edited by B. Lofts. New York: Academic Press. Van Pletzen, R. 1953. Ontogenesis and morphogenesis of the breast-shoulder apparatus of Xenopus laevij. Ann. Univ. Stellenbosch 29A: 137-184. Van Seters, W. H. 1922. Le dveloppernent du chondrocrne d3Al'es obstetncans avant Ia rntamorphose. Arch. Bwl. 32:373491. Vaz-Ferreira, R., and A. Gehrau. 1975. Comportamiento epimeletico de Ia rana comun, .ptodmty1us ocelbtus (L.)(Amphibia, Leptodactylidae).I. Atencion de Ia cria y actividades alimentaria y agresivas relacionadas. Physis 34: 1-14. Veltkamp, C. J., J. C. Chubb, S. P. Birch, and J. W. Eaton. 1994. A simple freeze dehydration method for studying epiphytic and epimoic communities using the scanning electron rnicroscope. Hydrobwlogia 288: 3 3 4 0 . Verrna, N., and B. A. Pierce. 1994. Body mass, developmental stage, and interspecific differences in acid tolerante of larval anurans. TuasJ. Sci. 46:319-327. Vermeij, G. J. 1976. Interoceanic differences in vulnerability of shelled prey to crab predation. Nature 260:135-136. Vesselkin, N. P., A. L. Agayan, and L. M. Nomokonova. 1971. A study of the thalamo-telencephalic af'ierent systems in frogs. Brain Behav. Evol. 4:295-306. Via, S., and R. Lande. 1985. Genotype-environment interaction and the evolution of phenotypic plasticis; Evolutiun 39:505-522. Viertel, B. 1982. The oral cavities of central European anuran larvae (Amphibia) morphology, ontogenesis, and generic diagnosis.Amphibia-Reptilia 4:327-360. . 1984a. Filtration, eine Strategie der Nahrungsaufnahme der Larven von Xenupus Iaem>, Rana temporaria und Bufo calamita (Amphibia, Anura). Virh. Ges. Okolooie (Berlin) 12:563575. . 1984b. Habit, melanin pigmentation, oral disc, oral cavity, and filter apparatus of the larvae of Baleaphyne muktensis. pp. 21-43. In Life histoy ofthe Mallmcan midwife toad, edited by H. Hemmer and J. A. Aicovet: Ciutat de Mallorca. . 1984c. Suspension feeding of the larvae of Baleaphlyne muletensis. pp. 153-161. In Lfe histoy of the Mallorcan midwfe toad, edited by H. Hemmer and J. A. Aicover. Ciutat de Mallorca. . 1985. The filter apparatus of Rana temporaria and Bufo bufo larvae (Amphibia, Anua). Zoonzorphology 105:345-355. . 1987. The fiiter apparatus of Xenupw W s , Bombina variegata, and Bufo calamita (Amphibia,Anua): A comparison of difFerent larvai types. 2001. Jahrb.Anat. 115:425452. . 1990. Suspension feeding of anuran larvae at low concentrations of 6hlmelb algae (Amphibia, Anua). Oemlogia 85: 167-177. . 1991. The ontogeny of the filter apparatus of anuran larvae (Amphibia,Anua). Zoommphology 110:239-266. . 1992. Funaional response of suspension feeding anuran larvae to different particle sizes at low concentrations. Hydrobwloga 234: 151-173. . 1996. Der Buccopharynx und der Filterapparat der Anurenlarven -Morphologie, Ultrastructur, Ontogenese,
-

412

LITERATURE CITED

Funktion, Leistung bei der Nahrungsaufnahme und okologopmentai biology, evolution and phylogenesis? Int. J. Deische Bedeutung.Mainzer Natumiss.Archiv. 34: 187-263. ve@. Bwl. 402359-869. Vietti, M., A. Cianca Perone, and A. Guardabassi. 1973. Pro- Wake, D. B., K. C. Nishikawa, U. Dicke, and G. Roth. 1988. nephric degeneration in normal and prolactin treated larvae Organization of the motor nuclei in the cemicai spinal cord from Rana temporaria, in correlation with metamorphosis. of salamanders.J. Comp.Neurol. 278: 195-208. Boll. Zool. 40:401404. Wake. M. H. 1977. Fetal maintenance and its evolutionarv significance in the Amphibia: Gymnophiona. J. ~ e r p e t o . Vigny, C. 1979. Morphologie larvaire de 12 espces et s o u 11:379-386. 865377-891. espces due genre Xenupus. Rev. Suisse 2001. . 1978. The reproductive biology of Eieutherodadylus Vaa, J., R. W. McDiarmid, and J. M. Gailardo. 1982. Arthrojmperi (Amphibia, Anura. Leptodactylidae),with comments pod predators of leptodactylid frog foam nests. Bvenesia 191 on the evolution of live-bearing systems. J. Herpetol. 20:577-589. 12:121-133. Via, J., and C. E. Valerio. 1982. Red, white, and brown. Pre. 1979. The comparative anatomy of the urogenital sysliminary observations on the color of the centrolenid tadpole tem. pp. 555-614. In Hyman3 cmnparative vertebate anat(Amphibia: Anura: Centrolenidae). Brenesiu 19/20:1-1 6. omy, edited by M. H. Wake. 3d ed. Chicago: University of Viy, F. 1890. The development of the ear and accessory organs in the common frog. Quart.J.Mimos. Soc. 30:523-550. Chicago Press. Viiter, V., and A. Lugand. 1959. Recherches sur le dtermini. 1980. The reproductive biology ofNectophlynoidesmalcolmi (Amphibia: Bufonidae), with comments on the evolusme interne et externe du corps jaune gestatif chez le crapaud tion of reproductive modes in the genus Nectoph~noides. Covivipare du Mont Nimba, le Nectophlynoides occidentalis Ang. peiu 1980:193-209. de Ia Haute Guine. C. R Soc. Bwl. Paris 153:294-297. Vine, I. 1973. Detection of prey flodrs by predators. J. Theol: - . 1982. Diversity within a framework of constraints. AmBwl. 40:207-210. phibian reproductive modes. pp. 87-106. In Environmental Viparina, S., and J. J. Just. 1975. The life period, growth, and adaptatwn and evolution, edited by D. Mossakowski and G. differentiation ofRanacatesbeiuna larvae occurring in nature. Rotk New York: Gustav Fischer-Verlag. . 1989. Phylogenesis of direct development and viviparCopeia 1975:103-109. ity in vertebrates. pp. 235-250. Ln ComplexorganismalfincVisser, J. 1985. The fanglike teeth of the early larvae of some twns: Integratwn and evolutwn in vertebrates, edited by D. B. Hewhlyne. South AjncanJ. Sci. 8 1:200-202. Vitalis, T. Z. 1990. Pulmonary and cutaneous oxygen uptake Wake and G. Roth. New York: Wiiey. and oxygen consumption of isolated skin in the frog, Rana . 1993a. The skuU as a locomotor organ. pp. 197-240. In The skull. Vol. 3. Functwnd and evolutwnary mechanisms. pipiens. R e s - Physwl. 8 1:391400. Vizotto, L. D. 1967. Desenvolvimento de anuros da regio norteEdited by J. Hanken and B. K. Hail. Chicago: Universiq of ocdental do Estado de So Paulo. Fac. Fiiosof. Cinc. Letr. Chicago Press. Zool. So Jose Rio Prto No. Especial. 161 p. . 1993b. Evolutionary scenarios, homology and converVogt, C. C. 1842. Untenuchungen uber die Entwickelungsgente of s t n i c t ~ ~specializations for vertebrate vivipariq. al geschuhte dev Geburtsheljierkrote.Solothurn. A m ' c a n Zool. 32:256-263. Vokov, V I. 1982. Morphological characteristics of the frog . 1993c. Evolution of oviductai gestation in amphibians. (Rana ridibunda) myocytes in the process of cardiogenesis. J. Exp. Zool. 266:394413. Arch. Anat, Cistol.Embryol. 82:73-80. Wake, M. H., and J. Hanken. 1982. Development of the skuil Voris, H . K., and J. P. Bacon, Jr. 1966. Differentiai predation of Demz~phismexicanus (Amphibia: Gymnophiona), with on tadpoles. Copeia 1966:594-598. cornments on skull kinesis and amphibian relationships.J. Wachtel, S. S., G. C. Koo, and E. A. Boyse. 1975. EvolutionMorphol. 173:203-223. ary conservation of H-Y ("maie") antigen. Naure Wakeman, J. M., and G. R. Ultsch. 1976. The effects of 254:270-272. dissolved 0, and CO, on metabolism and gas-exchangeparWager, V. A. 1929. The breeding habits and life-histories of titioning in aquatic saiamanders. Physwl. Zool. 48:348-359. some of the Transvaal Amphibia. -11. Trans. Roy. Soc. South Waldman, B. 1981. Sibling recognition in toad tadpoles: The A j k a 17:125-135. role of experiente. Zeit. Tierpsychol. 56:341-358. . 1965. Theji-08sofSouthAfzca.Cape Town and Johan. 1982a. Sibiing association among schoohg toad tadnesburg: Purneil& Sons Pty., Ltd. poles: Field evidence and implications. Anim. Behav. . 1986. Frogs of South Afzca. Theirfmcinating life stories. 30:700-713. Craighaii: Delta Books. . 1982b. Adaptive significance of communai oviposition Wagner, E., and M. Ogielska. 1990. Oogenesis and developin wood frogs (Rana sylvatica). Behav. Ecol. Socwbwl. ment of the ovary in European green frog, Rana ridibunda 10:169-174. (Pailas).11.Juvenile stages until adults. 2001. Jahrb.Abt. Anat, . 1984. Kin recognition and sibiing association among Ontog. TTiere 120:223-231. wood frog (Rana sylvaticu) tadpoles. Behav. Ecol. Socwbwl. Wakahara, M. 1978. Induction of supernumerary primordial 14:171-180. germ ceils by injecting vegetal pole cytoplasm into Xenupus . 1985a. OKactory basis of kin recognition in toad tadeggs. J. Exp. Zool. 203: 159-164. p01es.J. Gmp. Physwl. 156:565-577. . 1982. Chronological changes in the accumulation of . 1985b. Sibling recognition in toad tadpoles: Are kinPoly (A)+RNA in developingcells ofXenupusla&~,with speship labels transferred among individuais? Zeit. T ~ s y c h o l . cial reference to PGCs. Deve@. Growth Dzffm 24:311-318. 68:41-57. Wake, D. B. 1970. Aspects of vertebral evolution in the modern . 1986a. Preference for unfamiliar siblings over familiar Amphibia. F o m et Functio 3:33-60. nonsiblings in American toad (Bufo a&canus) tadpoles. Wake, D. B., and J. Hanken. 1996. Direct development in the Anim. Behav. 34:48-53. lungless salamanders: What are the consequences for devel. 1986b. Chemical ecology of kin recognition in anuran

LITERATURE CITED

413

ogy and general description. Occm. Pap. Mus. Nat. Hist., Univ.Knsm (48):1-23. Sih-erstein. New York: Plenum. . 1976b. Internai oral features in Hyh @h (Anura: Hylidae) larvae: An ontogenetic study. Occas. Pap. Mus. Nat. . 1989. Do anuran larvae retain kin recognition abiiities Woning metamorphosis?Anim. Behav. 37: 1055-105 8. Hist., Univ.Knsm (49):1-24. . 1991. Kin recognition in amphibians. pp. 162-219. In . 1976c. The identification of leopard frog tadpoles. CoKUi rmgnition, edited by P. G. Hepper. Cambridge: Campeia 1976:413414. bridge University Press. . 1980. Internai oral features of larvae from eight anuran WGkhnan,B., and K. Adler. 1979. Toad tadpoles associate preffamilies: Functionai, systematic, evolutionary, and ecological considerations. Mhc. Publ. Mus. Nat. H&., Uniu. Knsm aentiaiiy with siblings. Nature 282:611-613. Waliace, R A. 1978. Oocyte growth in nonmammalian verte(68):1-146. brates. pp. 469-502. In Thevmtebrateoaaty. Comparative bwl. 1984. The Pseuhhemkus tadpole: A morphologicallink qp and mlution, edited by R. E. Jones. New York: Plenum. between microhylid (Orton Type 2) and ranoid (Orton Type 4) larvae. Herpetologca 40: 138-149. lWiace, R L., and L. V. Diller. 1998. Length of the larvai cycle ofhxphus trcrei in coastal streams of the Redwood Region, . 1986. How does a tadpole know when to metamord e r n Ca1ifornia.J. Herpetol. 32:404-409. phose? A theory linking environmental and hormonai cues. W a k S. C., and A. R. Blaustein. 1995. Larvai marbled saiaJ. Theor:Bwl. 118:171-181. manders, Ambystoma opacum, eat their kin. Anim. Behav. . 1989a. Locomotion in amphibian larvae (or "Why 50:537-545. aren't tadpoles built like fishes?").American Zool. 29:65-84. Wdms, B. 1975. Studies of interspeciic predation within an . 1989b. What, if anything is a microhylid (Orton type amphibian community.J. Herpetol. 9:267-279. 11) tadpole?Fonch. Zool. 35 :534-538. Wangersky, P J. 1976. The surface iim as a physical environ. . 1992. The basic mechanisms of ascent and descent by anuran larvae (Xenoptslaems). ment. Ann. h Ecol. Syst. 7: 161-176. . Copeia 19922390-894. Warjnger-Loschenkohl,A. 1988. An experimental study of mi. 1996. The biology of Xenopus tadpoles. pp. 195-21 1. mhabitat selection and microhabitat shifts in European tadIn The bwlogy of Xenopus, edited by R. C. Tinsley and H. R. poles. Amphibia-Reptilia 9:2 19-236. Kobel. Oxford: Oxford University Press. Wackentin, K. M. 1992a. Effects of temperature and iumina. 1997a. Where the tadpole meets the world -observarion on feeding rates of green frog tadpoles (Ranachmitans). tions and speculations on biomechanical and biochemical Copnn 1992:725-730. factors that iduence metamorphosis in anurans. A ~ r i c a n . 1992b. Microhabitat use and feeding rate variation in Zool. 37: 124-136. g e a frog tadpoles (Rana chmitans). Copeia 1992:731. 1997%. Assessing and controlling amphibian popula740. tions from the larvai perspective. pp. 271-281. In Amphibi1995. Adaptive plasticity in hatching age: A response uns in decline: Canadian studies of agiubalpbiem, edited by D. M. Green. Herpetul. Consen,. (1). Victoria, British Coto predation risk trade-offs. Proc. Nat. Acad. Sci., United States 92:3507-3510. lumbia W a m e ~ C., W A. Dunson, and J. Travis. 1991. Interaction Wassersug, R., and W. E. D u e h a n . 1984. Oral structures and S. their development in egg-brooding hylid frog embryos and o f pH, density, and priority effects on the survivorship and p a z h of two species of hylid tadpoles. Oecologa larvae: Evolutionary and ecological implications.J. Mmphol. 88:331-339. 182:1-37. W x x q S. C., J. Travis and W. A. Dunson. 1993. Effect of pH Wassersug, R. J., and M. E. Feder. 1983. The effects of aquatic &ation on interspecific competition between two species oxygen concentration, body size and respiratory behavior on d-hyhd tadpoles. Ecology 74: 183-194. the stamina of obligate aquatic (BufameriEanw)and facultaWarren, E. 1922. Observations on the development of the nontive air-breathing (Xenopts laems and Rama berhnicti) anq m i c tadpole ofAnhydrophtynerattrayi Hewitt. SouthA+uran 1arvae.J. Exp. BwL. 105:173-190. a s J . Sn'. 19:254-262, plates I1 and III. Wassersug, R. J., K. J. Frogner, and R. F. Inger. 1981a. AdaptaWasserman, W. J., and L. D. Smith. 1978. Oocyte maturation tions for Me in tree holes by rhacophorid tadpoles from m mnmammalian vertebrates. pp. 443468. In The verteThai1and.J. Herpetul. 15:41-52. hrrtc ariaty comparative bwium an evolutwn, edited by R. E. Wassersug, R. J., and C. M. Hessler. 1971. Tadpole behaviour: Aggregation in larvai Xenopus laevis. Anim. Behav. Jcma. New York: Plenum. W ,R. 1971. On the comparative paiatability of some 19:386-389. dryseason tadpoles from Costa Rica. American M d . Nat. Wassersug, R J., and W R. Heyer. 1983. Morphologicai corre86:101-109. lates of subaeriai existente in leptodactylid tadpoles associ. 1972. The mechanism of ultraplanktonic entrapment ated with flowing water. C a d i a n J . Zool. 61:761-769. m anuran 1arvae.J.Morphol. 137:279-288. . 1988. A survey of internai oral features of leptodacty. 1973. Aspects of social behavior in anuran larvae. pp. loid larvae (Arnphibia: Anura). Smithsonian Contrib. Zool. 273-297. In Evolutwnaty bwlogy of the anurans. Contemporaty (457):1-99. rrrawch on majmproblemc, edited by J. L. Viai. Columbia: Wassersug, R. J., and K. Hoff. 1979. A comparative study of L-niversity of Missouri Press. the buccai pumping mechanism of tadpoles. Bw1.J.Linnean . 1974. Evolution of anuran life cycles. S c ~ m Soc. 12:225-259. 185:377-378. . 1982. Developmentai changes in the orientation of the anuran jaw suspension. A preliminary exploration into the . 1975. The adaptive significance of the tadpole stage uith comments on the maintenance of complex Me cycles in evolution of anuran metamorphosis. pp. 223-246. In Evoluanirans.A&n Zool. 15:405417. tionaty Bwiugy, Vol. 15, edited by M. Hecht, B. Waiiace, and . 1976a. Oral morphology of anuran larvae: TenninolG. Prace. New York: Plenum.
1-01. 4, edited by D. Duvai, D. Muller-Schwarze, and R. M.

amphibians. pp. 225-242. In Chemical signals in vertebrates.

414

LITERATURE CITED Weber, M. 1898. Ueber auffdende ecaudaten-larven von Tjibodas (Java).Ann. Jard. Bot. Buitenwrg Suppl. 2:5-10. Weigmann, D. L. 1982. Relationships between production and abundance of ranid (Ranapipiens,R. catesbeiana, and R. clamitans) tadpoles and the effeas of tadpole grazing on algal productivity and diversity and on phosphoms, nitrogen, and organic carbon. Dissertation, Michigan State University. Weisz, P, B. 1945a. The development and morphology of the larva of the South Mrican clawed toad, Xenopus mm I. The third-form tadpo1e.J Morphol. 77: 163-191. . 1945b. The development and morphology of the larva of the South Atrican clawed toad, Xenupus heras 11. The hatching and the first- and second-form tadpoles. J. Morphol. 77: 193-217. Weils, K. D. 1978. Courtship and parental behavior in a Panamanian poison-arrow frog (Dendrobatesauratus).Hnptologica 34:148-155. . 1980. Evidence for growth of tadpoles during parentai transport in Colosrethus ingztinalis.J Herpetol. 14:426-428. Wells, K. D., and K. M. Bard. 1988. Parenta1 behavior of an aquatic-breeding tropical frog, Leptoductylus bolivianus. J. Herpetol. 22:361-364. Wenig, J. 1913. Der Albinismus bei den Anuren, nebst Bemerkungen uber den Bau des Amphibien integuments. Anat. Anz. 43:113-135. Werner, E. E. 1986. Amphibian metamorphosis: Growth rate, predation rate, and the optimai size at transformation. Ameruan Nat. 128:3 19-341. . 1988. Size, scaling, and the evolution of complex life cycles. pp. 60-81. In Size-mctured populatwns: Ecology and evolutwn, edited by B. Ebenman and L. Persson. Berlin: Springer-Verlag. . 1991. Nonlethai effeas of a predator on competitive interactions between two anuran larvae. Ecology 72:17091720. . 1994. Ontogenetic scaiing of competitive relations: Size-dependent effeas and responses in two anuran larvae. Ecology 75: 197-213. Werner, E. E., and B. R. Anliolt. 1996. Predator-induced behaviorai indirect effeas: Consequences to competitive interactions in anuran larvae. Ecoio~y 771157-169. Werner, E. E., and J. F. Giiiiam. 1984. The ontogenetic niche and species interactions in size-structured popuiations. Ann. Rev. Ecol. Syst. 15:393-425. Werner, E. E., and M. A. McPeek. 1994. Direct and indirea effeas of predators on two anuran species aiong an environmental gradient. Ecolgy 75: 1368-1382. Werschkul, D. F., and M. T. Christensen. 1977. Differentiai predation by Lepomis macrochims on the eggs and tadpoles of Rana. Herpetoio~ica :237-241. 33 Wessels, N. K. 1968. Problems in the anaiysis of determination, mitosis and differentiation. pp. 132-151. In Epitheliamesenchymal interactwns. 18th Hahnemann Symp., edited by R. Fleischmajer and R. E. Biiiingham. Baitimore: Williams and Wiikins. Wessenberg, H. S. 1978. Opaiinata. pp. 552-581. In Parasitu p t o w a . Vol. I I . Intestinal jlqellates, histomonad, tt-ichomunads, amueba, opalinid, and ciliates, edited by J. P. Kreier. New York: Academic Press. West, N. H., and W. W. Burggren. 1982. Giii and lung ventilatory responses to steady-state aquatic hypoxia and hyperoxia in the bullfrog tadpole. Respir. Physiol. 47:165-176. . 1983. Reflex interactions between aerial and aquatic gas exchange organs in larvai bulkfrogs. AmericanJ. Physwl. 244R. 770-777.

. 1985. The kiematics of swimming in anuran larvae. J. Exp. Biol. 119:l-30. Wassersug, R. J., A. M. Lum, andM. J. Potel. 1981b. An analy- ' sis of school stmcture for tadpoles (Anura: Amphibia). Behav. Eml. Socwbwl. 9: 15-22. Wassersug, R. J., and A. M. Murphy. 1987. Aerial respiration faciiitates growth in suspension-feedinganuran larvae (Ximpus laevii).Exp. Bwl. 46: 141-147. Wassersug, R. J., R. Paui, and M. E. Feder. 1981c. Cardiorespiratoty synchrony in anuran larvae (Xenopw hais, Pachymedusa ducnuoiol; and Rana berlandieri). Comp. Bwchem. Physwl. 70A:329-334. Wassersug, R. J., and W. F, Pyburn. 1987. The biology of the Pe-ret' toad, Otophyne robusta (Microhylidae), with special consideration of its fossorial larva and systematic relationships. Zoo1.J. Linnean Soc. 9 1:137-169. Wassersug, R., and K. Rosenberg. 1979. Surface anatomy of branchiai food traps of tadpoles: A comparative study. J. Morphol. 159:393426. Wassersug, R. J., and E. A. Seibert. 1975. Behavioral responses of amphibian larvae to variation in dissolved oxygen. Copeia 1975:86-103. Wassersug, R. J., and K. A. Souza. 1990. The bronchial diverticuia of Xenopus larvae: Are they essentiai for hydrostatic assessment?Natunt.issenschaj?en 77:443-445. Wassersug, R. J., and D. G. Speny. 1977. The relationship of locomotion to differential predation on Pseudami triseriata (Anura: Hylidae). Ecology 58:830-839. Wassersug, R. J., and R. Viertel. 1993. Do cilia drive water through the buccopharyngeal and opercular cavities in the fossorial Otophyne robusta tadpole? Amphibia-Reptilia 14:83-86. Wassersug, R. J., and D. B. Wake. 1995. Fossil tadpoles from the Miocene of Turkey. Alytes 12:145-157. Watanabe, K., F. Sasaki, and H. Takahama. 1984. The uitrastructure of oral (buccopharyngeai) membrane formation and rupture in the anuran embryo. Anat. Rec. 210: 5 13-524. Watkins, T. B. 1997. The effea of metamorphosis on the repeatability of maximal locomotor performance in the Pacific tree frog, Hyla regilla.J Exp. Bwl. 200:2663-2668. Watson, G. F., and A. A. Martin. 1973. Life history, larvai morphology, and relationships of Australian leptodactylid frogs. Trans.Roy. Soc. SouthAustralia 97: 3 3 4 5 . Watt, K. W. K., and A. Riggs. 1975. Hemoglobins of the tadpole of the bullfrog, Rana catesbeiana. J Bwl. Chem. 250:5934-5944. Watt, P. J., S. F. Nottingham, and S. Young. 1997. Toad tadpole aggregation behaviour: Evidence for a predator avoidance function.Anim. Behav. 54:865-872. Webb, J. F., and D. M. Noden. 1993. Ectodermal placodes: Contributions to the development of the vertebrate head. Amwican Zool. 33:343-447. Webb, P. W. 1975. Hydrodynamics and energetics of fish propuision. BULI.Fish. Res. Bd. Canada 190:1-158. . 1978. Fast-start performance and body form in seven species of teleost fish.J Exp. Bwl. 74:211-226. . 1984a. Form and function in fish swimming. Sci. America Juiy:71-82. . 1984b. Body form, locomotion, and foraging in aquatic vertebrates.American 2001. 107-120. 24: Webb, R. G. 1963. The larva of the casque-headedfrog, PtevnohylafadiensBouienger. TexmJ. Sci. 15:89-97. Webb, R. G., and J. K. Korky. 1977. Variation in tadpoles of frogs of the Rana tarahumarae group in western Mexico (Anura: Ranidae).Herpetologica 33 :73-82.

LITERATURE CITED

415

Westerfield, M. 1994. The zebrafish book.Aguidefmthe laboratq use ofiebrafsh (Brachydaniorerio). Eugene, OR: Institute of Neuroscience, University of Oregon. Wever, E. G. 1973. The ear and hearing in the frog, Ranapipiens.J.Morphol. 141:461-477. . 1985. The amphibian eax Princeton, NJ: Princeton University Press. Weygoldt, P. 1980. Cornplex brood care and reproductive behavior in captive poison-arrow frogs, Dendrobatespumilw O. Schmidt. Behav. Ecal. Socwbwl. 7:329-332. . 1989. Feeding behavior of the larvae of Fritzianagoeidi (Anura, Hylidae). Amphibia-Reptilia 10:419-422. Weygoldt, P., and S. P. De Carvaho e Silva. 1991. Obse~ations on mating, oviposition, egg sac formation and developrnent in the egg-brooding frog, Fritzianagoeldi. Amphibia-Reptilia 12:67-80. White, B. A., and C. S. Nicoil. 1981. Horrnonal control of amphibian metamorphosis. pp. 363-396. In Metamorphosis a probiem in developmntal bw[ogy.22 ed., edited by L. I. Gilbert and E. Frieden. New York: Plenum. Whitear. M. 1976. Identification of the e~idermd "Stiftchenzellen" of frog tadpoles by electron rnicroscopy. Cell Tiss. Res. 175:391-402. . 1977. A functional comparison between the epidermis of fish and of amphibians. Symp. Zool. Soc. London 39:291-313. . 1983. The question of free nerve endings in the epidermis of lower vertebrates. Acta Bwl. Hungariw 34:303319. Whitford, W. G., and K. H. Meltzer. 1976. Changes in 0, consumption, body water, and lipid in burrowed desert juvenile anurans. Herpetoiogica 32:23-25. Whitington, E M., and K. E. Dixon. 1975. Quantitative studies of gerrn plasrn and gerrn ceils during early ernbryogenesis of Xenopus laevis.J. Embryol. Exp. Morphol. 33:57-74. Wiens, J. A. 1970. Effects of early experience on substrate patrern selection in Rana aurora tadpoles. Copeia 1970:543-548. . 1972. Anuran habitat selection: Early experience and substrate selection in Rana acadae tadpoles. Anim. Behav. 20:218-220. Wiens, J. J. 1989. Ontogeny of the skeleton of Spea bombzfi.ons (Anura: Pelobatidae).J. Morphol. 202:29-5 1. Wibur, H. M. 1972. Cornpetition, predation, and the structure of the Ambyrtoma-Rana sylvatica community. Eco[ogy 53:3-21. . 1976. Density-dependentaspects of metamorphosis in Amlytoma and Rana ylvatica. Eco&y 57: 1289-1296. 1977a. Density-dependentaspects of growth and rnetarnorphosis in Bufo americanus. Ecoiiyy 58: 196-200. . 1977b. Interactions of food leve1 and population density in Rana sylvatica. Eco&y 58:206-209. . 1980. Cornplex life cycles. Ann. h Ecol. Syst. . 11:67-93. . 1984. Cornplex life cycles and community organization i amphibians. pp. 195-224. In A new ecokgy: Nwel apn proaches to i n t h v e systems, edited by P. W. Price, C. N. Slobodchikoff, and W. S. Gaud. New York: Wiey. . 1987. Redation of structure in cornolex svsterns: Experimental teLporaq pond communities. ico[ogly 68: 14371452. . 1988. Interactions between growing predators and growing prey. pp. 157-172. In Size-mctured populatwns; eco[ogyand evolutwn, edited by B. Ebenman and L. Persson. New York: Springer-Verlag.

. 1989. In defense of tanks. Herpetologica 45:122-123. Wiibur, H. M., and R. A. Aiford. 1985. Priority effects in experimental pond communities: Responses of Hyu to Buj? and Rana. Eco[ogy 66: 1106-1 114. Wilbur, H. M., and J. P. Collins. 1973. Ecological aspects of amphibian metamorphosis. Science 182:1305-13 14. Wilbur, H. M., and J. E. Fauth. 1990. Experimental aquatic food webs: Interactions between two predators and two prey. Amm'can Nat. 135 :176-204. Wilbur, H. M., E J. Morin, and R. N. Harris. 1983. Salamander predation and the structure of experimental communities: Anuran responses. Eco[ogy 64: 1423-1429. Wibur, H . M., and R. D. Semlitsch. 1990. Ecological consequences of tail injury in Rana tadpoles. Copeia 1990:1824. Wilczynska, B. 1981. The structure of the alimentaq canal and the dimensions of the mucosa in ontogenetic developrnent of some Anura. Acta Bwl. Cracov. 23: 1 3 4 6 . Wilczynski, W. 1981. Afferents to the midbrain auditory center in the bulifrog, Rana catesbeiana. J. Comp. Neurol. 198:421-433. . 1988. Brainstem auditory pathways in anuran amphibians. pp. 209-23 1. In The evolution of the amphibian auditory system, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak. New York: Wiley. . 1992. The nervous system. pp. 9-39. Environmental physio&y ofthe amphibians, edited by In M. E. Feder and W. W Burggren. Chicago: University of Chicago Press. Wilczynski, W, and R. G. Northcutt. 1977. Afferents to the optic teaurn of the leopard frog: An HRP study.J Comp.Neurol. 173:219-230. . 1983a. Connections of the bulifrog striatum: Afferent organization.J. Comp.Neurol. 214:321-332. . 1983b. Connections of the bullfiog striatum: Efferent projections.J. Comp.Neurol. 2 14:333-343. Wild, E. R. 1995. New genus and species ofAmazonian microhylid frog with a phylogenetic analysis of New World genera. Copeia 1995:837-849. . 1996. Natural history and resource use of four Amazonian tadpole assernblages. O c a . Pap. Nat. Hist., Univ. I&sas (176):l-59. . 1997. Description of the adult skeleton and developmental osteology of the hyperossified horned frog, Ceratopbys cmuta (Anura: Leptodactylidae). J. Mwphol. 232: 169-206. Wiley, E. 0. 1979. Ventral gili arch musdes and the interrelationships of gnathostornes, with a new classification of the vertebrata. Zoo1.J. Linnean Soc. 67: 149-179. . 1981. Phyiiyenetics: The theory andpactice ofphyiyenetic systematics. New York: Wiley. Wiihelm, G. B., and R. E. Coggeshail. 1981. An electron rnicroscope analysis of the dorsal root in the fr0g.J. Comp.Neurol. 196:421-429. Wili, U. 1982. Efferent neurons of the lateral-line systern and the VIIIth cranial nerve i the brainstern of anurans. Cell Tiss. n Res. 225:673-685. . 1986. Mauthner neurons survive rnetamorphosis in anurans: A cornparative HRP study on the cytoarchitecture of Mauthner neurons in amphibians. J. Comp. Neueztml. 244: 111-120. . 1988. Organization and projections of the area octavolateralis in amphibians. pp. 185-208. In The evolutwn ofthe amphibian auditory ystem, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak. New York: Wiley.

416

LITERATURE CITED morphology and physiological irnportance. Endeavmr 35:27-31. Wise, R. W 1970. An irnmunologicai comparison of tadpole . and frog hemoglobins. Comp. Bwchem. Physd. 32: 89-95. Witschi, E. 1949. The larval ear of the frog and its transformation during metarnorphosis. Zeit. Naturfmh. 4B:230242. . 1950. The bronchial diverticula of Xenopus hm-is Daudin. Anat. Rec. 108:590. . 1955. The bronchial columella of the ear of larval Ranidae. J. Molphol. 96:497-5 11. . 1956. Dmelopwnt of v~ebrates.Phiiadelphia: W. B. Saunders. . 1962. Equivalent numericai designations for staging systems: Amphibians and fishes. pp. 272-273. In Grawth, including repmduction and mmphologcal hehpment, edited by P. L. Altman and D. S. Dittmar. Washington, D. C.: Fed. American Soc. Exp. Biol. Woiimuth, L. E, and L. I. Crawshaw. 1988. The effect of development and season on temperature selection in bullfrog tadpoles. Physwl. Zool. 61:461469. Woiimuth, L. P., L. I. Crawshaw, R. B. Forbes, and D. A. Grahn. 1987. Temperature selection during development in a montane anuran species, Rana C C L T C ~ Physiol. Zool. . 60:472480. Woilweber, L., R. Stracke, and U. Gothe. 1981. The use of a simple method to avoid cell shrinkage during SEM preparati0n.J Mimos. 121:185-189. Wong, A. L.-C., and T J. C. Beebee. 1994. Identification of a uniceiiular, nonpigmented alga that mediates growth inhibition in anuran tadpoles: A new mernber of the genus Prototheca (Chiorophyceae: Chiorococcales). Hydrobioiugia 277:85-96. Wong, S., and D. T. Booth. 1994. Hypoxia induces surfacing behaviour in brown-striped frog (Lzmnodynastesperonii) larvae. Comp.Bwchem. Physwl. 109k.437-445. Woodward, B. D. 1982. Tadpole competition in a desert anuran community. Oecoiugia 54:96-100. . 1983. Predator-prey interactions and breeding-pond use of temporary-pond species in a desert anuran community. Ecology 64:1549-1555. . 1986. Paternal effects on juvenile growth in Scaphwpus multiplicatus (the New Mexico spadefoot toad). American Nat. 128:58-65. . 1987a. Clutch parameters and pond use in some Chihuahuan desert anurans. Southwest.Nat. 32: 13-19. . 198% Intra- and interspecific variation in spadefoot toad (Scaphiopus) clutch parameters. Southwest. Nat. 32:127-156. 1987c. Interactions between Woodhouse's toadtadpoles (BufD woodhousii) of mixed sizes. Copeia 1987:380-386. . 1987d. Paternal effects on offspring traits in Scaphiopls couchi (Anura: Pelobatidae).Oecoiu~ia 73:626429. Woodward, D. B., and J. Travis. 1991. Paternal effects on juvenile growth and survival in spring peepers (Hyia crucifer). Eml. Ewl. 5:40-5 1. Woodward, D. B., J. Travis, and S. Mitchell. 1988. The effects of the mating system on progeny performance in Hyia crucifer (Anura: Hylidae). Evolutwn 42:784-794. Workrnan, G., and K. C. Fisher. 1941. Temperature selection and the effect of temperature on movement in frog tadpoles. AmericanJ. Physwl. 133P:499-500. Wratten, S. D. 1994. Videotechniques in animal ewlogy and behavwur London: Chapman and Haii.

Will, U., and B. Fritzsch. 1988. The eighth nerve of amphibians: Peripheral and central distribution. pp. 159-183. In The molutwn of the amphibian auditmy system, edited by B. Fritzsch, M. J. Ryan, W. Wilczynski, T. E. Hetherington, and W. Walkowiak. New York: Wdey. Wiii, U., B. Fritzsch, G. Luhede, and P Gorner. 1985. The area . octavo-lateralis in Xenopus W . Part 11. Second order projections and cytoarchitecture. Ce11 Tiss. Res. 239: 163175. Willey, A. 1893. Note on the position of the cloacal aperture in certain batrachian tadpoles. Trans. Nav T i k Acad. Sci. 12:242-245. Willhite, C., and P. V Cupp, Jr. 1982. Daily rhythms of thermal tolerante in Rana chmitans (Anura: Ranidae) tadpoles. Comp.Biochem. Physwl. 72A.255-257. Wfiams, G. C. 1964. Measurement of consociation among fishes and comments on the evolution of schooling. Publ. Mus. Michhan St. Univ. 2: 35 1-383. Wdiams, J. D., and E. Gudynas. 1987. Descripcion de Ia larva de Ehhistocleis bicoiur (Valenciennes 1838)(Anura: Microhylidae). Amphibia-Reptilia 8:225-229. Wiiliams, M. A., and L. D. Smith. 1971. Ultrastructure of the "germinal plasm" during maturation and early cleavage in Rana pipiens. Dmelop. Biol. 25 :568-580. Wiiliamson, I., and C. M. Buli. 1989. Life history variation in a population of the frog Ranihlia signifera: Egg size and early development. Copeia 1989:349-356. Wfi, I. A. 1936. The respiratory rate of developing Amphibia with special referente to sex differentiation. J. Exp. Zool. 73:481-510. Wiison, D. J., and H. Lefcort. 1993. The effect of predator diet on the alarm response of red-legged frog, Rana auria, tadpoles. Anim. Behav. 46: 1017-1019. Wiison, L. D., and J. R. McCranie. 1993. Preliminq key to the known tadpoles of anurans from Honduras. Life Sci. Octas. Pap. Roy. Ontario Mus. (40):1-12. Wiitshire, D. J., and C. M. Buli. 1977. Potential competitive interactions between larvae of Pseztdophryne bibrmi and Ei semim a m r a t a (Anura: Leptodactylidae). Australian J. Zool. 25 :449454. Wimberger, P. H. 1992. Plasticity of fish body shape. The effects of diet, development, family, and age in two species of Geophagus (lJisces: Cichlidae). Bwl. J. Linnean Soc. 45:197-218. . 1993. Effects of vitamin C deficiency on body shape and skuil osteology in Geophgus brrasiliensis: Implications for interpretations of morphological plasticity. Copeia 1993:343-351. Wmklbauer, R., and P. Hausen. 1983a. Development of the lateral line system inXenopuslaePis. I. Normal development and ceil movement in the supraorbital system. J. Embryol. Exp. Mophol. 76:265-28 1. . 1983b. Development of the lateral line system inXempus hm-is. 1 . Cell multiplication and organ formation in 1 the supraorbital system. J. Embryol Exp. Mmphol. 76:283296. . 1985a. Development of the lateral line system inXenopus hm-is. 1 1 Development of the supraorbital system in 1. triploid embryos and larvae. J. Embryol. Exp. Molphol. 88: 183-192. . 1985b. Development of the lateral line system inXenopus W .TV. Pattern formation in the supraorbital system. J. Embryol. Exp. Molphol. 88: 193-207. Wischnitzer, S. 1976. The lampbmsh chromosomes: Their

LITERATURE CITED

417

Wray, G. A. 1996. Pardel evolution of nonfeeding lawae in echinoids. Syst. Bwl. 45:308-322. Wright, A. A., and A. H. Wright. 1933. Handbook ofjro~5 and toads -the frogs and toads of the United States and Canada. Handbooks of Amm.can Natural Histmy. Vol. 1. Ithaca, NY: Comstock Publ. . 1942. Handbook offrogsand toads- thej?o&s and toads of the United States and Canada. Handbooks ofAmeriLanNatural Histmy. Vol. 1. 2d ed. Ithaca, NY. Comstock Publ. Wright, A. H. 1914. North American Anura-Me-histories of the Anura of Ithaca, New York. Carne@ Inst. Wmhington Publ. (197):1-98 . 1929. Synopsis and description of North American tadpoles. Proc. United States Natl. Mus. 74, 11 (2756):1-70, plates 1-9. . 1932. Life-hhries ofthejro~s Okefinokee Swamp, Gemof 8". North American Salientia (Anura) no. 2. New York: Macmiiian Publ. Wright, A. H., and A. A. Wright. 1949. Handboak offrogs and toads of the United States and Canada. 3d ed. Ithaca, NY: Comstock Publ. . 1995. Handbook offrogs and toa& ofthe United States and Canada. Ithaca, NY. Cornstock Classic Handbooks, Comeii Paperbacks, Cornstock Publ. Assoc. Wright, M. L., L. S. Blanchard, S. T. Jorey, C. A. Basso, Y M. . Myers, and C. M. Paquette. 1990. Metamorphic rate as a function of the lightldark cycle in Ranap$iens larvae. Comp. Bwchem. Physiol. 96A:2 15-220. Wright, M. L., L. J. Cykowski, L. Lundrigan, K. L. Hemond, D. M. Kochan, E. E. Faszewski, and C. M. Anuszewski. 1994. Anterior pituitary and adrenal cottical hormones accelerate or inhibit tadpole hindlimb growth and development depending on stage of spontaneous development or thyroxine concentration in induced metamorphosis. J. Exp. 2001. 175-188. 270: Wright, P. M., and P. A. Wright. 1996. Nitrogen metabolism and excretion in bullfrog (&na catesbeiana) tadpoles and adults exposed to elevated environmental ammonia levels. Phvswl. 2001.69: 1057-1078. wright, S. 1931. Evolution in Mendelian populations. Genetics 16:97-159. -. Wylie, C. C., and J. Heasman. 1976a. The formation of the gonadal ridge in Xenopus la&. I. A light and transmission elearon microscope study. J. Embqol. Exp. Mmphol. 35: 125-138. . 1976b. The formation of the gonadal ridge in X m p s W.11. A scanning electron microscope study. J. Embyol. Exp. Morphol. 35: 139-148. Wylie, C. C., J. Heasman, A. Snape, M. O'DriscoU, and S. Holwili. 1985. Primordial germ ceiis of Xenopus hmi5 are not irreversibly determined early in development. Deveiup. Bwl. 112:66-72. Wylie, C. C., S. Heasman, A. P. Swan, and B. H. Anderton. 1979. Evidence for substrate guidance of primordial gerrn ceiis. Exp. CeLLRes. 121:315-324. Wylie, C. C., and T B. Roos. 1976. The forrnation of the gonadal ridae in Xewus W.1 1 The behaviour of isolated 1. primordiz gerrn iells in vitro. J. Embyol. Exp. Morphol. 35: 149-157. Xavier, F. 1977. An exceptional reproductive strategy in Anura: Nectopbiynoides occidentalis Angel (Bufonidae), an example of adaptation to terrestrial life by viviparity. pp. 545-553. In Majmpatterns in vertebrate evolution, edited by M. K. Hecht, P. C. Goody, and B. M. Hecht. New York: Plenum.
-

Yang, D.-T. 1991. Phylogenetic systematics of the Amoiups groups of ranid frogs of southeastern Asia and the Greater Sunda Islands. Field. 2001.(N.S.) (1423):1 4 2 . Ye, C., and L. Fei. 1995. Taxonomic studies on the small type Megophqs in China including descriptions of the new species (subspecies) (Pelobatidae: genus Megophys). Acta Herpetol. Sinica 6:72-8 1. Yong, H. S., C. K. Ng, and R. Ismail. 1988. Conquest of the land: Direct development in a Malaysian Phihutus tree frog. Nat. Mahysiana 13:4-7. Yoshizaki, N. 1973. Ultrastructure of the hatching gland ceiis in the South African clawed toad, Xenupus W.J. Fm. Sci. Hokkaido Univ., Ser. W 2001. , 18:469480. . 1974. Ultrastructural cytochemistry of hatching gland cells in Xenupus ernbryos in relation to the hatching process. J.Fm. Sci. Hokkaidu Univ., Ser. VT, 2001. 309-3 14. 19: . 1975. Ultrastruaure of the frog hatching gland cell in Mag. relation to its role in the hatching process. 2001. (Tokyo) 84:39-47. . 1991. Changes in surface ultrastructure and proteolytic activity of hatching gland ceiis during developrnent of Xenopus embryo. Zool. Sci. 8:295-302. Yoshizaki, N., and C. Katagiri. 1975. Cellular basis for the production and secretion of the hatching enzyme by frog embry0s.J. Exp. Zool. 192:203-212. Yoshizaki, N., and M. Yamamoto. 1979. A stereoscan study of the development of hatching gland ceiis in the ernbyonic epiderrnis of Runa japonica. Acta Emblyol. Exp. 979:339-348. Yoshizaki, N., and H. Yamasaki. 1991. Morphological and biochemical changes in the fertiiization coat of Xenupus hmis during the hatching process. Zool. Sci. 8:303-308. Yoshizato, K. 1986. How do tadpoles lose their tails during metamorphosis?Zool. Sci. 3:2 19-226. . 1992. Death and transformation of larval cells during metamorphosis of Anura. Deveiup. Gomth Dzrer. 34:607-611. Youn, B. W., R. E. Keiier, and G. M. Malacinski. 1980. An atlas of notochord and sornite mor~honenesis several anuran in and urodelean amphibians. ' J . %mbryol. Exp. Mmphol. 59:223-247. Youngstrorn, K. A. 1938. Studies on the developing behavior oflAnura.J. Comp. Neuml. 68:35 1-380. Yung, E. 1904. De I'influence du regime alimentaire sur Ia longeur de I'intestin chez les larves de Ranu escuienta. C. R. Hebd. SeancesAcad. Sci. Paris 134:749-75 1. Zaccanti, E, G. Tognato, and S. Tommasini. 1977. Proposta di un rnetodo di valutazione de1 grado di differenziamento sessuale dei corpi genitale in corso di sviluppo di anfibi anuri. Arch. Italiano Anat. EmbrMl. 82:233-252. Zaccanti, F., and C. Vigenti. 1980. Quantitative observations on the development ofBufobufo genital bodies.Arch. Italiano Anat. EmbrMl. 85:353-362. Zakon, H. H. 1984. Postembryonic changes in the peripheral electrosensory systern of a weakly electric fish: Addition of receptor organs with age. J. Comp Neurol. 228: 557-570. Zamachowski, W. 1985. Changes in the body content of water in the cornmon frog, Rana tempmaria (L.), during its ontogenesis. Acta Bwl. C r w . 27: 1 1-18. Zamorano, B., A. Corts, and A. Saiibin. 1988. Ammonia and urea excretion in urine of larval Caudiverbera caudiverbera (L.), (Anura, Amphibia). Cmp. Bwchem. Physwl. 91A:153-155. Zamorano, B., and A. Salibin. 1994. Ionic net fluxes through
-

418

LITERATURE CITED

the in situ epithelia of larval Caudiverbera cuudiverbera (Anura, Leptodactylidae). Alytes 12:135-144. Zettergren, L. D., H. Kubugawa, and M. D. Cooper. 1980. Development of B-celis in Ranapipiens. pp. 177-185. In Phy[ogeny of immunology meviwy, edited by M. J. Manning. Amsterdam: Elsevier/North-Hoiiand. Zimmerman, H., and E. Zimmerman. 1981. Socialverhalten, Fortpflanzungsverhaltenund Zucht der Firberfrosche Dendrobates histrionicus und D. lehmanni sowie einiger anderer Dendrobatiden.Zeit. Kohlner Zoo 24:83-99. Zst, B., and K. E. Dixon. 1975. The effect of W irradiation of the vegetal pole ofxenupusl m i eggs on the presumptive primordial gerrn ceiis.J Embyol. Exp. Morphol. 34:209-220. . 1977. Events in the germ celi lineage after entry of the primordial germ ceiis into the genital ridges in normal and u.v.-irradiated Xencrpus l m i . J Embyol. Exp. Mmphol. 41:3346.

Zweifel, R. G. 1958. Ecology, distribution, and systematics of frogs of the Rana boylei group. Univ. Calqornia Publ. Zool. 54:207-276. . 1964. Life history of Phynohyas venuiusa (Salientia: Hylidae) in Panama. Copeia 1964:201-208. , 1970. Descriptive notes on larvae of toads of the debilis group, genus B u . . AmmicunMus. Nwitates (2407):1-13. . 1977. Upper thermal tolerances of anuran embryos in relation to stage of development and breeding habits.American Mus. Novitates (2617) :1-2 1. Zwiiling, E. 1940. An experimental analysis of the development of the anuran olfactory 0rgan.J Exp. Zool. 84:291-323. . 1941. The determination of the otic vesicle in Rana pipiens. J. Exp. Zool. 86:333-342.

AUTHOR INDEX

Abe, A. S., 204 Abelson, A,, 48 Abraham, M. H., 18 Abravaya, J. P., 314 Accordi, i?., 145, 147 Adams, M. J., 20 Adamson, L., 174 Adamson, M. L., 248 Adema, E. O., 250 Adler, K., 6, 168,230,233,235, 237,238 Adolph, E. F., 21,22, 254 Affa'a, F. M., 248 Aganvai, S. K., 45 Agganvai, S. J., 190, 191 Ahl, E., 297 Ahlgren, M. O., 246,247 Aho, J. M., 248 Aho, T., 266 Aichinger, M., 265 Akin, G. C., 255,272,273 AI Adhami, M. A,, 145 Aiberch, P., 28, 88, 171, 184, 188,279, 288, 289,290, 294 Aibuquerque, R. M., 296 Alcala, A. C., 172, 173, 174, 175, 178, 179, 180, 183, 186.333 Aicocer, I., 181 Alexander, R. D., 232,236 Alford, R. A., 26,48,206,218, 232, 242, 243,246, 248, 250,258, 264,268, 272,
, , , .

Anholt, B., 18 Anholt, B. R., 223, 238,253 Annandaie, N., 19,27,41,44, 228,297,298,329,330 Anstis, M., 37; 41 Antal, M., 154, 155 Archey, G., 173 Am~strong, B., 21 J. Arnauld, E., 138 Arnold, E. hT., 184, 296,334 Arnold, S. J., 79, 206,219,232, 260 Arnoult, J., 297,312,327,332 Arrayago, M.-J., 171 Arthur, W., 171 Aschoff, L., 97 Ashby, K R., 229 Ashley, H., 132 Atkinson, B. G., 106, 132, 190, 191,197 Atoda, K., 178, 180, 181 Attas, E. M., 18 Atweii, W. J, 145 Auburn, J. S., 238 Audo, M. C., 282,283 Augert, D., 235,251 Axelsson, E., 200 Azevedo-Ramos, C., 238 Babalian, A. L., 155 Babbitt, K. J., 226, 263 Baccari, G. C., 29 Bachmann, K., 202 Bacon, J. P., Jr., 200 Baculi, B. S., 101 Bagenal, T. B., 18 Bagnara, J. T., 30, 104, 105, 168,344 Bailey, C. H., 167 Baker, C. L, 197 Baker, G. C., 253 Baiasubramanian, S., 216, 247 Baidwin, G. P., 106 Balinsky, B. I., 24, 28, 104, 105, 106,121,131,140,189 Baiinsky, J. B., 132, 142 Banks, B., 242 Bantle, J. A., 16, 18 Barandun, J., 252 Barbault, R., 182 Barch, S. H., 138 Bard, K. M., 219 Bardsley, L., 271 Bargmann, W., 135, 140 Barker, J., 310 Barker, S. B., 212 Birlcher, E, 241 Bames, M. D., 159 Barreto, L., 200 Barrington, E. J. W., 123, 127, 129,269 Barrio, A., 316 Barry, T. H., 56,60 Basoglu, M., 296 Basso, N. G., 34, 327 Basu, S. L., 142, 144, 145 Bateson, P., 237 Battaglini, P., 248 Baaen, T. F. C., 145 Bauer, A. M., 296 Bavay, A., 171

294' Alford, S. T., 155 Ai Johari, G. M., 24 Aiiey, K. E., 159 Allison, J. D., 161 Ai-Mukthar, K. A. K., 140, 144 Altig, R., 3, 5, 9, 12, 13-14, 16, 17, 18, 19,25,26, 27, 30, 31, 34, 35,36, 37, 38,40, 42, 44,45,46,47,48,49, 50, 63,67, 79, 82,85, 87, 170, 173,178, 184,186, 201,216, 219, 221,227, 228, 229, 231,237, 238, 244,245,248,263,267, 268,280,290,292,296, 297,298,299, 300,302, 303, 305, 307, 308, 309, 310, 321, 326, 339, 340, 342,343,344,345,346, 347 Altman, J. S., 158, 164 Aivarado, R. H., 104, 106, 136, 210,211,212,213 Amboroso, E. C., 171 Amer, F. I., 142, 144 Amiet, J.-L., 5,27,248, 301, 312,313,348 Amos, W., 217 Anderson, J. D., 237 Andrn, C., 21,260,266 Andres, G., 105 Andres, K. H., 165 Andrews, R. M., 171 Angel, F., 296, 334

420 Beach, D. H., 168 Beachy, C. K., 251, 282 Beaumont, A., 42, 106, 132, 146 Beazley, L. D., 168 Beck, C. W., 171, 250,251,282, 284 Beckenbach, A. T., 192 Beckvar, N., 223, 244,292 Beddard, F. E., 79 Beebee, T. J. C., 21, 213,242, 248,253,255,261,271 Beigl, I., 142, 144 Beiswenger, R E., 203, 207, 208,216,229,230, 232, 236,260 Bell, B. D., 173, 180 Belloy Espinosa, D., 171 Belova, Z. A., 247 Bemelmans, S. E., 155 Benbassat, J., 97 Bennett, A. F., 194, 195 Benninghoff, A., 93 Benson, D. G., Jr., 20 Benson, W. W., 236 Bentley, P. J., 106, 132, 136, 198 Bernardini, G., 144 Bernasconi, A. E, 77 Berninghausen, F., 296 Bems, M. W, 21 Berry, P. Y., 31,41, 174, 296, 303,318,334 Berton, J.-P., 140 Bertram, B. C. R., 232 Berven, K. A,, 204, 205, 208, 209, 229,243,251,252, 257, 280, 281,283, 285, 286,288,293 Bhaduri, J. L., 142, 144, 145, 323 Bhalla, S. C., 237 Bider, J. R., 260 Bigaj, J., 135, 136 Binckiey, C. A., 18 Birks, R., 149 Bishop, J. E., 269 Bixby, J. L., 154 Black, D., 150 Black, I. H., 41,203,205,213, 260 Black, J. H., 221 Blackburn, D. G., 170 Blackier, A. W., 143, 144 Blair, W. F., 267 Blanc, C. P., 296, 312, 320, 323, 332 Blaustein, A. R., 209,229, 230, 231,232,233,234,235, 236,237,238, 253,259, 261 Blaustein, L., 242,243,252, 273,275 Bleakney, S., 247 Blight, A. R., 158,226 Blommers, I,. H. M., 228,267 Blommers-Schlosser,R. M., 34, 174, 175, 180,228,267, 296,312,320,323, 332 Blouin, M. S., 243,258,286, 288,292,293,294 Boas, J. E. V, 107 Boatright-Horowia, S. S., 162

AUTHOR INDEX Bobka, M. S., 243,263 Bogart, J. P., 170,290,297 Bogett, C. M., 202 Bokermann, W. C. A., 5,8, 309, 310,315 Bolt, J. R., 288, 290, 292 Boni, P., 248 Boonkoom, V, 106,212 Boord, R. L., 163, 164 Booth, D. T., 196,223 Boothby, K. M., 225 Bork, T., 168 Borkhvardt, V G., 25 Borland, J. R., 22 Bossard, H. J., 132 Boulenger, G. A., 4, 35, 50, 179, 180, 296, 298, 302, 306, 325,326,331,334 Bourne, G. R., 171 Bournoure, L., 143 Bourquin, O., 187 Bourret, R., 5,296, 332 Boutilier, R. G., 197,201 Bowen, S. H., 246,247,268 Boycott, R. C., 328 Bradford, D. F., 21, 181, 192, 194, 195, 203,207, 211, 212,229, 240, 249,260, 263,264 Bradley, L., 34 Bragg, A. N., 12, 21,27,40,41, 218,229,230 Bragg, W. N., 40,41 Braitenburg, V, 161 Branch, L. C., 232 Brandon, R. A,, 5,296 Brandt, F. H., 248 Brattstrom, B. H., 204, 205, 206.207.208,224,225, 229,237' Brauer. A.. 229. 296 Braus,'~.:25 ' Breden, F., 232,258 Breed, W. J., 22 1 Bregman, B. S., 155 Bresler, J., 21,40, 41 Bricout-Berthout,A., 163 Bridges, C. M., 253 Briggs, R., 21 Brink, H. E., 182 Britson, C. A., 200, 246 Brock, G. T., 24, 80 Brockelman, W. Y., 252 Brodie, E. D., 111, 236,237 Brodie, E. D., Jr., 30,49, 85, 211,227,229,232, 236, 243, 251, 260,261,262, 346 Bronmark, C., 226,273,302 Brook, A. J., 267 Browder, L. W., 189 Brown, C. A., 123, 152 Brown, D., 104 Brown, D. R., 154 Brown, F. A., Jr., 209 Brown, H. A., 202,203,206, 208,209,302 Brown, L. E., 245,246 Brown, R. M., 239 Brown, S. C., 210,211,212 Brown, T. E., 213 Brown, W. C., 172, 173, 174, 175,178, 179,180,333 Broyles, R H., 95,96, 97, 100, 190 Brust, D. G., 219,247 Buchan, A. M. J., 146 Buehr, M. L., 143 Bull, C. M., 243,258,261,262, 263,270 Bullock, T. H., 163,209 Burch, A. B., 16 Burd, C;. D., 160 Burggren, W W., 93, 181, 190, . 192, 193, 194, 195, 196, 197, 198, 199,200, 201, 209,214,224 Burns, R. K., 132, 144 Burton, J. A., 296, 334 Burton, l? R., 165 Bus; R. B., 202,203,206,207, 229 Bush, F. E., 145 Busk, M., 201 Byrd, C. A,, 160 Bytinski-Salz,H., 30 Cadle, J. E., 18, 27, 311 Caetano, M. H., 245 Caidwell, J. l?,30,41,217,226, 227, 230, 231, 232,240, 243,250,255,256,261, 262,297,308 Calef, G. W., 223,236,251 Callery, E. M., 181 Cambar, R., 92,133, 137,138 Carnerano, L., 34 Campbell, F. R., 103 Cam~beii. 140 G.. ~ambbell; ., 297, 308,309, J. 322 Campeny, R., 224 Candelas, G. C., 132,252 Cannatella, D. C., 6, 38, 52, 53, 54, 58, 59,61,63, 76, 77, 88,90,304,340 Capranica, R. R, 159, 160, 162, 163 Capurro, L., 174, 179, 316 Caraco, T., 232 Caramaschi, U., 307, 314, 315, 318 Cardoso, A. J., 3 15 Carlson, B. M., 189 Carmona-Osaide,C., 245 Carpenter, C. C., 229,236 Carpenter, K. L., 97, 135 Carr, K. E., 132 Carr, K. M., 16,25,38,49,67, 79,339 Carrizo, G. R., 27,314,334 Carroll, E. J., Jr., 34, 125 Carver, V H., 129 Casinos, A., 224 Castanet, J., 245 Castanho, L. K., 315 CasteU, L. L., 18 Casterh, M. E., 203, 204,205, 206,207 Castle, W. E., 203 Caston, J., 163 Catton, W. T., 168 Cecil, S. C., 251 Cei, J.M.,5,34,35,44, 174, 179, 181, 195, 296, 311, 314, 315, 316, 317, 321, 334 Chacko, T., 56 Chadra, B. G., 243,257 Chang, C.-T., 144 Chang, J. P., 135 Channing, A., 5, 28, 29, 30, 37, 51, 112, 304, 306, 312, 328,334 Chen, A,, 152 Cherdantsev, V G., 38 Cherdantseva, E. M., 38 Chernov, S. A., 296 Chester, M., 153 Chester-Jones,I., 147 Chibon, P., 180, 181 Chieffi, G., 143 Chopra, D. P., 137 Chou, W.-H., 5,296, 322 Chovanec, A., 225,238,243, 263 Christensen, M. S.,201,231, 236,263 Christman, S. P., 213 Cianfoni, P., 147 Ciantar,D., 132, 134, 135, 137, 138 Cioni, C., 158 Claas, B., 158, 164 Clark, A. G., 293 Clarke, B. T., 34,297, 322, 330 Clarke, F. M., 309 Clarke, J. D. W., 152, 155 Claussen, D. L., 202,203, 204, 205,206 Clemens, D. T., 196 Cline, M. J., 95 Cocroft, R. B., 228 Coggeshall, R. E., 153 Coggins, L. W., 140, 142 Cohen, l?P., 131,132 Collins, J. P., 5, 194, 204, 208, 215, 229,243,249,250, 257, 279, 280, 281,285, 291,296 Comer, C., 158, 160 Conneil, J., 229,232,235,249 Constantine-Paton, M., 149, 153,160 Cooke, A. S., 242 Cooper,A. K., 132, 181, 183, 185 Cooper, E. L., 97, 101, 103 Cooper, K. W., 34 Cooper, R. S., 24, 58 Capenhaver, W. M., 93 Cordier, R., 135, 136, 138 Corkran, C. C., 296 Corless, J. M., 166 Cornell, T J., 233,234,237 Cortwright, S. A., 200, 274 Corvaja, N., 155 Costa, H. H., 198,216,247 Cowden, R. R., 102 Cowles, R. B., 202 Cox, T C., 212 Crawshaw, L. I., 203, 205,206, 207,224,260

A U T H O R INDEX Crespo, E., 295 Cronin, J. T., 262 Crossland, M. R., 218,246,247 Crowder, W. C., 21, 193, 194, 197,198,199,200 Cmce, W L. R., 155, 160 Crump, M. L., 23, 172, 184, 206,208,211,217,218, 219,232,242,244,246, 247,249,251,262,265, 266,268,303 Cuellar, H. S., 321,322 Cuiien, M. J., 168 Cuey, D. D., Jr., 245,247 Cummins, C. I?, 21,249,260 Cupp, P. V, Jr., 203,204,205, 206,207,224 Curtis, S. K., 103 Cusimano-Carollo,T., 35,45 Czechura, G. V, 297 Czh, G., 154 Czolowska, R., 143 Da Cniz, C. A. G., 303,308, 310,311,314,321 Dainton, B. H., 205 Daniel, J. C., 323 Daniel, J. E, 16 Danos, S. O., 221 DarneU, R. M., Jr., 197 Das, I., 107 Das, S. K., 77 d'hcanio, E, 155 Dash, M. C., 245,246, 256, 258 Daua, M., 131, 138 Davenpoa, C. B., 203 Davidson, E. H., 24 Davidson, M., 21 Davies, M., 34, 55, 175, 297, 310,324 Davis, M. R, 153 Davis, W. M., 151 Dawes, E. A., 158, 164 Dawson, J. T., 232,233,235 Dawson, W. R., 194 De Albuja, C. M., 175 De Arajo, M. C., 217 De Bavay, J. M., 174 De Beer, G. R., 55, 56, 59,61, 63,88 DeBenedictis, P. A., 235, 236, 273 De Boer-Van Huizen, R, 155 Debski, E. A,, 160 De Capona, M.-D. C., 312 De Carvalho, A. L., 5, 9,28,31, 33, 37,41,50,296,323 De Carvalho e Silva, S. P., 171 De Cerasuolo,A. M., 181 Decker, R. S., 154 Degani, G., 210,223,269 De Gueldre, G., 168 DeHaan, R L., 93 De Jongh, H. J., 24, 53, 56,58, 61,66,67,68,69, 70, 75, 81,82, 84,216 De Lahunta, A,, 153 De Ia Riva, I., 56, 315 Delbos, M., 135, 137, 140, 143 Deffino, G., 105 Deiidow, B. C., 249 De1 Pino, E. M., 22, 171, 172, 173, 175, 177, 178, 180, 181,186 Delsol, M., 93 Dempster, W T., 180 Denis, H., 58 Denton, J., 231,261 Denver, R. J., 225 De Prez, G. R., 187,229 De Quiroga, B. G., 197 De S, R. O., 36, 55, 56, 57,63, 64, 78,91,298, 307, 309, 317,327,331 De Saint-Aubain, M. L., 104, 106,196,197,198 Desser, S. J., 248 Deuchar, E. M., 142 D e d ; J., 42 Deutsch, M. J., 97 De Villiers, C. G. S., 174, 175, 179,181,186,327 Devis, R. J., 140, 142 De Vlaming, V L., 202,203, 205,206,207,229 De Waal, S. W P., 136, 138 . De Witte, G. F., 296 Deyrup, I. J., 132, 133 Diamond, J. M., 186,209 Diaz, N. F., 296, 316, 317, 318, 334 Diaz-Paniagua, C., 216,243, 248,267,269 Di Bemardino, M. A,, 143 DiBona, D. R., 136 Dickman, M., 223,248 Diener, E., 97 Dieringer, N., 159 Diesel, R., 171 Diea, T. H., 212,213 Di Grande, F., 142, 144 Dionne, J.-C., 221 DiStefano, R. J., 12 Ditrich, H., 138 Dittrich, S., 135 Dixon, J. R., 140, 142, 143, 267,297,305,327 Dodd, J. M., 125,137 Dodd, M. H. I., 137 Donnelly, M. A., 9, 34, 38,263, 265,296,306,311, 321, 322,334 Donner, K. O., 166, 167 Donoso-Barros, R., 316 Dornfeld, E. J., 17 Doughty, I?, 171 Douglas, R. M., 26 Dowling, J. E., 167 Downie, J. R., 183,218, 219, 230,246,255 Doyle, M., 199 Dressler, G. R., 171 Drewes, R. C., 28,48, 50, 228, 312,313,327 Dreyer, T. F., 55 Drysdale, T. A., 34 Dubois, A., 40,298, 329 Dudley, R., 225 D u e h a n , W E., 5 , 6 , 8 , 9 , 10. 11,28,29,34, 36,50, 54, 107, 112, 113, 123, 172, 173, 179, 180, 182, 184, 185, 190,206,207,215, 216,220,229, 230,265, 267,290,291,296,299, 300, 304, 305, 307, 308, 309, 310, 311, 314, 316, 321 Dugs, A., 4, 53, 55, 107 Duke, J. T., 192, 193, 196, 198 Duke, K L., 144 Dumas, P. C., 270 Duncan, J. T., 93 Dunham, A. E., 243 Dunlap, D. G., 29,237 Dunlop, S. A., 168 Dunn, E. R, 41,187,308 Dunn, R. F., 163 Dunson, W A,, 202,203,211, . 212,213,214,229,241, 249,267 Du Pasquier, L., 104 Dupr, R. K., 199,203,205, 206,224,238,260 Dutta, S. K., 50 Dziadek, M., 143 Dzieduszycka, S., 328 Eakin, R. M., 34, 145 Eaton, R. C., 158 Eaton, T. H., 288, 290 Ebbesson, S. O. E., 152, 154, 155, 156, 157, 158, 159, 160 Ebenman, B., 243 Eberhard, M. L., 248 Echeverra, D. D., 5, 25,40,45, 50,107,109,144 Eddy, E. M., 140,143 Edelman, A., 255 Edenhamm, E, 226 Edgeworth, F. H., 53, 55,67, 68,69, 70, 72, 74, 75, 76,
77 ..

42 1 Estrada, A. R, 172 Etkin, W., 9, 145, 180, 191, 208,346 Exbrayat, J. M., 182 Faber, J., 8, 33, 53, 70, 77,92, 93, 131, 136, 137, 145, 146,147,327 Fabrezi, M., 55, 56 Faier, J. M., 35,49 Faivovich, J., 27, 247, 297, 314, 334 Falconer, D. S., 284,286,287, 289,291,293 Fang, H., 182 Fanning, S. A., 296 Faragher, S. G., 240 Farel, P. B., 153, 155,226 Farlowe, V, 223, 247 Farrar, E. S., 146 Farrell, A. P., 93 Farris, J. S., 5, 9, 50, 296 Fauth; J. E., 254,274,275 Fawcett, D. W., 142 Feder, M. E., 88, 191, 192, 193, 194, 195, 196, 197, 199, 200,209,223,224,225, 226,243,261 Fei, L., 319 Feinsinger, P., 248 Felsenstein, J., 87 Feminella, J. W., 261 Feng, A. S., 159, 160, 161, 163 Fenster, D., 255 Ferguson, D. E., 168 Fennin, E, 171 Fernndez, K., 5 Fernndez, M., 5 Fernndez-Marcinowski, K., 5 Ferns, M. J., 155 Ferrari, L., 210 Fetter, R. D., 166 Figge, F. H. J., 197 Figiel, C. R, Jr., 261 Filipello, A. M., 40 Filipski, G. T., 18 Fink, W. L., 171, 188 ~ i o n t o Lopez, L. E., 45, 50, de 107 Fisher, K C., 203,205 Fisher, M. D., 160 Fisher, R A,, 236 Fishwild, T. G., 233, 234, 235, 236 Fite, K. V, 160, 161, 162, 166 Flatin, J., 93 Fletcher, K., 191 Flock, A., 163, 164 Flores-Nava, A,, 21, 22 Flower, S. S., 297 Floyd, R B., 202,203,204, 205,206,207,260 Ford, L. S., 6, 52, 53, 54,61, 63,77,90 Ford, T. D., 221 Forehaiid, C. J., 153 Forge, P., 182 Forman, L. J., 97 Formanowicz, D. R., Jr., 30, 171, 211, 232, 236,243, 251,260,261,262,263

Edwards, M. L. O., 21 Egami, N., 140,143 Ehmann, H., 175 Ehrl, A., 175, 326 Eichler, V B., 154 Eiger, S. M., 205,214,224 Eiselt, J., 60 Eisworth, L. M., 164 Ekblom, P., 138 Ekman, G., 93, 115 Eldred, W. D., 168 Elepfandt, A., 160,164 Elias, H., 30,104 Elinson, E. E, 29,34, 172, 175, 177, 178, 179, 181, 182, 184,186,188 Emerson, S. B., 67,292,293, 294 Emlet, R. R., 171 Engel, W., 144 Engels, H. G., 136, 138 Engermann, W.-E., 296 Epple, A., 145 Erasmus, B. de W, 201 Erdmann, R., 93 Escher, K., 163 Escobar, B., 171, 172, 175, 178, 180 Estes, R., 3

422 Formas, J. R, 5, 174, 179,227, 316,317,318,334 Forunan, J. R., 50 Foster, M. S., 201, 230, 231 Foster, R. G., 162, 168 Foster, S. A,, 191 Fouquette, M. J., Jr., 308 Fox, H., 72, 104, 105, 107, 119, 121, 123, 125, 129, 131, 132, 133, 135, 136, 137, 146,238,288,346 Foxon, G. E. H., 97, 135 Franchi, L. L., 144 Francis, M. G.. 200,260 Frank, B. D., 167 Frank, E., 155 Frank, G., 96, 103, 104, 135, 136 Frankino, W A., 247 . Fraser, E. A., 132 Freda, J., 211,212,213,214 Freeman, J. A,, 159 Frieden, E., 97, 129 Friedman, B.. 158 Fnedman, R. T., 104 Friet, S. C., 195 Fritsche, R., 93, 214 Fritzsch, B., 158, 162, 163, 164 Frogner, K. J., 31,322 Frohiich, J., 104 Frost, D. R., 27, 172,298 Frost, J. S., 331 Frost, S. K., 30 Frye, B. E., 146 Fujisawa, H., 160 Fder, P. M., 158 Funkhouser, A., 191,210 Gale, E. A,, 28, 184 G d , J. G., 142, 175 Gdardo, J. M., 29 Gdien, L., 92, 132 Gans, C., 168 Garca, G., 55, 56 Gardener, L. W., 137 Gardiner, J., 6 Gam, R, 104 Gascon, C., 20, 260, 262, 265, 266,275,302 Gates, W. R., 50 Gatherer, D., 177 Gatten, R. E., Jr., 191,192, 193, 194,195,209 Gaupp, E., 53,56 Gavasso, S., 238,263 Gavrila, L., 247 Gaze, R M., 160, 167 Gee, J. H., 224 Gegenbaur, K., 115 Gehrau, A., 317 Grard, P., 135, 136, 138 Gerhardt, E., 104, 115 Gerschenfeld, H. M., 162 Gesner, C. von, 3 Geyer, G., 140 Ghate, H. V, 260,263 Giarretta, A. A,, 309, 315 Gibbons, J. W., 262,264 Gibley, C. W., Jr., 135 Giesel, J. T., 171 Giies, M. A., 18

A U T H O R INDEX Gi, D. E., 180,208,209, 272, 280,291 Giliiam, J. F., 243, 269 Gines, H., 326 Giorgi, P. i?,137, 143 Gipouloux, J.-D., 92, 132, 137, 138,143 Girard, C., 137, 143 Gitlin, D., 174, 175 Glaw, R., 174, 296, 320,323, 332 Gluecksohn-Waelsch, S., 171 Goater, C. P., 243, 248 Godinho, H. P., 204 Goette,A., 55,93,107,112,115 Goicoechea, O., 184 Goin, C. J., 3, 175, 178 Goin, O. B., 3 Goldacre, R. J., 221 Golden, D. W., 95 Goiichenkov, V A., 104 Gollman, B., 2, 34, 181, 229, 259,295 Goiiman, G., 2,34, 181,229, 259,295 Gomez, M., 132, 252 Gona, A. G., 159 Goniakowska-Witalnska, L., 106,107 Gonzaiez, A., 154,159 Goodyear, C. P., 238 Gordon, F. J., 12 Gordon, J., 167 Gordon, M. S., 132,210,213 Gorham, E., 213 Gorlick, D. L., 150 Gorner, P., 164 G o r d a , S., 173, 303 Gosner, K. L., 8, 10-11, 12,28, 41,42,49, 50, 53,92, 131, 142, 143,144,145,171, 178, 189,202,203,205, 213,215, 240, 260, 281, 339,340,343,345,347 Gotte, S. W., 23 Gouras, P., 167 Gradweli, N., 38,40,48, 49, 53, 55,56,57, 58,63,64,65, 68, 69, 70, 71,72, 73, 74, 75,80, 81,82,83,84,85, 86, 87, 88,90, 109, 112, 113,119,121, 198,216 Grafen, A,, 235 Graham, J. B., 201 Grandison, A. G. C., 28,50, 173,182,240,303,304 Grant, P., 167, 168 Grant, S., 168 Grass, P.-P., 142 Grassi Milano, E., 145, 147 Gray, E. G., 168 Gray, J., 226 Gray, P., 138, 172, 173, 180, 184,307 Graybeai, A., 90, 304 Graziadei, P. P. C., 165 Green, N. B., 297 Greenwald, D. J., 136 Greenwood, P. J., 236 Greii, A., 115 Greven, H., 121 Griffin, J. R., 213 Griffiths, I., 5,9,27, 28, 31, 33, 37, 41, 50, 53, 89, 123. 125, 127, 129; 130,296, 323,326 Grifiths, R. h, 200,223,231, 238,243,253,255,271 Grigg, G. C., 209, 260, 310 Griiiitsch, B., 40,42, 51, 109, 331 Grillitsch, H., 42, 296 Grillner, S., 226 Gritis, P. A., 298, 328 Grobstein, P., 155, 158, 160, 238 Gromko, M. H., 255 Grover, B. G., 159 Grubb, J. C., 260 Gruberg, E. R., 160 Gmse, P., 171 GNsser, O.-J., 168 GNsser-Cornehis, U., 159, 168 Gudynas, E., 321 Gugiielmotti, V, 160, 161, 165 Guib, J., 173,297, 313 Guilford, T., 232 Guiette, L. J., Jr., 171 Guraya, S. S., 144 Gutzeit, E., 4 Gutzke, W. H. N., 253 Guyer, C., 263,265 Guyer, M. F., 18 Guyetant, R., 21,254 Haas, A., 53, 54, 55, 56,57, 58, 59,61,62,63,65, 68,69, 70, 71, 72, 73, 74, 75, 77, 80,81,90,91, 181 Haddad, C. F. B., 315 Haertel, J. D., 50 Hah, J. H., 135 Hailman, J. P., 167, 237 Hairston, N. G., 254 Hakim, J., 132, 138 H d , B. K., 58 Hall, J. A., 24,230, 231, 234, 235 Haii, J. C., 161 H d , R. W., 132 Hamburger, V, 149 Hamilton, L., 135 Hamilton, W. D., 232,236,291 Hammerman, D. L., 107, 112 Hammerschlag, R., 224 Hampton, S. H., 41 Hanke, W., 145, 146 Hanken, J., 16, 17, 24, 55, 58, 59,62,63,66, 77,78,88, 171, 178, 181, 185, 188, 194 Hardy, J. D., Jr., 179 Harkey, G. A,, 21, 22, 205, 208, 260 Harpur, R. P., 252 Harris, R. N., 247, 248,250, 277,281,282,283,284, 288,292 Harrison, J. D., 216 Harrison, P. H., 216 Harrison, R. G., 164 Hart, M. W., 171 H a d , D. L., 293 Harvey, I. F., 261 Harvey, J. M., 210,260 Harvey, P. H., 88, 236 Hassinger, D. D., 235,236 Hastings, D., 209 Hathaway, E. S., 204 Hausen, P., 165 Hauser, K. F., 149, 159 Havenhand, J. N., 171 Hawkins, C. P., 261 Hay, J. M., 54,90 Hayes, B. P., 238 Hayes, L., 262 Hayes, S., 41 Hayes, T. B., 105, 146,249 Hazard, E. S., 111, 190 Hearnden, M. N., 242,247,276 Heasman, J., 143, 144 Heathcote, R. D., 152 Heatwole, H., 174, 203,204 Hedrick, J. L., 34 Hegner, R. W., 21,248 Heisler, N., 201 Heiff, O. M., 107, 110,112, 194,201 EIenderson, R. W., 173, 174 Hendricks, F. S., 223,247 Hendrickson, A. E., 168 Hendrickson, J. R., 221 Hennen, S., 143 Henrickson, B.-I., 260, 261,262 Hensley, F. R., 250,281,282, 283,284 Hero, J.-M., 246, 247,262,265, 267,291,296,309,325 Hron-Royer, 3 , 4 , 4 9 Herreid, C. F., 11,229, 251,260 Herrick, C. J., 158, 159, 160, 161 Hessler, C. M., 230, 231 Hetherington, T. E., 59, 60, 302 Heusser, H., 27,218,246, 247 Hewitt, J., 298 Hews, D. K., 236,238,243, 253,261,262 Heyer, W. R., 5,27,44,48,49, 107, 109, 111, 112, 113, 121, 173, 182, 196,200, 216, 223,227, 228,240, 247, 262,265,267,268, 296, 305, 307,309,310, 311, 315, 316, 317, 322, 332 Hibiya, T., 121 Higgins, G. M., 58 Hildrew, A. G., 48 Hill, C. G., 203,205,206,260 Hill, S., 57 Hdis, D. M., 34, 297, 331 Hlllman, D. E., 166 Hinckley, M. H., 4 Hirakow, R., 93 Hird, D. W., 248 Hirsch, G. C., 138 Hisaoka, K. K., 21 Hlastala, M. P., 196 Hochachka, P. W., 192 Hoegh-Guldberg, O., 171 Hoff, K., 5,48, 55, 56, 57, 87, 88,91, 109, 119,216,219,

A U T H O R INDEX 224,225,226,239, 243, 288 Hohann, M. H., 160 Hokit, D. G., 229,231,237, 253,259 Holder, N. J., 153 Holley, A., 160 Hoiiyfxld, J. G., 97, 135, 167 Holomuzki, J. R., 248 Honda, E., 110 Honegger, R. E., 314 Hood, D. C., 167 Hoogmoed, M. S., 179,311 Hoppe, D. M., 202,203,204, 205 Hora, S. L., 27 Horat, E, 243 Honuchi, T., 245 Horseman, N. D., 247 Hoaon, J. D., 97, 100, 101, 103,135,136 Hoaon, P., 142 Hoskins, S. G., 167 Hosoi, M., 49 Hota, A. K., 246,256,258 Houillon, C., 92, 132 H o u r h J., 21, 104, 131, 182, 188,254 Honrer, R. O., 17 Hoyer, M. H., 101 Hoyer, W. J., 239 Hoyt, J. A., 58 Hraouibloquet, S., 182 Hrbiek, J., 236 Hsu, C.-Y., 144 Hu. S. Q., 319,320, 330 Huber, I., 237 Huey, R. B., 204,205,209,261 Hughes, A., 29, 154,155,168, 180,181,182,227 Hulsebus, J., 146 Humphries, A. A,, Jr., 175 Hurley, M. B., 135, 137 Huxhke, E., 107 Hutchinson, T.. 3, 6, 203,204 Hutchison, V H., 190, 205, 206,209,260 Hudey, T. H., 62 I a d o , J. M., 93 I&y-Oka, A., 127, 130 Ideker, J., 261 Ij.K. L., 140, 143 ikenishi, K., 143 In,w, F., 5,6,9, 20,25,27, R 29, 30, 33, 35, 41, 45,46, 48,49,50, 109, 123, 175, 182, 184, 190,216,223, 227,229, 240,269,279, 295,296,297, 298, 302, 303, 304, 318, 319,320, 321, 322, 328, 329, 330, 331,333,334 wennan, R. L., 171 hgie, D., 168,237,239 iqieron, P. M., 145 hpm, . J., 187 G lngram, V. M., 96,97,190,191 Iwhchi, T., 125 Iranova, N. L., 21, 247 Lshizuya-Oka, A., 127 Ison, R. E., 94 Istock, C. A,, 292, 293,294 Ito, S., 140 Iwasawa, H., 138,140,142, 143,144,145 Izecksohn, E., 5, 30, 174, 303, 310,311,314 Izquierdo, L., 317 Jablonski, D., 294 Jackson, D. C., 201 Jackson, G. D., 26,248 Jackson, J. F., 254, 3 14 Jacobson, A. G., 93 Jacobson, C. M., 55 Jacobson, M., 168 Jacoby, J., 158, 164 Jaeger, C. B., 166 Jaeger, R. G., 167, 237,240, 254 JaEee, O. C., 133, 135 Bgersten, G., 290 James, V G., 22 Jameson, D. L., 173, 180 Jande, S. S., 163 Janes, R. G., 129 Janies, D. A,, 9 Jasienski, M., 237,259 Jaslow, A. P.. 162 Jayatilaka, A. D. P., 146 Jennings, D. H., 182 Jennings, R. D., 28,29, 295, 297,331 Jensen, F. B., 216 Jenssen, T. A,, 223,247 Jia, X. X., 196 Jiang, T., 160 Johansen, K., 190 Jolui, K. R., 255 John-Alder,H. B., 243 Jolmson, D. S., 213 Jolmson, E. A,, 271,275 Johnson, S. R., 136 Johnston, G. F., 5,9, 12, 13-14, 19, 27, 31, 34,40,42,44, 48,49, 50,63,82, 87, 170, 173, 184, 186, 216,227, 228,244,280,290,297, 298,299,300,342,344, 345,346,348 Johnston, L. M., 223,247 Jolivet-Jaudet, G., 147 Joliie, M., 72, 74 Joly, E, 235,251 Jonas, L., 135 Jones, R. E., 172 Jones, R. M., 132 Jargensen, C. B., 140 Jordan, F., 226,263 Jordan, H. E., 97 Jorquera, B., 131, 175, 181, 317 Joseph, B. S., 154, 159 Junc, F. A., 173, 178, 184 Jungfer, K.-H., 217, 219, 220, 247 Junqueira, L. C. U., 104 Jurgens, J. D., 58 Just, J. J., 97, 106, 136, 190, 191, 197, 201, 208,210, 25 1 Justis, C. S., 168, 238 Kahn, T., 5 1 Kaiser, H., 173, 178, 184, 305 Kalt, M. R., 140, 142, 143 Kam, Y.-C., 217,219, 332 Kamat, N. D., 223 Kamel, S., 243,251 Kamimura, M., 143 Kamprneier, O. F., 101 Kanamadi, R D., 143 Kaplan, M., 41 Kaplan, R. H., 225,246,253, 258,261 Karlson, P., 132 Karlstrom, E. L., 203,204 Karnovsky, M. J., 16 Kartn, I., 237 Kashin, S., 226 Kasinathan, S., 143 Katagiri, C., 34 Kats, L. B., 240, 260, 263, 264 Katz, L. C., 230,231 Katz, M. J., 280 Kaufn~ann, C., 188 T. Kaung, H. C., 42, 46, 48, 146 Kawada, J., 212 Kdwai, A., 104 Kawamura, T., 50 Keating, M. J., 160, 168 Kehr, A. I., 34,250,257,262, 269,296,327 Keiffer, H., 4 Keiier, R., 24 Kelley, D. B., 150 Keiiy, C. H., 258 Kelly, D. E., 168 Kelly, J. P., 244 Kernali, M., 160, 161, 165 Kemp, A., 16 Kemp, N. E., 58,130 Kermy, J. S., 49, 56, 80, 81, 84, 109, 112, 113, 216,243, 244,327 Kenward, R. E., 232 Kerr, J. B., 142, 143 Kerr, J. G., 115, 129 Kessel, R. G., 24, 34, 143 Kesteven, H. L., 55 Kett, N. A., 155 Khan, M. S., 12, 15, 21, 41,44, 109,296 Khare, M. K., 223 Kiditer, E., 156, 160, 162 Kiesecker, J., 260 Kilham, P., 213 Kindahl, M., 135, 138 Kindred, J. E., 119 King, M., 310 Kingsbury, B. F., 163 Kinney, E. C., 213 Kinney, S., 229, 260 Kinoshita, T., 92 Kirtisinghe, P., 31. 44, 322, 323, 330 Kiseleva, E. I., 238 Kisseii, R. E., Jr., 200, 246 Kisteumacher, G., 315 Kiein, S. L., 165 Kiimek, M., 247 Kiuge, A. G., 5,9, 50,90,205, 296

423
Kluger, M. J., 209 Kiymkowsky, M. W., 17 Knudson, C. M., 95, 135 Kobayashi, M., 50, 140, 142, 144 Kok, D., 219,231 Koilros, J. J., 8,9,42,46,48, 92,94, 151, 159, 160, 166, 168, 189, 194, 198, 224, 229,281,288 Kopsch, F., 53, 64, 127 Korky, J. K., 40,41,49,295, 297,339 Koshida, Y., 245 Koskela, P., 240 Kotani, M., 143 Kothbauer, H., 229 Kotler, B. P., 242 Kotthaus, A., 55 Kraemer, M., 55, 59,61 Kramer, D. L., 201 Kramer, H., 20 Kratochwill, K., 56, 81, 84, 107, 112,113 Kress, A., 140 f i g e l , P., 21, 174 Krujitzer, E. M., 55 Kruse, K. C., 200,260 Ku, S. H., 137 Kubin, T. M., 309 Kunst, J., 136, 138 Kunte, K., 228 Kuna, A., 125 Kmle, H., 154 Kupferberg, S. J., 238,243, 244, 245, 248,253, 263,270, 273 Kuramoto, M., 332 Kuthe, K. S. M., 216 Kuzmin, S. L., 243 Lainson, R., 248 Laird, A. K., 280 Lajmanovich, R C., 216,247 Lamar, W W., 315 . La Marca, E., 297, 306 Lamb, A. H., 155 Lambem, G. A., 248 Lambiris,A. J.L.,28, 179, 181, 296,306 Lamborghni, J. E., 154 Lametschwandtner,A., 28,93, 138,140 Larnotte, M., 5, 9, 35, 140, 142, 144, 173, 179, 181, 183, 254,297, 301, 312, 313, 323, 326, 327, 328, 329, 331,333,334 Lancaster, J., 48 Lande, R, 285 Langone, J. A., 31,34,303 Lannoo,M. J., 5, 18,27, 88, 107, 112, 113, 121, 163, 164. 198, 217. 227. 237, . . 238; 240; 308' Lanza, B., 296,302,306, 325, 326,331,334 Largen, M. J., 313 Larseii, O., 158, 159, 163 Larson, P., 91 Larson-Prior,L. J., 160

424 Lataste, F., 4 Latsis, R. V., 135 Lauder, G. V., 90,288 Laurens, H., 237 Laurila, A,, 253,266 Laviiia,E. 0.,5,9,27,31,34, 37,41, 55, 56, 57,91,295, 298,318,321,327,346 Lawler, K. L., 262 Lawler, S. P., 243,263,275, 276 Lawson, D. P., 28,228,301 Lzr,G. Y., 160 Leaf, A., 136 Lebedinskaya, I. I., 28 Lee, A. K., 324 Lee, J. C., 12, 298 Lee, R. K. K., 158 Lefcoa, H. G., 205,209,214, 224,238,253 Leibovitz, H. E., 21,247 Leimberger, J. D., 264 Leips, J., 250,281,282,283, 284 Leloup-Hatey, J., 147 Lemckea, F., 207 Lenfant, C., 190 Leon-Ochoa,J., 316 Le Quang Trong, Y., 104 Lescure, J., 5,9,35, 183,215, 230,306,308 Leucht, T., 238 Levine, R. L., 160 Lewinson, D., 121 Lewis, E. R., 162, 163 Lewis, M. A., 194 Lewis, S., 159,208 Lewis, T. L., 145 . Lewis, W M., Jr., 201 Lewontin, R. C., 237 Leydig, E, 4, 104 Li, J. C., 245 Licht, L. E., 252,256,260,272 Licht, P., 105, 195 Lieberkund, I., 24,34 Liern, K. F., 29, 30, 149 Ligname, A. H., 15 L i e , R. D., 20 Liiiywhite, H. B., 202,204,205, 207 Lima, A. P., 265 Limbaugh, B. A., 50 Lin, C. S., 245 Lin, J.-Y., 5, 296, 322 Lin, W., 159,160 Lindquist, E. D., 302 Ling, R. w., 210 Linsenmair, K. E., 238,261 Linss, W., 140 Lips, K. R., 21,296 List, J. C., 21, 58, 78 Little, G. H., 132 Littlejohn, M. J., 5, 174, 179, 325 Liu, C.-C., 5,9,296,297,298, 319,320,322,330 . . Liu, H., 27 Lofts. B.. 140. 142. 144 Lombard, R ., 162,163 Loor-Vela, S., 177 Lopez, K., 140, 142, 143, 144, 145

AUTHOR I N D E X
Lopez, P. T., 227 Loschenkohl, A., 223,269 b m p , S., 34, 105, 191, 197 Lowe, D. A., 151,160,164 Lucas, E. A., 203,205,206, 207,260 Luckenbiil, L. M., 42,46,48,49 Ludwig, D., 251,282 Lugand, A., 181, 184 Lum, A. M., 164,238 Luna, L. G., 20 Luther, A., 55,68 Lua, B., 170, 174, 179,180, 229,290,291 Lyapkov, S. M., 215 Lynch, J. D., 41, 50,90,302 Lynch, K., 25 Lynn, W. G., 55, 172, 174, 178, 180,181,182,188,255 Macgregor, H. C., 172 . Mackay, W C., 211 Maeda, N., 5,296,332 Maetz, J., 104 Maglia, A. M., 55 Magnin, E., 55 Magnusson, W. E., 246,247, 265,291 Mahapatro, B. K., 246,258 Maher, S. W., 221 Main, A. R., 202,324 Majecka, K., 267 Majecki, J., 267 Malacinski, G. M., 21 Maiashichev, Y B., 25 . Maiik, S. A., 109 Maione, B. S., 183 Malvin, G. M., 196,197 Manahan, D. T., 184 Maness, J. D., 173,204,206, 209 Maness, S. J., 172 Mangia, F., 96 Mangold, O., 115 Maniatis, G. M., 96,97 Manis, M. L., 203, 204 Mann, R., 18 Mannen. H.. 153 ~ a n n i n gM J., 24,97, 100, ,. 101,103,136 Manteifel. Y B.. 238., 253., 263 . ~ a n t e k lG., 237 , Marcus, E., 104, 115 Marcus, H., 115 Marescakhi, O., 142, 144 Margalit, J., 242,243,252,273, 275 Marian, M. P., 199,208, 209, 224,251,283 Marie, Sister Ared De, 182 Marinelli, M., 45 Maritz, M. F., 26 Marks, J. C., 294 Marlier, G., 213 Marrot, B., 92 Marshall, E., 209,260 Marshall, G. A., 21,247 Martin, A. A., 5, 132, 181, 183, 185,296,310,323,324, 325,334 Martin, M. S., 92 Martin, W., 267 Martinez, I., 21,22,28 Matesz, C., 163, 164 Mathies, T., 171 Mathis, U., 237,263 Matsui, M., 5, 296, 332 Matsurka, Y., 135 Matthews, L. H., 171 M a m a , H. R., 168 Matz, G., 22 Maurer, F., 107, 115 May, R. M., 278 Mayhew, W. W., 202 Maynard Smith, J., 232 McCann, R. G., 246 McCarrison, R., 249 McClanahan, L. L., 132, 181, 252 McClendon, J. F., 189 McCoiium, S. A,, 30,264 McCormick, C. A,, 158,162, 163 McCranie, J. R., 296, 302 McCutcheon, F. H., 97, 190, 191 McDearman, W., 245,248,292 McDiarmid, R W., 1, 16, 18, 36,38, 173, 179, 184, 187, 201,217,230, 231, 232, 266,291,303,309,310 McDonaid, D. G., 211,212, 213,214 McEdward, L. R., 9 McIndoe, R., 105, 106, 196, 197 McLachlan, A. J., 200,241,242, 243,248,251,291 McLennan, I. S., 155 McMurray, V. M., 159 McPeek, M. A., 240 Mecham, J. S., 171, 172 Medewar, P. B., 171 Mhely, E. G., 5 Meiiicker, M. C., 107, 110, 112 Meltzer, K. H., 192 Mendeii, L. M., 155 Mendelson, J. R., 111,297 Menke, M. E., 202,203,205, 206 Menzies, J. I., 44, 50, 71 Merchant-Larios, H., 140, 144 Merriil, E. G., 167 Mertens, R., 49,296, 301, 304, 334 Meseguer, J., 97, 136, 181 Metter, D. E., 182, 302 Meyer, D. L., 160 Meyer, M., 104 Michael, M. I., 55, 56,60, 70, 72, 73, 74, 75, 77,92, 106, 133, 135,137,145,163 Michaiowski, J., 50 Mijares-Urrutia, A., 304, 306, 309 Millamena, O. M., 22 Miiiard, N., 97,98,99, 136 Mills, K. S., 191 Miranda, L. A., 21,205 Mishra, P. K., 245,258 Mitcheii, S. L., 48, 186 Moaiii, R., 196 Mobbs, I. G., 249 Mocquard, F., 298 Moffat, A. J. M., 163 Moger, W. H., 187 Mohanty, S. N., 246 Mohanty-Hejmadi,P., 30, 50, 323 Mohawald, A. P., 143 Monayong Ako'o, M., 131 Montgomery, N., 159,160 Moody, A., 104,210,211,212 Mookerjee, H. K., 77 Moore, J. A., 177, 186, 189, 202,205,209 Moore, M. K., 261 Moran, N. A,, 279 Moreira, G., 200,265 Morgan, B. E., 182 Morgan, M. J., 161 Morin, P. J., 230,243,248, 254, 260,261,263,269,271, 273,274,275,276 Morris, J. L., 140 Moss, B., 190, 191 Moulton, J. M., 158 Moursi, A. A,, 77 Mous; J. D., 16,178 Mudry, K. M., 163 Mufii, S. A., 41,44 Mdally, D. l?,229 Mder, F., 106 Mumme, R. L., 266 Munn, N. L., 168,239 Munro, A. F., 132,212 Muntz, L., 29,226 Muntz, W. R. A,, 167 Murphy, A. M., 224 Murray, D. L., 258 Muto, Y., 21 Mwaiukorna, A,, 190,196,197 Myant, N. B., 191 Myers, C. A., 217, 246, 247 Myers, C. W., 305 Nachtigall, W., 34,49 Nadal, E., 172 Nadkarni, V. B., 142 Nagai, Y., 245,247 Naito, I., 136, 137, 140 Naitoh, T., 123,293 Nakata, K., 255 Nanba, H., 146 Nathan, J. M., 22 Nation, J. L., 16 Naue, H., 115 Neary, T. J., 159, 161, 162, 163 Nelson, C. E., 200, 274, 321, 322 Nelson, G., 90 Netsky, M. G., 161 Nevo, E., 210 Newrnan, R. A., 243,252,258, 281,282,284,285,286, 288,293 Nguenga, D., 217 Niazi, I. A,, 45 Nichoiis, T. J., 142 Nichols, H. W., 22 Nichols, R. J., 3, 5,49, 50 Nicholson, C., 153 Nicoii, C. S., 283

A U T H O R INDEX

425 Reinbach, W., 55 Reiss, J. O., 55,62, 63 Rengel, D., 140, 144 Reques, R., 208,242,243,250, 251,284 Resetarits, W J., Jr., 266, 275 . Retallick, R. W., 296, 325 Reuter, T., 166, 167 Reyer, H.-U., 243,252 Reynolds, R P., 12,23 Reynolds, W A., 203, 205,206, . 207,260 Reynolds, W. W., 203,204,205, 206,207 Rhodin, J. A. G., 28,93 Rice, T. M., 18 Richards, C. M., 255 Richards, S. J., 29, 34, 37,243, 261,262,263,297,310 Richmond, J. E., 210 Richmond, N. D., 201 Richter, S., 132, 133, 135, 136, 137,138,140,144,174 Ridewood, W G., 53,55,56, . 57,71,75,76-77 Riedel, C., 196 Rieger, R. M., 279 Riggs, A,, 190, 191 Xis, N., 251 Rist, L., 240 Ritke, M. E., 266 Rivero, J. A., 297 Rivero-Blanco, C., 305 Roberts,A., 152, 153, 155,162, 168,225,226,238 Roberts, J.D., 174, 175, 181, 325 Robinson, D.C., 307 Robinson, S. J., 30 Rotek, Z., 55, 58, 61, 71 Rodel, M.-O., 220,238,261, 296 Rohlich, P., 135 Roig, V. G., 297 Roos, T. B., 143 Root, R. B., 249 Rosati, R. R, 245,246 Rose, B. R., 171 Rose, M. J., 187,286,289,293 Rose, S. M., 255 Rose, W., 303 Rosel von Rosenhof, A. J., 3 Rosen, D. E., 121 Rosenberg, E. A., 213, 260 Rosenberg, K., 107, 112, 113 Rosenthal, B. M., 155 Rossi, A,, 92 Rostand, J., 179 Roth, A. H., 254 Roth, G., 53,55,56, 72,73, 74, 75,91,151,160,181 Rouf, M. A., 102 Rougier, C., 109 Rovainen, C. M., 199 Rovira, J., 130 Rowe, C. L., 21,40,242, 267 Rowe, L., 251, 282 Rowedder, W., 58 Royce, G. J., 165 Rubin, D. I., 155 Rubinson, R, 158, 160, 164

Xr J., 190,198,200
S i i e n h u y s , R., 156, 157, 158,159,160 X ~ ~ ~ - k o o p , P . D . , 8 ,34,53, 33, 70, 77,92,93, 106, 131, 136, 137, 145, 146, 147, 296,327 Siiq-asova, E. N., 104 S i i d i w e , A. M., 157,158, 139,160 \-in, G., 266 S H. L., 175 i m Siiiukawa, A,, 25 S s h h w a , K . C.,79, 150, 153, 134,226 S d m a r a , M., 246 Soble, G. K, 5,27,29, 34, 38, 42,50,62,68,71, 74,75, 85. 89, 179, 180,228, 229, 288,290,296,320 Soden. D. M., 153, 164 S&nski, E., 25,45,46,48, 50,135,144,182,196,245 Sdan4 R, 192,193,194, 197, 199,200,203,205,207, 209,260 S d e r t , A., 296,334 S d e r t , C., 296, 334 S o m q S., 107,110 Sordlander, R. H., 150, 153, 155,158 Sorthcutt, R. G., 121, 158, 160, 161,162,165 sassai, G. J. v, 97 S o a ogrodzka-Zagrska, M., 166 Sunnemacher, R. F., 151, 153 S ~ m a nS., 248 ,
Odendaal, F. J., 269, 270 Ogjeiska, M., 140,142,144, 145 OXara, R. K., 229,230,231, 232,233,234,235,236, 237 Ohler. &I., 40 Ohrani, O., 140 Ohzu. E., 34 Ok Y , 152, 154, 157 . Olrada, Y., 5,296,332 O ~ V., 248 L Ohitomi, K., 57 OLughlin, B. E., 282 Okiharn, R S., 227,228 O k a G. J., 70 O h n , D. H., 20,232 Omerza, F. F., 159 Opatm!; E., 221 Opdam, P., 156, 157, 158, 159 Orlando, K., 197 Orr. P. R, 204 Orron.G.L.,5,6,9,38,41,52, 55,63, 88, 89, 121, 168, 173,179,180,240,244, 280, 288,290,295,296, 298, 307, 321, 333, 339, 344,347,348 Osbome, P. L., 241, 248,291 On.M.E., 193 O m n , D., 168 OI-de, W. K., 104

Ovaska, K., 174 Overton, J., 137 Oxner, W M., 225 Ozeti, N., 296 Pace, W L., 49 . Padhye, A. D., 260,263 Pagel, M. D., 88 Paicheler, J.-C., 3 Pancharatna, M., 143 Pandian, T. J., 208,209,224, 251,283 Papernal, I., 248 Parenti, L. R., 17 Parichy, D. M., 225, 258,261 Parker, G. E., 192,209 Parker, H. W., 297 Parker, W K., 55, 56, 76, 107 . Partridge, B. L., 231 Pasztor, V. M., 53, 80 Paterson, N. F., 24, 55 Patterson, J. W., 199-200,242, 243,251 Patton, D. T., 153 Pavignano, I., 244,248 Peacor, S. D., 275 Peadon, A. M., 137, 178, 181, 182 Pearniati, P. B., 266, 267 Pearse, A. G., 20 Pehek, E. L., 211 Peixoto, O. L., 5,28, 308, 310, 311,314,316,317,321 Pelaz, M. P., 109, 255 Prez-Rivera, R A., 172, 187 Perret, J.-L., 5, 173, 304, 312, 313,331,333 Persson, L., 261 Petersen, H., 45 Peterson, J. A., 261 Petranka, J. W., 183, 199,203, 205,206,219,224, 229, 238,242,243,246, 251, 253,255,260,262,266, 2 72 Petrini, S., 145 Pettigrew, A. G., 160 Pfeiffer, W., 104,235,263 Pfenning, D. W., 27,91,230, 237,246,247,250 Pflugfelder, O., 104 Phiiiips, C., 200,260, 263 Piavaux, A., 131 Pick, J., 151 Picker, M. D., 213,249 Pickersgili, M., 312 Pierce, B. A., 210, 211, 213, 260 Pign, A., 34, 105, 197 Piiper, J., 192 Pdiai, R. S., 24, 297, 303, 330 Pinder, A. W., 190, 192, 194, 195,196,197,199 Pintar, T., 320 Pisan, A., 21, 140, 144 Pitcher, T. J., 231,232 Pivorun, E. B., 21 Plasota, K., 55, 56 Plassman, W., 159 PS.tycz, B., 136,247, 286,293 Poinar, G. O., Jr., 248

Polis, G. A., 217,242,246,247, 261,269 Pollack, E. D., 155 Poole, T. J., 135 Pope, C. H., 6, 7,297 Porges, R., 2 13 Porter, R A., 154 Poska-Teiss, L., 104 Postek, M. T., 16 Potel, M. J., 230 Potter, H. D., 159 Potth08, T I,., 41 Pough, F. H., 243,251 Pouyet, J. C., 146 Poweii, G. V. N., 232 Poweii, T. L., 136 Power, J. H., 297, 303,327, 331 Prakash, R, 93 Prasadmutthy, Y. S., 143 Precht, W., 163 Prestige, M. C., 29, 155,227 Preston, F. W., 26 Pretty, R., 92 Price, G. C., 132 Prigioni, C. M., 303 Pronych, S., 224 Prosser, C. L., 104,209 Pgener, L. A., 55 Pugn, E., 5,179,316 PuiLiam, H. R., 232 Punzo, F., 168,230,239 Pusey, H. K., 53,55,56,58,60, 62,63,70, 71, 72, 73, 74, 75,76 Pyastolova, O. A., 237,259 Pyburn, W. F.,9,31,44,45,57, 121, 123, 174, 196,219, 297,298,308,322,334 Pyles, R. A., 62, 63 Qualls, C. P., 171 Quinn, D., 194, 195,201 Rabb, G. B., 187 Rabb, M. S., 187 Race, J., Jr., 132 Rada de Martinez, D., 297 Radakov, D. V., 232 R&, R. A,, 170, 182, 188,289 Rafols, C., 18 Rahwan, R. G., 22 Raies, D. E., 18 Ramaswami, L. S., 55, 56, 57, 63,64 Rao, C. R. N., 19,44,329,330 Rappaport, R., Jr., 133 Rastogi, R. K, 142 Rathke, H., 107 Ratiba, R, 143 Rau, R. E., 327 Rauthner, M., 121 Rautio, S. A., 233 Razarihelisoa, M., 297, 312, 332 Read, A. W., 21 Reading, C. J., 200 Redshaw, M. R, 140,142 Reed, S. C., 142 Regos, J., 3 15 Reh, T. A., 149, 160 Reiliy, S. M., 288

426 Ruby, J. R, 140 Rugh, R., 93,95,106,255 Ruibal, R., 31, 35,44, 55. 113, 213,296,314,340 Rui. Garcia, F. N., 304 Rusconi, M., 107 Russeli, A. P., 296 Russeli, I. J., 158, 163, 164 Ruthven, A. G., 173 Ryan, M. J., 162 Ryke, P. A. J., 77, 78, 79 Sabater-Pi, J., 329 Sabbadin, A,, 138 Saber, P. A,, 249 Sabnis, J. H., 216 Sahu, A. K., 223 Saidapur, S. K., 142, 143 Saint-Ange,M., 4 Salas, A. W., 321 Salibin, A,, 212 Salthe, S. N., 171,172,185 Salvador, A., 296 Sampson, L. V, 27, 175, 180 Saraeva, N. Y.,135 Sarikas, S. N., 20 Sarnat, H. B., 161 Saruwatari, T., 17 Sasaki, F., 92 Sasaki, H., 153 Satel, S. L., 88,90 Sater, A. K., 93 Sato, K., 104 Satterfield, C.K., 237 Saunders, R., 22 Savage, J. M., 41,295,296, 305, 306,308 Savage,RM.,6,49,51,80, 81, 107, 112, 123, 181, 196, 240,243,244,254,268 Saxn, L., 138 Sazima, I., 315, 318 Scadding, S. R., 17 Scalia, F., 160, 165 Schechtman, A. M., 16 Scheel, J. J., 173, 231 Scheid, P., 192 Schenkel-Brunner,H., 229 Schiesari, L. C., 217,219 Schjfsrna, K.,296 Schiatz, A,, 231,296, 312,313, 328 Schlaepfer, M. A., 18 Schley, L., 224 Schlosser, G., 53, 55, 56, 72, 73, 74, 75,91, 151, 181 Schluga, A,, 135,136,138 Schluter, A., 315, 321 Schmalhausen, I. I., 29 Schmid, M., 144 Sdimidt, B. R., 272 Schmidt-Nielsen, B., 211 Schmidt-Nielsen, K., 197,206, 209 Schmiedehausen, S., 259 Schmuck, R., 211,252 Schmutzer, W., 248 Schnack, J. A., 262 Schneider, L., 104 Schoonbee, H. J., 221 Schorr, M. S., 248

A U T H O R INDEX Schubert, G., 104 Schulze, F. E., 4, 38,55,63,68, 74, 75, 80, 81, 84, 107 Schutte, F., 175, 326 Schutte, M., 167 Schwartz, A,, 173, 174,308 Schwartz, J. J., 162 Scott, N.J., Jr., 28,29, 295, 297, 331 Scott-Birabn,M. T., 5 Seale, D. B., 88, 107, 189,216, 217,221,223, 240,243, 244,248,263,292 Sedra, S. N., 55,56,60,63,66, 68,69, 70, 72, 73, 74, 75, 77,84,92, 106, 163 Seibert, E. A,, 200,201,217, 223,224,260 Seigel, R. A,, 252 Sekar, A. G., 216,247 Selenka, E., 181 Semiitsch, R. D., 21,22, 185, 205,208,226,238,243, 247,250,256,259,260, 261,262,263, 264,266, 267,272 Senn, D. G., 158 Serra, J. A., 296 Sesarna, P., 130 Sever, D.M., 20 Severtsov, A. S., 55,56, 84,216 Sexton, O. J., 200, 260,263 Seyrnour, R S., 21, 181, 192, 194,229 Shder, H. B., 20,279,295 ShafFer, L. R., 171 Shapovalov, A. I., 155 Shaw, E., 231 Shaw, J. P., 42 Shelton, P. M. J., 163, 164 Sheratt, T. N., 261 Sherman, E., 199,203,204, 205,208,224 Shield, J. W., 198 Shirnozawa, A., 127, 130 Shine, R, 171 Shrane, T., 140, 144 Shiriaev, B. I., 150 Shoemaker, V H., 132, 181,252 Shofner, W. i?,163 Shurnway, W., 8,92,93 Shupliakov, O. V, 150 Shvarts, S. S., 237, 259 Siegfried, W. R., 232 Siekrnann, J. M., 296 Sih, A., 273 Siiver, M. L., 152 Siiverstone, P. A., 173, 305, 306, 315 Simnett, J. D., 137 Simon, M. P., 174 Sirnons, B. P., 162 Sirnpson, H. B., 158 Sin, G., 247 Singer, J. F., 153 Sinsch, U., 240, 316 Sive, H., 34 Skeliy, D. K., 229,238,243, 253,263,291 Skiles, M. P., 158, 160 Slade, N. A., 88, 123,240 Slater, D. W., 17 Slater, P., 324 Srnirnov, S. V, 21 Srnith, C.L., 155 Smith, D.C., 200,208,232, 237,243,252, 259, 263, 264,280,282 Srnith. D. G., 105,106,196, 197 Srnith, E. N., 308,309 Smith, L. D.,143,144 Srnith, M.A., 28,41,229,296, 298,320,321,329,330 Smith, R J. F., 263 Srnith, S. C.,164,165 24,25 Smithberg, M., Srnith-Gill, S. J., 204,205,208, 209,229,251,272,281, 283,285 Snetkova, E., 224 Sobotka, J. M., 22 Sokol, A., 256 Sokol, O. M., 5, 6,9, 33, 50, 53, 54, 55,56,57,58, 59,60, 61,62,63,64,68,70,71, 72, 73, 74, 75, 76, 84, 88, 90, 107, 109, 113, 123, 237,296,326, 339, 344, 347,348 Somero, G. N., 192 Song, J., 17 Sotairidis, P. K., 245, 247 Sotelo, C.,159 Souza, K. A., 61 Spallanzam, L., 171 Spannhof, I., 180 Spannhof, L., 135,180 Spemann, H., 56 Spencer, K., 10-11,23, 51, 337 Sperry, D. C., 155 Sperry, D. G., 25,79,206,219, 260 Sperry, R. W., 149 Spinar, Z. V, 3 Spitzer, J. L., 154 Spiker, N. C., 154 Splechma, H.,137, 138 Spornitz, U. M., 140 Spray, D. C., 168 Sprumont, l?, 106 Stanley, H. P., 142 22 Starr, R. C., Starrett, P. H., 5,9, 31,33, 38, 42, 53, 56,57,62,67,68, 70, 71,82, 89,90, 121, 240, 244,296, 302, 303, 339,344,347,348 Stauffer, H. P., 226 Stebbins, R. C., 221, 302 Steele, C.W., 239 Stefanelh, A,, 158 Stehouwer, D. J., 160,226, 238 Stein, J. R., 22 Steinberg, M. S., 135 Steinke, J. H., 20 Steinman, R. M., 34 Steinwascher, K., 216, 223,239, 243,244,245,246,248, 255,256 Steneck, R. S., 49 Stensaas, L. J., 155 Stensaas, S. S., 155 Stephenson, E. M. T., 63 Stephenson, N. G., 55, 173,290 Sterba, G., 101, 103 Stevrnson, H. M., 296 Stevenson, R. D., 204,209 Stewart, M.M., 16, 171, 172, 174, 177, 178, 179, 180, 181 Stiffler, D. E, 213 Stilling, H., 147 Stirling, R. V, 167 Stohr, P., 56, 57 Stokely, l? S., 58, 78 Stone, L. S., 18, 164 Storm, R. M., 50,230 Straka, H., 159 Strathmann, R. R , 171,182 Strauss, R. E., 26,296 Straw, R. M., 203,204 Strawinski, S., 104, 106 Straznich C., 159,160 Streb, M., 254 Strickler-Shaw, S., 168,239 Strijbosch, H., 265 Strong, D. R., Jr., 278 tua<^. C.,201,232,297, 298.322 ~tubblefield, I., 194 K. Stuesse, S. L., 151, 157, 158 Sugi, Y., 93 Sullivan, B., 190 Summers, C. H., 181 Summers, K., 217 Summey, M. R., 263 Sunaga, Y., 64 Sutasurya, L. A., 296 Sutherland, R. M., 153 Swammerdam, J., 4, 189 Swan, G., 175 Swanepoel, J. H., 57, 58, 181 Swart, C.C., 55,63 Sweet, S. S., 279 Swofford, D. L., 70 Syuzyamova, L. M., 21,247 Szabo, T., 160 Szarski, H., 290 Szkkely, G., 154, 159, 164 Tachibana, T., 42, 104 Tachihama, H., 45 Tahin, Q. S., 132 Taigen, T. I,., 172 Takahash, H., 140, 144 Takasu, T., 142, 144 Takisawa, A,, 64 Tan, F. L., 29,227,298, 331 Tanaka, S., 246 Tang, Y.-Y., 199 Tanimura, A., 118, 140, 142, 143,144,145 Tanner, G. W., 263 Taturn, J. B., 26 Taugner, R., 135 Tai, J., 153 Tay, D., 160 Taylor, A. C., 8,9, 92,94, 189, 194,198,251,281 Taylor, C.L., 36,40,223, 227, 247 - -. Taylor, D. H., 168,238,239

A U T H O R INDEX Taylor, E. H., 5, 298, 307,332, Uchiyama, H., 167 Uchiyama, M., 163,211,213 Udin, S. B., 158,1 0 6 Ueck, M., 119,123, 125,127, Vigny, C., 327 Vilia, J., 30, 227, 260 Viliaipando, I.,140,1 4 4 Viy, F., 1 2 6 Vilter, V, 1 1 1 4 8,8 Vine, I.,232 Viparina, S., 208 Visser, J., 44, 5 4 Vitaiis, T. Z., 6 1 9 Vizotto, L. D., 3 1 2 vogt, C., 4 Volkov, V: I., 93 Volpe, E.P., 41,50 Voris, H. K., 200,229

427
Webb, R. G., 40,41,49,295,

33 3
Taylor, J. D., 105, 344 Taylor, J. S. H., 1 3 5 Taylor, R. E., 212 Jr., Taylor, R. J., 232 Taylor, W. R.,2 1 Tejedo, M., 208,242,243, 246,

297,309,339
Webster, H. D., 1 8 6 Webster, T. P., 2 5 0 Weigmann, D. L., 4 28 Weir, A., 8 13 Weisz, P. B., 5 5 Welis, K. D., 162,187,219,220 Wenig, J., 1 4 0 Werduiius, B., 1 1 9 Werner, E. E., 198, 223,238,

129,130,131
Ueda, H., 217 Uhlenhuth, E., 208 Ultsch, G. R., 190, 192,193,

250,251,258,284
Ten Donkelaar, H. J., 154, 5 15 Teran, H . R., 1 1 8 Terent'ev, P. V ,296 : Terhivuo, J., 267 Test, F. H., 229,246,260 Tevis, L., Jr., 207, 229,266 Theil, E. C., 1 0 9 Theirry, T., 1 - 1,23 01 Thibaudeau, D. G., 38,45,46,

240,243,25, 1 263,269, 275,282,292 Wersall, J., 1 3 6 Werschkul, D. F., 3 26 Wachtel, S. S., 1 4 4 Wessels, N. 1 5 K.,4 Wessenberg, H . S., 248 Wager, V: A., 173,175, 5, 180, West, N. H., 192,197,198, 181,220,228, 296,297, 47,48, 184, 50, 340,3 3 4 Vagnetti, D., 4 5 199,201 303, 304,306, 313,323, Thibier-Fouchet, C., 142 Vaira, M., 262 Westerfield, M., 1 8 328,329,331,334 Thele, J., 24, 34 Valencia, J., 296, 316, 317,318, Wagner, E., 140,142,144,1 5 4 Wever, E. G., 61,162 334 Thiesse, M. L., 6 10 Wakahara, M., 1 3 4 Weygoldt, P., 173, 184,219,307 Thornas, D. A.G., 219 Vaientine, J. W, 294 Wake, D. B., 3,77,149, 155, White, B. A., 2 3 8 Thornas, E., 31, 35,44, 113, Valerio, C. 30,227 55, E., Whitear, M., 1 4 0 171,185 Vaiett, B. B., 1 3 1 0 7,8 Whitford, W G., 1 2 9 296,314,340 Wake, M. H., 35, 59,85, 132, Thornas, E. I., 9 3 Van Bambeke, C., 3,4,49 Wlutington, P. M., 1 3 4 142,145, 7 ,174,181, 11 Thompson, R. I. 1 4 ,, 6 Van Bergeijk, W. A,, 61, 6 13 Whitlock, D. G., 154,159 182,185,304 Thorns, C., 296 Van Beurden, E. K., 1 2 9 Wiens, J. A., 229,230 Wakernan, J. M., 199,201 Thrail, J. H., 248 Van Buskirk, J , 30,263, . 264, Walcott, B., 64,83 Wiens, J. J., 55,64, 75,78 Tihen, J. A,, 1 3 2 274,275 Waidick, R C., 224 Wilbur, H. 6,200,206,208, M., Toai, K. 111, 297 R., Vance, W. H., 1 1 5 Waidnian, B., 230, 232,233, 215,229,230,232,240, Toews, D. P., 201 234,235,236,237 2 1 242,243,247,249, 4. Van de Kamer, J. C., 1 8 6 Toloza, E. M., 1 6 209 8, Vandenbos, R. E., 248 Wallace, R. A,, 1 4 4 253,254,256,257,261, Toner, P. G., 1 2 3 Van Den Broek, A. J. P., 1 2 3 266, 270, 274,275,276, Walls, S. C., 237,254 Tosney, K. W., 1 6 Van Der Horst, C. J., 1 4 6 277,279, 281,285, 280, Waiters, B., 260 Townsend, D. S., 171,172, 16, Van Der Linden, J. A. M., 154, Wang, C.-S., 332 291,296 Wangersh P. J., 2 1 2 Wilqnska, B., 1 5 2 1 4 177,179,180,1 1 7, 8, 19 5 Van Der Westhuizen, C. M., 55, Waringer-Loschenkohl, A,, 230, Wilczynski, W., 149,158,159, 187,188 Townsend, V: R., Jr., 2 1 6 160,161,163 243,269 57,58,60,62,63,64 Tracy, C. R., 204,205 Van Dijk, E. D., 3,5,8,9, 14, Warkentin, K. M., 209,216,264 Wild,E.R.,55,268,315,320 Trainor, F. R., 2 2 Wdey,E. O., 72,74,90,5 11 34, 1, 7 28,29, 35,37,48, Warner, S. C., 260,276 153,189 Travis, J., 223,230,241, 243, 179,220,295, 296,297, Warren, E., 175, 179,180, 181, Willielm, G. B., Wd, U., 158,159,164 246, 250,253,259,262, 298,306, 327, 328,329, 16 8 272,279,280,281,282, Wasserman, W. J., 1 4 4 Wdey, A., 3 4 331,333,342,347,348 206 283,284,285,286,287, VanEeden, J.A., 55, 58,61,62, Wassersug, R. J., 3,5,9, 6 20, Willhite, C., 203, 1, 288,293,296 63,64 25,27, 30, 35,41, Wdliams, G. C., 232 29, 33, Treisrnan, M., 232 Van Gansen, P., 140,1 4 4 Wfiams, J. D., 3 1 296, 2 44,48,49,52, 56,57, 55, Trenerry, M. P., 248,269,271 Van Geettmyden, J., 137 61, 79,80,87,88,90,91, Wfiams, M., 1 3 4 Trewavas, E., 7 5 Van Kampen, P. N., 297 104,107,109,111,112, W ~ a m sM. A., 4 , 13 Trexler, J. C.,5 23 Van Mier, E,1 5 5 Wfiamson, I.,258 113, 119,1 1 123,150, 2, Tmeb, L., 8, 10-11,23, 6, 9, 38, Vannini, E., 137,138,1 3 4 Wdson, D. 238 J., 153,154,166,179,182, 5 ,54,55,57,58, 61, 0 59, Van Oordt, P G. W. J., 1 5 . 4 Wdson, L. D., 6 2 9 186,192,194,195,196, 62,63, 76, 78, 64, 77, 88, Van Pletzen, R., 7 9 Wdson, M. V: H., 1 8 1 7 199, 9, 200, 201,206, Wiltshire, D. 270 J., 90,123,162, 180, 206, Van Seters, W H., 55,64 . 216,217, 219,223,224, 207,215, 220,229, 216, Veltkamp, C. J., 22 W~rnberger, H., P. 2 1 225,226,227, 228,230, Windnim, G. M., 20 Vences, M., 174, 296, 320,323, 230,291,296, 300, 299, 231,232,236, 237,240, 6 332 241,243,244,249,260, Winkibauer, R., 1 5 308,340 Tschugui~ova, J., 61 T. 4 Vera, M. A., 174,1 9 7 279, 280,283,288,290, Wischnitzer, S., 1 2 9 Tsuneki, K., 3 16 Verma, N.,11 2 291, 292,296,297,298, Wise, R. W., 1 0 Tubbs, L. O. E., 41,45,48,50 Vermeij, G. J., 2 1 9 Witschi, E., 8, 97, 61, 101,162, 307,322, 323, 331,333, Tucker, V: A,, 132,210,213 13 6 Veself, M., 55, 8 5 334,340,341,344,348 Tung, T. C., 132.137 VesseUun, N.P ,160 . Watanabe, K., 24,35 Wollmuth, L. P., 203,205,206, Turner, R. J.,97 Via, S., 2 5 8 Watkins, T. B., 2 3 2 207,208,224,260 Turner, S. C., 1 7 3 Woliweber, L., 1 6 Viertel, B., 9, 33,49, 102,105, Watling, L., 49 Turnipseed, G., 268 107,108,109,1 0 111, 1, Watson, G. F., 296, 310, 323, Wong, A. L.-C., 248,255 Turpen, J. B., 95,97, 0 ,1 5 10 3 Wong, S., 2 3 2 112,114,115,1 7 119, 1, 324,325,334 Twitty, V C., 1 : Watt, K. W. K., 1 0 9 Wood, S. C., 224 121,122,1 3 1 5 126, 2,2, Twornbly, S., 1 6 8 128,216,223,244,298, Watt, P. J., 2 5 1 Woodward, B. D., 200,230, Tyler, M. J., 5,12,24, 37,41, 242,259,260,263,271, 346,348 Webb, A. C., 140,1 4 4 49,67, 172,175,202,296, Vietti, M., 1 7 3 Webb, J. F., 164 286 310,324,325 Vigenti, C., 1 2 4 Webb, P. W.,5 2 2 Workman, G., 203,205

194,196, 197,198,199, 200,201,203, 205,207, 209,260 Underhill, D. K, 2 1 Underhill, L. G., 2 2 3 Uray, N. J., 149,1 9 5 Ustach, P., 6 Utsunomiya, T., 332 Utsunomiya, Y., 265,332

428 Wratten, S. D., 18 Wright, A. A., 4,296,307, 309, 321,326,331 Wright, A. H., 4,5,296, 307, 309,321,326,331 Wright, M. L., 29,249,260 Wright, P. A., 132 Wright, P. M., 132 Wright, R K., 97 Wright, S., 292 Wu, T. H., 146 Wylie, C. C., 143, 144

AUTHOR INDEX
Xavier, F., 142, 144, 173, 179, 181,182 Yacob, A. Y., 133, 135, 137 Yamaguchi, K., 142, 143, 144 Yamamoto, M., 34 Yamasaki, H., 229 Yang, D.-T., 296, 329, 330 Yazdani, G. M., 303 Ye, C., 319 Yoshizaki, N., 34 Yoshizato, K., 25,28,92,229 Yoshizawa, H., 211,213 Youngstrom, K. A., 168 Yung, E., 245 Yurewicz, K. L., 275 Zaccanti, F., 140, 142, 145 Zakon, H. H., 164,346 Zamachowski, W., 211 Zamorano, B., 132,212 Zeikus, J. A., 22 Zhushev, A. V, 238 Zimmerman, E., 219 Zimmerman, H., 219 Zuber-Vogeli,M., 297, 30 1, 328,329 Zust, B., 140, 143 Zweifel, R G., 44, 50,260, 303, 308 Zw~ling, 24,31 E.,

SUBJECT INDEX

Abdominal veins, 95 Abducens nerve (VI), 59 Abiotic factors, in evolution, 182 Acclimation, 206,284-85 Acid-base balance, 132,201-2 Acidic mine drainage, 213 Acidosis, 201 Acid rain, 201,213-14 Acinar cells, 132 Acoustic foramina, 60 Acoustic nerve VIII, 60 Actin iiaments, 104 Adaptations, tadpole stage as, 279-80 Adenohypophysis, 146,162 Adenosine triphosphatase (ATPase), 106 Adenosine triphosphate (ATP), 190 Adepidermal membrane, 104 Adhesive glands, 32, 34,47, 183 Aditus laryngis, 106 Adrenal glands, 147 Adrenocorticotropic (ACTH) cells, 145 Aggregations. See also Schooling choice tests, 232-34 geometric suucnire of, 230 and kin recog~tion, 232-37 schooling, 231-32 of sibling groups, 235 and social attraction, 230 subdivision of, 236 and thermoreguiation, 207-8, 224 and vuinerability to predation, 26 1 Aggressive encounters, 258 Air breathing of, 197, 220, 223, 224,261 guiping of, 201 Aiarm cells, 104 Aician Blue 8GX, 17 Aicohol, as fixative, 12 Algae, as food, 22 Aiimentary tracts. See &o Digestive system development in endotrophs, 181 dorsolateral views, 108 ventral views, 109 Aiizarin Red S, 1 7 Aiielopathy, 259 a-Grandes, 146 Amacrine cells, 167 !unmonium ions, 212 Amphibian papiiia, 162 Amygdala, 160 Amygdaloid nucleus, 165 Anerobiosis, 194-95 Angular bones, 64 Anguiosplenial bones, 64 Annuius cricoideus, 75 Anterior cardinal vein, 101 Anterior copula, 71 Anterior cupola, 5 7 Anterior parathyroid gland, 103 Antibody staining, 1 7 Anus. See vents; vent tube

Aortic arches, 93,96 Aomc tmnks, 93 Archenteron, 46 Arcus subocuiaris, 62 Arteries, anatorny of, 93 Articular bones, 64 Artificial pond technique, 25455 Ascendmg process, 62 Astrocytes, 162 Atrial chamber, 111 Auditory nerve mI), 60 Auditory system, 1 6 2 4 3 Autonomic nervous system, 151 Axial skeleton, 77-80 Axons, spinal organization, 153 BABB, 1 7 Balbiani body, 142 Barbels, 38,47, 104. See &o Papiiiae Basal process, 62 Basement lamella, 104 Basibranchial, 70 Basicranial fenstra, 54 Bashyale, 7 1 Basii cranii, 54 Basilar papiiia, 162-63 Basophilic cells, 145 Behaviors, 215-39. See &o spenjic behaviors breeding, 173 dispersion, 235-36 and ecological grouping of tadpoles, 244 emetic, 293 escape, 239 and kin selection, 291 locomotion, 2 2 4 2 9 male, 186-87 parent-offspringcommunication, 219-21 predators and changes in, 243 respiration, 223-24 schooling, 231-32 social, 229-37 startle, 239 and thermoreguiation, 224 and trunk myogenesis, 226-27 and vuinerab~hty predators, to 260,261 Belly "finger," 32 Belly sucker, 27 p-Cells, 146 Bicarbonate ions, 210-14 Bidder's organ, 138-39, 142, 145 Bile duct, 131 Bin boxes, for storage, 14 Bipolar cells, 167 Birefringence, 16-17 Blast ceiis, 103 Blood. See spectjic a1types 1 Blood, formation of, 95-97, 100 Blood islands, 95 Bobbing rates, 223 Body plans, shape changes, 25 Body shapes ecomorphologicalviews, 48 measurement of, 25-26

SUB JECT INDEX

Body shapes (contznued) order of examination, 16 representative configurations, 301 Body sizes and feeding efficiency, 292 at metamorphosis, 280 and predation rates, 262 and predators, 243 and temperature, 202,204, 209 and vulnerabllity to predators, 260,261 Body wall lateral, 178 modifications to, 27 musculature, 77-79 Bohr effect, and tadpole Hb, 190 Bone, staining of, 17 Boss, keratinized, 44 Bouin's fixative, 14, 15 Bowman's capsule, 135-36, 138 Brain anatomy of, 155-62 cell types, 151 transverse sections, 156 views, 150,151,152 Braincase, ossiications, 61 Brain stem and cranial nerves, 157 cross sections of, 152 key nuclei, 157 organization of, 155 Branched tubular glands, 127 Branchial arches, 113 Branchial arteries, 81, 105 Branchial baskets, 70-71, 71-75 Branchial food traps, 105, 107, 119 Branchial levators, 72 Branchial pump, 84 Branchiohyoideus extemus muscle, 72, 73 Breathing, 102 air, 197,200 water, 194-97 Breeding, 282. See alro Reproductive system; Spawning Bronchi, 106 Buccal cavity, 107 demarcation of, 80 floor of, 64-68, 110,111-12 innervation, 166 roof of, 110,111-12 taste buds, 166 Buccal flters, 280 Buccal pockets, 111 Buccal pumping aspects of, 87-88 efficiency of, 216 and feeding morphology, 88 in gill irrigation, 82-84 rate and food concentration, 22 1 and suspension-feedmg,224 Buccal volume, 87 Buccophaygeal epithelium, 212 Buccopharynx, 106 anatomy of, 107-13 anterior, 109-10 dorsolateral aspects, 128 histology and ultrastructure, 113-23 origin and evolution, 119 Buoyanq, 224 Cabbage, 22 Caenogenesis, 88 Calciform cells, 131 Calcium, endolymphatic deposits, 180 Calcium ions, regulation of, 21 1 CaVing phenology, 267 seasonal patterns of, 167 seasonal records of, 266 site choice, 266 and spawning, 267 Canales olfactorii, 59 Canalis semicircularis, 59 Cannibahsm, 217-18,245-46 Cannibal morphotypes, 27,41 Capdaries, 104, 106, 196 Carbon &oxide, 105,195,201-2 Carnivores, 280 Carotenoids, 30, 105 Carotid foramina, 54 Cartiiage, staining of, 17 Cartiiago orbitalis, 58 Cartilago tecti, 58 Cadoging. See Collections Catecholamine cells, 147 Cathepsin E, 125 Caudal artery, 93 Caudal lymphatic hearts, 101 Cavum cran, 58 Cell bodies, spinal, 152 Cell division, in growth, 280 Ceratobranchial muscles. 84 Ceratohyal cartilages, 70-71,75, 82.84 Ceratohyoideus externus, 72 Cerebellum, anatomy of, 159 Characters abbreviations use4 298 phenotypic, 286-88 Chemicals, interference by, 253 Chemoreceptors, 110,235 Chloride ions, 210-14 Choanae, 112 Chondrocranium,49 anatomy of, 53-61 braincase, 59 dorsal view, 60 ethmoid region, 57-58 lateral view, 58 ontogeny in endotrophs, 181 orbital region, 58-59 otic capsules, 59-61 summary of literature, 55-57 Chordamesoderm, 143 Chromfin ceiis, 147 Chromatophores differentiation of, 179 examination of, 18 identification of type, 30 morphology of, 29-3 1 Ciliary cells, 46, 115, 117, 123 Ciliary cushions, 115, 117 Ciliary grooves, 112, 115, 117, 119 Ciliated cells, 34,46 Circadian rhythms, 30 Circulatory system, anatomy of, 93-104 Circumoral ligaments, 63 "Clear" ceiis, 104 Clearing, 12, 17, 18 agents for, 17 Cloaca, 33, 131, 181 Coccygeo-iliacus, 78, 79 Coelom, 136 Cold-sensitive neurons, 224 Collagen fibrils, 104 Collections cataioging illustrations, 20 cataloging of, 15 curatorial practices, 14 of specimens, 20-21 Colon, 131 Coloration. See alsa Pigmentation changes in, 104-5 effect of katives, 14 hypothalamus and, 162 and photography of specimens, 19 and vulnerability to predators, 260,261 Columella, 60, 162 Columnar secretory cells (SC3), 123 Commensals, 248 Commissura quadratico-orbitalis, 58 Commissura quadratocranialis anterior, 62 Communication, parentoffspring, 219-21 Conal valve, 93 Conditioned water, and growth, 255 Cones, 166-67 Constrictor brandiialis muscles, 72-73,81 Constrictor colli muscle, 66-67 Constrictor lacyngis muscle, 75 Conus arteriosus, 93 Coo1 streams, and temperature tolerance, 202 Coprodaeum, 131 Coprophagy, 256 Copulae ligaments, 82 Copula I, 71 Copula 11, 70 Copula posterior, 70 Comeas, pigmentation of, 29 Comua trabecdarum, 57 Corpus lymphaticum, 101 Corticosteroids, 147 Cranial cavity, 58 Cranial floor, 54 Cranial lymphatic hearts, 100 Cranial muscles, 65,66 Cranial nerves, 60, 72, 151, 157 Cricoid ring, 75 Crista hyoidea, 71 Cristate mitochondria, 131 Critical point drying, 16 Critical thermal maximum ( a - ) , 199,202-7 Critical thermal minimum (R,), 202-7 Crowding stresses, 254-55 Cultures. See d o speczfic tecbniques of tadpoles, 21-22,42 Cupola anterior, 57 Cusps, 35 Databases, use of, 20 Data coiiection, sampling methodology, 20 Dehydration. Seealro Desiccation diemical, 16 Density stress. See Population densities Dentary bones, 64 Denticles, 37 Dent's kative, 17 Depth strata, within ponds, 269 Dermis, nerves in, 168 Desert pools, 184 Desiccation and egg jeiiy, 186,252 prevention in d t u r e , 22 Development compartmentalization of, 294 energy allocation during, 283-84 link between growth and, 28 1 patterns of, 296 and social behaviors, 237 trajectories of, 283, 290 Developmend categories, 8-9 Developmentai rate (DR), 283 Developmentai timing, 184-85 Diaphragm, 67 Diaphragmato-brachialismuscles, 74-75 Diel cycles, and thermoregulation, 206-7 Diencephalon, 160, 162 Diet. See alm Feeding and growth, 244-46 and habitat selection, 230 in nature, 216 and relationship of species, 247-48 Dderentiation in growth, 280 and temperature, 202, 204, 208-9 Differentiation rates, 250 Diffision, Fick's law, 196 Digestive system anatomy of, 107-32 efficiency of, 245 endotroph development, 181 enzymes, 131 exit of, 33-34 and PBTs, 207 Digits, 28, 178 Dilator lacyngis muscle, 75 Diplotene oocytes, 142 Direct developer g d d , 170 Direct development, 291 Dispersal behaviors, 235-36 Dissections, 21 Diverticula, anatomy of, 12332 DNA, nuclear, 175 Documentation, 15 Dorsal aorta, 101, 136 Dorsal cavity body, 101

SUBJECT INDEX Dorsal filter plates, 105 Dorsal gi remnant, 101 Dorsalis trunci, 78,79 Dorsal pericardium, 93 Dorsal tonsils, 101 Dorsal vela, 113, 115 Ductus cysticus, 131 Ductus thoracici, 100-101 Ducnis thyreoglossus, 146 Duodenum, boundary, 131 Ears, 159, 180 Ecology. See aho Environments factors affecting, 241 interactions in. 241-43 relationships between species, 24748 Ecomorphological classifications, 9-12 basis of, 1 2 and body shapes, 48 diversity of anuran tadpoles, 13 and oral apparatus, 48-50 Egg jeuy and desiccation, 186 and egg preservation, 15 embryo recognition of kin, 234 in endotrophic development, 186 and hatching gland function, 34 illustration of layers, 20 physical properties of, 20 removal of, 1 8 Eggs coiiection of specimens, 20-21 desiccation of, 252 of endotrophs, 172, 175 examination of, 18 as food source, 218-19 illustration of, 20 netting of, 21 pigmentation, 179 preservation of, 1 5 retention of, 183, 185 staining of, 20-21 terrestrial masses, 291 turgidtv of.. 186 , Egg & and composition, 184,18586 and developmental timing, 184 and arowth rate. 285 and ietamorphic parameters, 257 and sprint speeds, 258 variations in, 284 Egg tooth, 179 Electrolytes, 132 Electroshockers, 20 Elevation, 202,204 Ehptical hyoquadrate joint, 82 Elygia, 2 9 Embryology, of endotrophs, 177-78 Embryonic discs, 177 Embryonic stages, Gosner equivalent, 92 Embryos as category, 9 cdture of, 180 endotrophs, 176-77 pigmentation, 179 Endocardial tube, 9 3 Endocrine pancreas, 146 Endocrine system, anatomy of, 14547 Endolymphatic fluid, 162 Endolymphatic foramen, 6 1 Endolymphatic sacs, 180 Endoplasmic reticdum (ER), 106,115 Endotrophc anurans, 170-88 body development, 178 breeding biology, literature review, 173-75 characterization of taxa, 171 comparison of staging systems, 178 definition, 12 developmental energy, 170 development of, 172-82 digestive system, 181 diversity of tadpoles, 13-14 early embryology, 177-78 egg tooth, 179 example taxa, 13 families and genera, 172 gills, 179-80 guilds, 170 hatching, 181-82 integument, 179 limb development, 178 literature review of anciilary subjects, 171 mouthpart development, 179 nervous system, 180-81 operculum, 180 physiology, 181-82 reproductive system, 1 8 1 respiratory system, 180 sensory systems, 180-81 skeletogenesis, 181 spiracle, 180 mil development, 178-79 urogenital system, 181 Endotrophy evolution of, 182-87 exotrophs approaching, 183-84 and oviposition, 183 Energy. SCCa150 Resources aiiocation during development, 283 and egg size or composition, 185-86 Environmental uncertain~; 275-77 Environments. See also Ecology and aggregations, 230 and coloration, 30 and evolution, 3 and growth, 280 hypercarbic, 190 intluence o n keratinized mouthparts, 2 1 Enzymes, digestive, 131 Eosinophils, 97,101 Ependymal layer, 157 Ephemeral pools, food sources in, 184 Epidermis composition of, 104 nerves in, 168 permeability of, 106 removal of, 1 6 vascdarization, 104 Epiphysis, 161, 168 Episternum, 181 Epithalamus, 161 Epithelial body, 103 Erythrocytes, 101 differentiation of, 9 7 organic phosphates in, 190 Erythropoiesis, 100 Ethmoid plate, 54, 5 7 Ethmoid region, 57-58 Eustachian tubes, 60,180 Evaporation, and temperature tolerante, 202 Evolution aspects of tadpole morphology, 48-50 and environrnent, 3 of Me history stages, 289 Exercise, 195,201 Exotrophic tadpoles definition, 1 2 example taxa, 13-14 Gosner staging system, 10-11 Exotrophy, approach t o endotrophy, 183-84 Exovivipary, 170, 183, 186-87 Exploitative competition, 243, 256,257 External gills, 34 Eyes diameter, 26 endotroph development, 180 movement of, 238 order of examination, 1 6 orientation of, 29 position and visual field, 237-38 position of, 2 9 of semi-terrestrial tadpoles, 228 Facial nerve (VII), 59 Fascia dorsalis, 79 Fat body, 143 composition of, 145 origin of, 144 Feces, 33 nutritional value of, 156,248 Fecundity, and growth rate, 280 Feedmg. See also Food; speczfiE
orps

4 31 and morphological models, 216-17 morphology and buccal pumping, 88 and Orton's tadpole types, 8991,243 rates and particle size, 244 suspension, 223,224 Femoral veins, 9 5 Fenestra frontoparietalis, 59 Fenestra ovalis, 59 Fenesua rotundum, 6 1 Feduation, of endotrophs, 172-77 Fiber optic lights, 1 5 Fick's law of diffision, 196 Field studies competition and predation, 270-72 enclosures, 252-54 marking techniques, 18 photography, 1 9 Filter apparatus, 107 cytology of, 120-21,122 histogenesis, 116 morphogenesis, 115-17 ontogeny of, 115, 117, 1 2 4 25,126 origin and morphogenesis, 128 structure of, 114 Filter plates, 112, 117 Fiitration rates, and particle size, 243-44 Fins, 26,178 Fissura metotica, 6 1 Fitness definition of, 289 and Me-history stages, 281 maximization of, 279 Fixation, 12, 1 4 1 5 , 2 9 5 Flageiium, tail, 28 Flask ceiis, 104 Fluorescent elastomers, 1 8 Food. See aho Diet; Feeding and buccal pumping rate, 221 for cultured tadpoles, 2 1-22 density, body mass and, 257, 258 and growth, 245 and metamorphic parameters, 257 and metamorphic timing, 218-84 particulate, effect of supply, 184 sources of, 216 Foramen jugulare, 6 1 Foramen magnum, 6 1 Foramen orbitonasalis, 59 Foramina carotica primaria, 54 Foramina olfactoria evehentia, 59 Forebrain, anatomy of, 160-62 Foregut, 123,129,130 Formalin, 12, 1 4 Fossiis, tadpole, 3 Fourth ventricle, 157 Frontal organ, 168 Front legs, eruption of, 25 Frontoparietal fenestra, 59 Frontoparietals, 6 1

behaviors, 215-23, 244 branchial food traps, 105 comparative aspects, 84-87 coprophagy, 256 dynamics of, 223 efficiency of, 292 grazing, 223,248 hyobranchial apparatus, 70 ingestion, 119 jaw muscles, 82 jaw sheath morphology, 49 larval specializations, 88

SUB J E C T I N D E X
Frozen lakes, and temperature tolerance, 202 Funnel traps, 20 Gabladder, 131, 181 Gametes, 172-77 Ganghon cells, 167 Gas exchange, 104 and egg jeiiy, 186 partitioning, 196-20 1 physiology of, 191-96 and temperature, 202,204 Gastric vein, 95 Gastdation, 175, 177 Gene expression, sequential, 187 Geniohyoideus muscle, 67 Geniohyoid muscle, 82 Genotypes, and vulnerability to predators, 260 Geotaxis, 228,229 Gi apparatus, summary of literam e , 55-57 Gi arches, in endotrophs, 180 Gi arch skeleton, 64 Gi cavity, anatomy of, 80-81 Gi irrigation A . truei, 85-86 branchial pump, 84 buccal pump, 82-84 buccal pumping, 87-88 evolutionary and comparative aspects, 84-88 and frequency of breathing, 106 Gradwell's model, 85 hydrostatic pressures, 83 and hypoxia, 199 muscle activity, 83 E annectens, 86 pharyngeai pump, 84 R. mtesbeianu, 80-84 viscerai skeleton during, 82 X. l a r , 86-87 Gis anatomy of, 105-6 CO, and growth of, 105 contribution to gas exchange, 198 endotroph development, 179-80 ion uptake through, 212 persistent, 105 remnants, 104 Gi tufts, diagram of, 106 GiU ventilation cate, 224 Gland ceiis, 125 Glanduia interposita, 101 Glial astroqes, 162 Glomus, 135 Glossopharyngeal nerve (E), 61 Glottis, 106,107,109 Glycerin, 17 Goblet cells (SC2), 104, 115, 117, 123, 131. Seealso Mucous cells Goiter, 22 Golgi complex, 104, 106, 115 Gonads. See also Ovaries; Testes development in endotrophs, 181 development of, 138-39 differentiation of, 1 4 0 4 1 and fat bodies, 143 male, 140 cate of development, 144 sex differentiation, 140 Gosner system of staging approximation to other system, 8 for exotrophic tadpoles, 10-11 of staging, 8-9 use of, 298 Grazing, 223,248 Growth early, 280-81 factors decting, 245 link with development, 281 and social behaviors, 237 unrestrained exponential, 280 Growth rates correlation with larval period, 286 and fimess, 279-80 genetics of, 285-86 iduences on, 248-60 models of, 280-81 and ppuiation density, 243 and survival, 280 and temperature, 202,208-9 Guilds, 48,170,186-87 Gustation, 166 Gut, 212,244-45. See alro Digestive system; Intestines Habenuia, 161 Habitats factors decting ecology of, 24 1 general features, 240-41 partitioning of, 268-69 and vulnerability to predators, 260 Habitat selection sensory stimuli and, 239 and substrates, 229-30 Hair ceh, 162,163-64 Hassa's corpuscles, 101 Hatching and egg jelly, 186 in endotrophs, 181-82 factors iduencing, 229 "late," 183 and the oxygen supply, 183 Hatching glands, 34 Hatchlings behavior of, 229 as category, 9 endotrophs, 176-77 guild assignment, 186-87 size and available energy, 186 Head, width of, 294 Head flap, 32 Heart, 93, 94. See also Circulatory system Heat exchange, 105 Heights, measurement of, 26 Helminth parasites, 248 Hematocrits, 2 10 Hematopoietic precursor cells, 100 Hemoglobin (Hb), 100,190-91 Hemopiesis, 95-97, 100, 135 Hemoproteins, 96 Hepatic ducts, 96 Hepatopancreatic duct, 129-30, . 131 Herbivores, competition with, 243 Hibernada, 195 Hindbrain, 155-59 Hind limbs, 178 Hitwiae Animalium (Gesner), 3 Hormonal profiies, and breeding, 187 Hucker's Crystal Violet, 18 Hydrochloric acid, 125 Hyoanguiaris muscles, 70, 81 Hyobasai process, 62 Hyobranchial apparatus, 70-77 Hyobranchial plates, 75 Hyoid arch muscles, 70 Hyoid muscles, 70, 71 Hyoquadrate process, 62 Hypercarbic environments, 190, 20 1 Hypophyseal fenstra, 54 Hypothalamus, 161-62 Hypoxia anerobiosis durine. 194-95 and capillary mesrdensity, 196 and gas exchange, 196-99 and hibernada, 195 and lung venulation, 223-24 rhythms in, 190 temperature and tolerance of, 199 Ileolumbalis pars lateralis, 79 Ileolumbalis pars medialis, 78, 79 Ileum, 131 Iliac arteries, 93 Iliacus externus, 79 Iliac veins, 95 Ilium, 79 Illustrations,of specimens, 18-20 Infections, risk of, 97 Inferior perilymphatic foramen, 61 Infrarostral cartilage, 49 Infundibuium, 162 ingestion rates, 221,222 Inner sensory layer, 104 Inspiration, pressure dynamics, 84 Integument endotrophs, 179 and gas exchange, 196 morphology of, 29-31 and pigmentation, 104-5 Integumentaq glands, 30 Interarcuales, 78 Interauricular septum, 93 Interference competition, 256 and growth, 257 and kin recognition, 237 between sibhgs, 259-60 Interhyoideus musde, 82,83 Interhyoideus posterior muscle, 66,67,68, 84 Interhyoid ligaments, 70,82 Intermandibuiaris muscles, 40, 66 Internal carotid arteries, 54 interrenal vein, 93 Interspecific competition dects of pond drying, 271-72 and growth inhibition, 273 in the pond environment, 279 interspecilic competition and predation, 270-77 dects of environmental uncertainty, 275-77 field studies, 270-72 interactions between, 273-75 laboratory studies, 272-73 Interspecific variations, 295 Intertransversarii, 78 Intestinal tract, 106 intestina, 181. See alro Gut intrageneric relationships, 5 intrahyoid ligaments, 71,82 intraspecific competition, 259, 279 Involution, 177 Ion balance, 210-14 Iridiophores, 20,30, 105 Irises, 29, 180 Ischium, 181 Islets of Langerhans, 146 Istock's paradox, 292,293 Jaw c d a g e s , 35-36 Jaw depressor muscles, 70 Jaw levator muscles, 68-70 Jaws anatomy of, 6 1-70 development of, 180 lower, 64 opening and closing, 81-82 terminology for musdes, 68 view of musdes and ligaments, 69 Jaw sheaths cannibal morphotypes, 41 chronology of research, 4 development, 4 2 4 5 ecomorphologicai views, 49 illustration of, 19 and oral disc morphology, 37-38 in oral ontogeny, 46,47 removai of, 17 shapes of, 44,45 terminology, 36 Jugular foramen, 61 Karnovsky's fixative, 16 Keratinization, 44 Keratinized ceh, 42 Keratinized mouthparts, 21 Keratinized strucnires, dearing of, 18 Keys to families of anuran tadples, 336-37 inadequacy of, 2 Kidneys anatomy and development, 133 hematopoiesis, 97

S U B JECT INDE mesonephros, 181 opisthonephnc, anatomy of, 6, 93,135-36 stones, 22 Kin association, 232-37 Kin recognition, 232-37 Kin selection, 291 K i r ~ e l z e l h ,104 Labeling, 15 Labial flaps, 38 Labial teeth chronology of research, 4 ecornorphologicalviews, 49 examination of, 17-18 keratinized, derivation, 42 rnargins, 37 and mode of feeding, 280 ontoaenv, 46 and & I i & s crnorphology, 37-38 replacernent, 42 size of, 42 termuiology, 35 and tooth ridges, 38-42 variations in, 43 Labial tooth row formuia (LTRF),40,41,46,50 Labial tooth rows (LTRs), 41 Lactate, 195,201 Lagena, 158,163 Lamina orbitonasalis, 57, 59 Lamina temiinaiis, 167 Larval periods correlation with growth rate, 286 genetics of, 286 length of, 285-86 and popuiation density, 256-57 Larval samples, storage of, 14-15 Larval stages, Gosner equivalent, 92 Larval stornach, 123 Larval traits ecological genetics of, 284-88 intrageneric relationships, 5 Laryngeal cartilages, 106 Larynx, anatomy of, 75 Lateral cephahc vein, 59 Lateral circumoral ligament, 63 Lateral hypogiossal nucleus, 158 Lateral lemniscus, 158 Lateral line nerve, 164 Lateral line organs, 18 Lateral line system, 163-65 and school structure, 238 Lateral posterior cardinal veins, 93 Lateral sacs, 27 Lateral tecti inferius, 58 Lateral tem superius, 58 Latissirnus dorsi, 79 Learning and habitat preference, 239 vision, 237-38 Lectins, 127 Length, termuiology for, 36 Lens, 166 Leptotene oocytes, 140,142 Leptotene spermatocytes, 142 Lettuce, 21 Levator arcus branchiaiis, 72 Levator bulbi muscles, 166 Levatores arcuurn branchialium, 72 Levator rnandibuiae anterior, 68, 82 Levator mandibuiae anterior articula~%, 68-69 Levator mandibuiae anterior pars intermedialis, 70 Levator rnandibuiae anterior pars lateraiis, 70 Levator mandibuiae anterior subextemus, 69 Levator mandibuiae externus, 69, 70,82 Levator rnandibuiae posterior pars profunda, 70 Levator mandibuiae posterior pars superficialis, 70 Levator mandibuiae posterior profunda, 68,82 Levator mandibuiae posterior superficiaiis, 68,82 Levator mandibuiae subexternus, 69,70 Leydig ceUs, 104 Life cycles, complex evolutionary stability, 294 evolution of, 289,293-94 fossil evidence of, 290 general features, 240-41 maintenance of, 291-93 origin of, 288-93 Ligamentum circumorale, 63 Ligamentum cornu-quadratum, 63 Ligamentum cornu-quadratum lateraie, 63 Ligamentum cornu-quadraturn rnediale, 63 Ligamentum interhyale, 70 Ligamentum interhyoideum, 70 Ligamentum intrahyoideum, 71 Ligamennun quadratoethmoidale, 63 Ligamentum rostrale superius cartilago Meckeli, 63 Ligamentum rostrale superius quadrati, 63 Ligarnennun tecti, 57-58 Light egg sensitivity to, 22 phototaxis, 207 response to, 167, 168 sensitivity to, 237,238 Light Green stain, 17 Limbs buds, 178,184 developrnent in endotrophs, 178 front, 28-29 hind, 28-29, 30-31, 178, 294 morphology of, 28-29 Lipases, synthesis of, 132 Lipid granules, 106 Lipid vacuoles, 131 Liver anatomy, 131 development in endotrophs, 181 hematopoiesis, 96,97 Locomotion anatomy of, 79-80 behaviors, 224-29 neurornuscular control, 226 semi-terrestrial,228, 229 tail flipping, 228,229 and temperanire, 209 terrestrial, 227-28 using oral discs, 227 wiggling, 228 Longissirnus dorsi, 78, 79 LTRF (labial tooth row formula),, 40,41,46,50 LTR (labial tooth row), 41 Lung buds, 106 Lungs anatomy of, 106-7 contribhion to gas exchange, 198 dfierentiation of, 104 and gas exchange, 196 and hypoxia, 195-96, 22324 Lymphatic organs, 97, 100-101 Lymphatic sacs, 100 Lymphatic vessels, 97, 101 Lymph glands, 97,101-3 Lyrnphocytes, 101,102,103 Lyrnphomyeloid organs, 101, 103 Lyophylization, 16 Macrophages, 97, 101 Macrophagy, 229 Macula densa, 138 Malpighian corpuscles, 135 Mandibuiae tentaculi, 70 Mandibulolabialis rnuscle, 38, 40,66, 67 Mandibuio-suprarostral ligarnents, 63,81,82 Manicotto glandulare, 123, 125, 127, 129, 130. See &o Stomach Marking techniques, 18 Mastication, 159 Mate choice, 259 Matemal behavior, 219, 220 Maternal dects, 284-85 Mauthner cells, 158 Mechanoreceptors, 163 Meckel's cartilages, 49 lower jaw, 64 rnetamorphosis of, 75, 76 rnovernent of, 8 1 rnuscles attached to, 82 Medial longitudinal fasciculus, 158 Medial posterior cardinal veins, 93 Meduiiary nucleus mediaiis, 164 Meiosis, 142 Melanin, 30, 105 Melanophores, 30, 104, 105 Melanophore-stimuiating horrnone (MSH), 105 Membrana vasculosa opercularis, 81 Mentomeckelian bones, 64,76 Merkel ceh, 104 Merkel grandes, 104 Mesencephalic nucleus of V, 160 Mesencephalon, 159-60 Mesentrial arteries, 93 Mesoderm, and the circulatos. system, 93 Mesogoniuni, 140 Mesonephros, 181 Metabolic rates, and crowding stress, 254-55 Metabolic wastes, 255 Metachrosis, 30 Metamorph, as category, 9 Metamorphc period, 9 Metamorphosis body size at, 280 c h a x stage, 92-93 and deteriorating environments, 250 hyobranchium, 75-77 and locornotion, 79 lymph gland disappearance, 101 rneiosis in, 142 rnodels of timing, 281-84 and rnortaiity rates, 25 1 nervous systern changes, 154, 168-69 palatoquadrate, 63 pancreas reconstruction, 132 size at, 243,286 and ternperature, 208-9 temperanire and rnass at, 258 temperature tolerance at, 205 tongue anlage, 100 Metoptic foramen, 59 Microhabitats, species within, 265 Microhylid larvae, 89-90 Micropinocytotic vesicles, 135 Microscopes, 15 Microsporidian parasites, 248 Microvilli, 115 Microvillous adsorptive ceUs, 131 Midbrain, anatorny of, 159 Middle ear, 162 Mitochondna, cristate, 131 Mitochondria-nch cells, 104, 106 Monocytes, 102 Morphology, phylogenetic aspects of, 88-91 Mortaiity rates, 251,282 Mounting, of soft parts, 17 Mouthparts chronology of research, 4 in cuiture, 295 development in endotrophs, 179 iustration of, 297 iduence of environment on development, 21 keratinized, 38 terminology for, 35-36 Mowiol, 17, 18 MS-222, 16, 18

434 Mucopolysaccarhides, 186 Mucous, 119,283 Mucous ceiis, 104. See also Goblet ceiis Mucous cords, 123 Mderian ducts, 142,145 Murrafs Clear, 17 Muscular process, 62 Musculature axial skeleton, 77-80 body w d , 78 branchal baskets, 71 buccai floor, 64,66-68 craniai, 65, 66, 77 hyoid, 71 jaw, 68-70 of the larynx, 75 summary of literature, 55-57 tnink, 78,226-27 Musculo-abdominal veins, 95 Musculus constrictor branchialis, 105 Muscdus dilator laryngis, 106 Myocardium, growth of, 93 Myxozoan parasites, 248 Nares apemre, 32 diameter, 26 external, morphology of, 31 internai, 112 order of examination, 16 water flow through, 84 Nasal capsde, 57 Nasal septum, 57 Nasolacrimal ducts, 29 Nematode parasites, 248 Neotropical ponds, diorama, 217 Neotropical streams, diorama, 218 Nephric ducts, 131 Nephrons. See d o Kidneys innervation of, 136 morphogenesis of, 138 Nephrostomes, fusion of, 136 Nervous systems endotroph development, 180-81 overview, 150-52 Nets, use of, 20 Neurai arches, ossification, 78 Neural crest cells, 143 Neuromasts, 163-64, 181 Neutrophils, 97, 101, 102, 103 Nidicolous continuum, 183, 184 Nidicolous guild, 170 Nitrogenous excretion, 132, 136,252 Nomenclature, conventional names, 2 Notocord, 54, 77 Nucleases, synthesis of, 132 Nucleus isthmi, 158 Nucleus lentiformis, 160 Nucleus profundus, 160 Nutrition. See also Feeding; Food of cdtured tadpoles, 21-22 and gland cell development, 125 and PBTs, 207

SUBJECT INDEX
Obiiquus abdominis externus superficialis, 78 Observation, rnethods of, 15-18 Occipital arch, 54 Ocular reticles, 15 Oculornotor foramen, 59 Oculornotor nerve (111), 59 Offspring, communication with parents, 2 19-21 Of Scientists a n d Salamandevs ( w., Ti ) 1 Olfactos. bdbs (CNI), 160, 165 Olfactos. canals, 59 Olfactos. epithelium, 164 Olfactos. organs, 180 Olfactos. systern, 165-66 Omphalomesenteric veins, 95 Oocytes, 140,142,175 Oogonia, 140, 144 Oophagy, 218-19 Opercular chamber, 111 Operculum, 162 definition, 3 1 endotroph development, 180 fusion with the body w d , 31, 33 Operculunl (otic), 60 Ophthalmic artery, 59 Ophthaimic branch of V; 59 Opisthonephros, 93,96, 1 0 3 4 , 134-35 Optic foramen, 59 Optic nucleus, 160 Optic placodes, 180 Oral apparatus atrophy of, 25,48 ecomorphologicalviews, 48-50 functionai morphology, 49 iliustrations of, 19 morphology, 48-50 ontogeny of, 45-48 oral disc, 35 representative contigurations, 300 terminology for, 35-36 Oral discs extension during ixation, 17 extrinsic muscdature, 38-39 and feeding ecology, 3 iilustration, 19, 35 in length measurement, 25 morphology, 36-38 order of examination, 15-16 and tadpole movement, 227 terminology, 35 umbeiiiform, 36-37 Oral flap. See Labial flaps Oral pad, 46 Orbital cartilage, 58 Orbital region, 58-59 Orbitohyoideusmusdes, 71, 82, 84 Orbitonasal foramen, 59 Orbitosphenoid, 59 Orbits, 26 Orton's tadpole types, 52-53, 54, 89-91, 112 Osmoregulation, l 3 2 , 2 10-14 Osmotic pressures, 2 13 Ossification craniai, 88 sequence in endotrophs, 181 Ostia pharyngea, 60 Ostium, 142 Ostium abdominale, 144 Otic capsules, 59-61, 75 Otic iigament, 59 Otic process, 62 Otocyst of Chibon, 180 Otoliths of Hughes, 180 Ovaries, 140, 144. See also Gonads Overwintering and rnetamorphosis, 208 and Q,", 192 and temperature tolerance limits, 202 Oviductal secretions, 182 Oviducts, 142 Oviposition, 293 endotrophy and, 183 and maie behaviors, 187 Ovisacs, 142 Ovoviviparous guild, 170 Oxygen carrying capacity in tadpoles, 190 changes in whole-body conductance, 192-93 dissociation curves, 190 dissolved, and growth rates, 249 dissolved, and temperatiire, 200 extraction by filter plates, 105 and hatching, 183 limitations of uptake snidies, 191 and lung function, 106 tempera&re and consumption, 209 uptake of, 191-96 Oxygen tension (PJ, 192, 194, 229 Pachytene oocytes, 142 Pachytene spermatocytes, 142 Paedornorphosis, 294 Pain, sense of, 168 Paiatabihty, 260-61 Paiatines, 59 Paiatoquadrate cartilage, 57, 62-63 Pancreas, 132, 181 Pancreatic enzymes, 131 Papdiae buccai floor arena, 110 buccal roof arena, 110 lingual, 110, 111, 112 marginal, 46, 47 and mode of feedmg, 280 morphology, 36-38 postnarial, 112 on the snout, 29 Parachordals, 54 Parahyoid bone, 76 Parasites and inhibition of growth, 255 in the pond environment, 279 248 of tad~oies. Parasyrnpathetic nervous system, 151 Paraviviparous guild, 170 Paravivipaq 186-87 Parentai care, post-hatching, 219 Parents, cornmunication with offspring, 219-21 Parietal nerve, 161 Pars abdominalis, 79 Pars anterior, 70 Pars convoluta, 142 Pars externa, 60 Pars interrnedia, 105, 145 Pars locomotorius, 68 Pars nervosa, 145 Pars posterior, 70 Pars reuniens iigaments, 82 Paternal behavior, 219,220-21 PCO,, 201 Pear-shaped secretos. ceiis (SC4), 123 Pectoraiis, 79 Pelvic girdle, 79 Pen techniques, 252-53 Pepsin, 125 Peptidases, 132 Peribranchial chamber, 111 Peribranchial w d . See Peribranchial chamber Periderm, 104 Periiyrnphatic ducts, 60-61 Periiymphatic fluid, 162 Perilymphatic foramina, 60 Periiymphatic sac, 60-61 Perisinusoidai space, 131 Peritoneal funnels, 138 Permanent waters breeders in, 200 effect on reproduction, 242 predation pressure, 291 Peroxisomes, 131 PH acid tolerance, 260 effects of, 212-14 and formalin storage, 14 and growth rates, 249 and hatching, 229 Pharyngeai cavity, 80 Pharyngeai pump, 84 Pharyngeai slits, 179 Pharyngobranchialtract, 112 Pharynx, 107 arterial system, 93 endotroph development, 180 formation of, 117 ontogenesis of, 110, 111 in pipoid tadpoles, 108 terminology of, 121 Phenotypes changes and predation, 264 character correlations, 286-88 plasticity of, 295 Pheromones, 229 Phosphates, organic, 190 Photography, 18,20 Photoperiods, and PBTs, 207 Photoreception, 166-67,238 Phototaxis, 207
L ,

S U B JECT INDE Phylogenetic analyses, 50 Phylogeny, and the first tadpole, 52-53 Physiology, and the environment, 189-214 Pigmentation. See also Coloration changes in formalin, 12 chronology of research, 4 endotroph eggs and embryos, 179 eyes, 29 integument and, 104-5 iridiophore, photography of, 20 oral disc, 36 in tadpoles, 29-3 1 Pila antotica, 59 Pila ethmoidalis, 59 Pila prootica, 59 Pin cells, 104 Pineal gland, 160,161,162, 168,238 Pipoid larvae, 89-90 Pipoids, cranial ossifications, 88 Pituicytes, 162 Pituitary gland, 105, 148, 162 Piniitary vein, 59 Planum antorbitale, 57 Planum trabeculae anterior, 54 Plasma composition, 210-1 1 p H of, 201 salinity and osmotic pressure of, 213 Plectrum, 60, 162 Plesiomorphic larvae, 89-90 Pluripotent hemapotoietic stem celis (PHSCs). 95 PO,, 105: 192,195, 197, 199, 200 Po~ulation densities body mas, density and, 257, 258 and growth, 245 larval period length, 256-57 and metamorphic parameters, 257 stress and, 250 and survival, 252-53 frorn a temporary pond, 242 Population stability, 266-67 Porphyropsin, 167 Portal vein, 95 Positioning, for examination, 16 Posterior copula, 70 Posterior parathyroid gland, 103 Posterior vena cava, 95 Posture, and thermoreguiation, 204 Potassium hydroxide, 14, 1 7 Potassium ions, 213 Predation and buoyancy, 224 and the competitive hierarchy, 273 &ect on survivors, 274 interspecific,246 and palatabilis; 260-61 and phenological variation, 277 in the pond environment, 279-80 pressure, 291 risks, 243 Predators aversive reactions to, 238 evasive burst swirnming, 226 gape-lirnited, 211,243,280 interactions with, 260-64 olfactory clues from, 261-63 and prey body si=, 262 and schooling behaviors, 232 and tadpole coloration, 30 tadpole interamons with, 243 Preferred body ternperanires (PBTs), 203,205 Premetamorphic period, 9 Prernetamorphic stage, 92 Preoptic root, 59 Preservation, 12 Primordial germ cells (PGCs), 140 Primordial germ cells, presumptive (pPGCs), 143 Principal component analyses (PCA), 287-88 Processus antorbitaiis, 57 Processus ascendens, 62 Processus basalis, 62 Processus hyoquadrati, 62 Processus muscuiaris, 62 Processus rnuscularis capsulae auditivae, 59 Processus oticus, 62 Processus pseudobasalis, 62 Proctodeum, 33 Profundus nerve, 59 Progonium, 140 Prolactin, 283 Prometamorphic period, 9 Pronephric ducts, 135, 137-38 Pronephric slus, 135 Pronephros. Seealso Kidneys anatomy and develo'prnent, 134-35 anatomy of, 103-4 develoiment in endotrophs, 181 hematopoiesis, 96, 97 Prootic foramina, 59 Prostaglandin E (PGE,), 249 Proteases, 132 Protowans, enteric, 248 Pteridines, 30, 105 Pterinosomes, 105 Pteryogoquadrate, 62-63 Pulmonary arteries, 93 Pulmonary hyperventilation, 201 Pupils, 180 Purkinje ceiis, 154, 159 Quadratoanguiaris, 70 Quadratoanguiaris muscles, 8 1 Quadratocranialcommissure, 62 Quadratoethmoid iigament, 63 Ramus communicans, 153-54 Recessus scalae tympani, 60-61 Rectum, 131 Recnis abdominis rnusclc, 25, 32, 78, 79 Rectus abdominis profundus, 79 R e m abdominis su~erficialis. 78,79 Reinis cervicis, 79 Referente coliections, 2 Rehydration, of specimens, 14 Renal arteries, 136 Reproduction, timing of, 265-66 Reproductive phenology and ecological interactions arnong tadpoles, 276-77 d e c t of short-terrn variations on, 275-76 Reproductive system anatomy, 140-43 development of, 143-45 endotroph development of, 181 Resources and exploitation competition, 243 and population density, 268-70 use of, 281 Respiration, at the surface, 104 Respiratory system anatomy of, 104-7 behaviors, 223-24 endotroph development, 180 physiology of, 189-202 responses to exercise, 195 Respiratory toxins, 20 Reticle units, 15 Reticuiar celis, 101 Reticular formation, 157 Reticuioendothelial cells, 97 ReticuloendotheLial organs, 97, 100-101 Reticuiospinal tracts, 155 Retinas, 166-67,180 Rhodopsin, 167 Rhombencephalon, 155-59 Riesenzellen, 104 R k ribosornal, 172, 175 N Rods, 166-67 Rohon-Beard cells, 154 Roteone, 20 Round window, 61,162 Rubrospinal tracts, 155 Saccuies, 158, 163, 180 Saccus perilymphaticus, 60-61 Sacral Liiapophyses, 79 Salamanders and gas exchange, 196 in hypoxic environrnents, 198 oxygen uptake snidies, 191, 193 Salinity, 212-13 Sarnpling methodology, 20,268-69 problerns in naniral ponds, 254 Saponification, 12 scnning electron microscopy, 16,18
Schleimkphchenzellen, 127 Schooling. See also Aggregations and egg rnasses, 291 and lateral line systems, 238 Schools, spatial arrangernents in, 237 Seasonal cycles pattems in habitats, 241 and the pineal gland, 168 and therrnoreguiation, 206-7, 269 Secretory cells (SCs) bottle-shaped (SCl), 113-21 columnar (SC3), 115-21 goblet (SC2), 115, 119 pear-shaped (SC4), 115 Secretory grooves, 117 Selection for adult specialization, 288 of habitats, 229-30 heuristic models of, 287, 289 and mortahty factors, 279 Semicircular canals, 59, 158, 163,180 Seminal vesicles, 142 Sensory systems endotroph development, 180-81 integration of, 162-68 Septorndae, 58 Septum, 160 Septum nasi, 57 Serous glands. 29 Sertoli cells, 144 Sex, differentiation, 144 Shdow ponds, and temperanire tolerante, 202 Silicon cement, 16 Sinus venosus. 93 ~keleto~enesis', endotrophs, in 181 Skin carbon dioxide eliniination, 20 1 contribution to gas exchange, 198 and gas exchange, 196 S k d , openings, 61 Smooth vesides, 106 Snout external nares on, 3 1 length and buccal volume, 87 in length measurement, 25 shape of, 26 Social interactions, 230-32,259 Sodium ions active transfer, 136 factors influencing uptake, 212 reguiation of, 210-14 Solar radiation, heating, 105 Solum nasi, 57 Space of Disse, 131 Spatial memory, short-tem, 239 Spawning and calling, 267 site choice, 2 6 4 6 5 Species assemblages of, 2 6 4 7 0 patterns of co-occurrence,264 Sperm, motility, 175

SUBJ E C T I N D E X
Spermatogenesis, 142, 144 Sperm extmion, 293 Sphenethmoid commissure, 59 Spides, 71, 84 Spinach, 21-22 Spinal cord anatomy, 152-55 fiber types, 154 b c t i o n of, 150-51 organization of, 153 views of, 150, 151, 153, 154 in young tadpole, 155 Spinal nerves, 153, 154 Spiracle(s) anatomy of, 80-8 1 chronology of research, 4 definition, 3 1 endouoph developnient, 180 location of, 26 morphology, 3 1,33 order of examination, 16 sinistra], 163 tube arrangements, 33 Splanchnotome, 131 Spleen, 96,97 Stages and critical thermal maximums, 199 definition, 8 and effect of temperature on growth, 25 1 and PBTs, 205-6 uncoupling of, 289 Staging systems comparison of, 8 and endotrophs, 171-72 endotrophs, comparisons of, 178 review of, 8-12 Staining, 17,20-2 1 Stamina, 195-96,224 Standard rates of nletabolism (SRM) and oxygen uptake srudies, 192 and temperature, 202,204 Stapes, 6 0 , 6 l Statoacoustic nerve VIII, 60 Stemohyoideus, 78, 79 Stemohyoid muscle, 158 Stemum, ossification of, 181 Steroid hormones, 145 STF, fixative, 12 St~Pchenzllen,104 Stiing's celis, 147 Stirn organ. See Pineal gland Stomach, 216, 249. See also Manicotto glandulare Stoniodeal-hypophysealanlage, 145 Stomodeum, 4 5 4 6 Storage improper, 295 of larval samples, 14-15 Stratum albumen, 157 Stratum griseum, 157 Stretch receptors muscular, 168 pulmonaq 197 Striatum, 160, 161 Subarcualis obliquus niuscles, 74, 77,84 Subarcualis rectus niuscles, 73-74 Subbranchialis niuscle, 66,68 Subclavian arteries, 93 Submentalis muscle, 66 Substrates, and habitat selection, 229-30 Subthymic tonsils, 101 Summer cells, 147 Superior olivary nucleus, 158 Superior olive, 160 Superior perilymphatic foramen, 60 Suprarostral cartilages, 49,6162,81-82 Suprarostral-quadrate ligament, 63 Suspension feeders, feeding apparatus in, 280 Suspensorioangularismuscles, 70,81 Suspensoriohyoideus muscles, 71,84 Suspensorium, anatomy of, 62-63 Swimming, 2 2 4 2 9 and evasion of predators, 238 kinematics of, 79 mechanical efficiency, 225 speeds and egg size, 258 and wlnerabihty to predation, 26 1 Sympathetic nervous system, 151 Sympathetic tmnk, 153 Tddpole fama, selected literature, 296 Tddpole research, chronology of, 3-6 Tddpoles as adaptation, 279-80 as category, 9 collection of specimens, 20 enigrnatic taxa, 298 Gosner equivalent stages, 92 holes formed by feeding, 221 methods of observation, 15-18 Orton's types, 52-53, 54, 89-9 1 previously misidentified, 297-98 previously unidentified, 298 rearing of, 2 1-22 Taeniae tecti marginales, 59 Taeniae tecti mediaiis, 59 Taeniae tecti transversaiis, 59 Taii bud, 178 Taii fins, 26, 179 Taiis amplitude of oscillation, 79 effects of damage to, 261 locomotion by flipping, 228,
2 29

morphology of, 28 muscle, composition of, 28 regression of, 25

in swimming, 225 waves of bending, 226 Taii tip, 28 Taste buds, 166 Tectal cartilages, 59 Tectospinal tracts, 155 Tectum, 160 Tectum nasi, 57 Tectum posterius, 59 Tectum synoticum, 59 Teeth. See a& Labial teeth calcified, 42 preparation of, 17 Telencephalon, 160 Temperate ponds, thermoreguiation in, 207 Temperature acclimation to, 206 and aggregations, 207-8 behavioral thermoreguiation, 204-5 and dissolved oxygen, 200 effect on water uptake, 2 12 and growth rates, 249, 25 1 influentes of, 229 and larval periods, 258 and locomotor activity, 209 and mass at metamorphosis, 258 ontogeny of tolerance, 205-6 and oxygen consumption, 209 PBTs, 203 physiologic effects, 204 precision of selection, 205 sense of, 168 thermal physiology, 202-10 tolerance limits, 202-4 and tolerance of hypoxia, 199 Temporas. ponds and absence of fish predators, 263-64 advantages of, 267 breeders in, 200 effect on reproduction, 242 and fitness, 279-80 and gas exchange, 200 population densities, 242 predation pressure, 291 productivity cycles of, 291 thermoreguiation in, 207 unpredictability of, 268 Terminal nerve (CNO), 160 Terminology chronology of research, 4-5 standardization of, 3 Testes. See also Gonads differentiation, 142 formation of, 144 lmal, 140 Thalamus, b c t i o n of, 160 Thermoreguiation and aggregations, 207-8 behaviod, 2045,224 factors influencing, 206-7 ontogeny of, 205-6 Thrombocytes, 97 Thymus gland, 101,102-3 Thyroid gland, 145-46 Thyroid hormones, 105, 137, 182,283

Thyrotropic cells, 145 Thyroxine, 137, 182. Seealso Thyroid hormones Tongue, 107,166 Tonoiiaments, 104 Tooth ridges gaps in, 40 and labial teeth, 3 8 4 2 notations for, 40 in oral ontogeny, 4 6 4 7 Tooth rows absence of, 36 chronology of research, 4-5 and gaps, 40-41 illustration, 35 species specific differences, 40 T o m semicirdaris, 158, 159 Touch, 168,238 Trabedae, 54 Trabedar homs, 57 Trabedar plate, 54, 57 Trabedar quadrate ligament, 63 Trachea, anatomy of, 106-7 Transepithelial potential (TEP), 212 Transport of larva, 220 preparation of samples, 15 Transversus abdominis, 78 Transversus ventraiis I1 muscle, 71,84 Transversus ventralis IV musdes, 74 Trigeminal nerve (V), 59,68, 159 Trigeminospinal tracts, 155 Trochlear foramen, 59 Tropical forests, and temperature tolerance, 202 Tropical ponds, productivity cycles of, 291 Trunci lymphatici lateraies, 100 Trunci lymphatici lateraies corpo=, 100 ris ( ) Turgidiq and egg jelly, 186 Tympanic membranes, 162, 180 Tyrnpanopharyngeus muscles, 74 Type B celis, 104 Umbelliform oral discs, 221, 222 Urinas. bladder, 131, 136, 181 Urine, and ion regulation, 212 Urobranchial process, 71 Urodaeum, 131 Urodelan branchial depressors, 72 Urogenital system, 132-40, 181 Urwirbhjktsatz,77, 78 Utrides, 158, 163, 180 Vagus nerve (IX), 61 Vagus nerve (X), 75 Vasa lymphatica caudaie dors.de, 100 Vasa lymphatica caudaie v e n d e , 100 Vascular system, right side view of, 95 Vas deferentia, 135

'ISZ 'oSZ 'I~PO" suvo3-mqr!! ~ '1a ~ o n x ~ s 9E '103 ~~o[ou!uuJI'tpplfi 81 buaurdola~ap oiroddns ~ j TTZ 50 a w d n 8 ~ >o uonpred 1 9 ~ )o irodsuea 1 "" 'aem~ sno1osrp~ u! 1E 1 ' ~ T U O ~ ~ ~ F J 9 8 5 8 1 'aze~eap uaga uo ZSZ 'uoness!sap BBa pue 'jo sTaAa1 8T '8u!q p~uaurdo~a~ap pue I SZ ' ~ u a u d o l a ~ a p 50 slaAa1 pue 781 30 UO!I&OS~E noA SSZ '30 S T ~ J 9s-SSZ 'w~JB SSZ LpMoIS pue c s l s e a ~ j o uon!qTqu! pue 'pauogrpuos
'uo!llq!lp

L 1 '8 anI8 e!Jou!A 91 'uralsiis Jepq!lsaA 6~ ' g 'aelqaira~ ~ S9T 'LR810 @SEUOJJUIOA ?E-EE '91 'aqm maA sz 'S '"uaA 102-961 '6-I61 '061 "OA O T 'P+ ~ n o x d ! ~ ! ~ ST T 'E 1T 'Z I 1 '-laA L FJluaA 6 assa^ auqal!A E-101'sarpoq hves p m a A 6 ' s v a ~ u ~ n a , ! ~ a 6 ' E U O ~@JlUJA ~ 02 '91 'sauelquau a q a ~ ! ~ 66 30 UO!UMSUOSaJ 89-997 'ua~s6s pns!A 86 30 m a u d o ~ a ~ a p 85 1 ' ! q s ! snapnu atp pue ura~sds snouaA ~ 'uon!sod a h pue 2 6 s ' s p ~ a sn!des euaA ~e~ SPIJY @ns!A S6 'E6 '~U!JA g-~z' B p x a ~ m 'uors!~ p 692 28 'uo~alays ~ a s s ~'spuod unp!m 'sauoz u o n a a B a ~ p ~ 01 'P qsnod p a s s ! ~ T Z "561 " 0 3 ~ O 01 'E qsnod p a s s r n o01 ' ? n (r) 81 'maurd~nbaoap!A apale1 appnes iunsgeqdd1 se^

LEP

x a a ~~ 3 a i a n s r

TAXONOMIC INDEX

Names of families and subfamilies are indexed only for the taxonomic accounts in chapter 12.

Acanthixalus, 3 11, 334 spimsus, 13,334 Acrls, 30,33,111,268,307,334 crepitans, 247,261 pyllus, 261, 264,268, 334 pickminji, 56 Adelastes, 172,320 Adelophryne, 172,316 Adelotw, 323,334 brevis, 334 Aabwmera, 172,173,315 hylaeahqla, 173 marmorata, 112, 173 Aermnm, 248 hydruphiiu, 205 Aeshina, 262 Aeshnu, 262 cyanea, 238,263 umbrosa, 262 AfNCalus, 3 12,334 Agalychnis, 14, 111,311 mIcanrfer; 13 callidwm, 32, 56, 88, 264 spumeili, 2 16 A g l y p t o ~ l u s332 , Ahaetulla nmuta, 228 AlbericMs, 172, 320 ~lhphiyie; 53,301 Aiophrynidae, 301 Ahodes, 112, 3 16 banWi, 55 Altigius, 320 Altiphiyioides, 172, 173,302 malcomi, 13,173 Alytes, 54, 62, 63, 70, 71, 73, 74, 75,76,77, 112, 187,306 cisternmi, 111 muletensis, 107, 111, 112 obstetnkans, 4, 55, 64,92, 104, 107, 111, 112, 125, 127, 135, 136, 137, 138, 146, 224 Ambystoma, 12,21, 30, 36,253 jraciie, 260 jeffersonianum, 262,274 macuiutum, 246,274 opacum, 253,274 talpoideum, 30,262 tignnum, 253,260,263,273 Amolops, 14,27,29,44, 51,227, 298,328,329,342 ahhana, 56 rickemi, 68, 85 sp. A, 298 sp. B, 298 Amphiuma, 198, 201 Anabaena spherica, 223,244 A w , 238,264,274 junius, 238,252,261,262, 263,274 Andiwphryne, 172,302 Anhydrophiyie, 172,175,180, 181.327 rama$, 175, 178, 179, 180, 181 Amdontbyiu, 172, 174,184,3 19 bouienjeri, 174 Amtheca, 37, 111,307 spimsa, 13, 56, 87, 88, 112,

217,219, 220,239,247, 299 Ansonia, 14,27, 36,44, 302, 334 Im@i&ita, 45, 334 Aparmphenoh, 307 Aphantophyne, 172,320 Aphdiscw, 307 Arcavomq 172,320 Arenuphyne, 172,174,324 rotunda, 174 Argenteobiu, 307 Ariequinw, 3 12 Armbates, 305 Arthroleptea, 172, 175, 179, 181, 182,327 hm'm; 175,184 lightfaoti, 175, 180 Arthroleptidae, 301 Arthroleptides, 27, 28, 228, 229, 327 murtiensseni, 228 Arthroleptinae, 301 Arthroleptis, 13, 172, 173, 176, 301 crusculum, 173 poecilonotus, 173 wahlber@, 173, 181 Ascaphidae, 301 Rcmphw, 25,27,29, 31, 33, 34, 44,48,49, 52,53,54, 58, 59, 61, 63, 64,67, 70, 71, 72, 73,74,75, 84, 85,86, 87, 88, 89, 111, 115, 164, 181, 198, 216, 270, 301, 302,334,340,345 truei, 14,33, 38,52, 53,55, 62, 63, 70, 71, 74, 80, 84, 85. 86, 97, 104, 131, 182, 185,202,203,204,205, 206,207,208,227,237, 229,244,248,261,334 Assa, 172, 175, 179, 183, 187, 229,324,340,343 darlinsoni, 13, 175 Asterophryinae, 3 19 Asterophrys, 172,319 Astylosternw, 301,334 comjatw, 27 Atelognathw, 316 reverberii, 112 Atelophyniscw, 27,227,302 chrysophorus, 14 Atelopus, 14,26,27,227, 302, 334,342 crucigeu, 187 ignescens, 32,299, 300 pachydemzw, 334 Atupuphrynus, 316 Aympanophys, 318 Aubria, 328 subsigiliuta, 23 1

Baiebrmiceps, 172, 319 Barbouruh, 172, 302 Barycholos, 172,316 Bavygenys, 172, 3 19 Batrachuphynw, 316 Ba~achyiu,181 , 316 taeniata, 181

TAXONOMIC INDEX
Batrachylodes, 172,328 Beiastoma, 261,262 oqumm, 262 Bombinn, 25,53,54,59,61,63, 70, 71, 72, 73, 74, 75, 89, 93, 111, 112, 115, 158, 225,302,334 bombina, 55, 225,238, 334 ignew, 55 orientalu; 42, 55, 135, 140, 142, 143,144, 145,246, 253,258,261,334 pachypus, 115 variegata, 42, 55,93, 107, 111, 112, 115, 117, 121, 123, 125, 132, 140, 142, 252,259 Bombinatoridae, 302 Boophis, 14, 34, 38,41,44,216, 332,334 Brachycephalidae, 302 Brachycephalw, 172,302 Brachpamphrys, 318 Breviqs, 172, 174,319 d p m u s , 57, 174, 178, 180, 181,280 fiscus, 178 Brevicipitinae, 319 Bryobatrachus, 172,324 nimbus, 13 Buergeria, 332 buergeri, 57 Buergeriinae, 331 Buf, 2, 3, 13, 14, 19,26,29, 31, 37,42, 50, 51,92, 101, 104, 107, 111, 112, 150, 155,163,164, 172,184, 191, 196, 198, 199,200, 206,207,221,225,226, 227,228,229, 231,232, 235,236,246,254,256, 261,265,269, 270,271, 272,273,274,275,276, 277,292,295,297,298, 302,303,334 americanw, 56,61,145, 195, 198,203, 204, 205, 206, 207,221,223,224,225, 226,230, 232, 233,234, 235,236,237, 238,239, 244,246,252,256,259, 260,262,263,264,268, 269,270,271, 274,275, 276,277,284 angwticeps, 56 arenamm 144,262 bweas, 146,203,207,230, 231, 232, 233, 234, 235, 236,238,261,262 buf, 56,92,93,97, 104, 106, 107, 109, 111, 112, 114, 115, 119, 121, 123, 131, 132, 135, 136, 137, 138, 140, 142, 143, 144, 147, 168, 196, 197, 198,212, 225,238,244,245,247, 262,263,269,297 calamita, 104, 111, 112, 115, 117, 119, 121, 123, 125, 244,246,250,258,263, 266,269,271,284 canoms, 1,203,204,207 cavfrons, 297 debilis, 50, 303 ewul, 203,204 f m s u s , 135 gaparimns, 137 haematiticus, 184 hemiuphrys, 203,207 japonicwfmms, 140, 142, 143,144,145 lentiginosw, 56 macro&atw, 297 maculahts, 228 marinus, 97, 136, 140, 154, 160,202,203,204,205, 206,207,217,246,247, 258,260,262,265,266, 272,276 melanomctus, 56,246 pentoni, 182 punctatus, 207,266 regularir, 56,69,70,77,84, 92,142,163 spinhsus, 56 stomaticus, 246 terveh, 158, 193, 194, 199, 200, 203, 205, 207, 209, 221, 226, 253, 273, 274 valliEep, 10,260 vevaguensis, 27, 297 viridh, 56, 135, 137, 213, 252,269,275 vulgans, 56 woodbousii, 193, 194, 196, 199, 201, 203, 204, 205, 223, 224, 232, 242, 243, 247, 250,256,259,263, 271,276,334 woodbousii@leri, 223,244, 273,274,281,292 Bufides, 303 Bufonidae, 302 Camstemum, 327 Callixalw, 312 Calluella, 320 Callulina, 172,319 Callulops, 172, 319 Calyptahyla, 308 Calyptocephalusgayi, 58 Cambams bartonii, 254 Candida humicola, 255,256 Capensibuf, 303 Cardwglossa, 26, 37, 301, 334 gracilU; 334 Caudiverbera, 3 16 caudiverbera, 55,61, 132 Centvolene, 14, 304 buckleyi, 297 Centrolenidae, 304 Ceratobatrachus, 172,175, 177, 178, 179, 180,328 guentheri, 175, 180 Ceratophryinae, 314 Ceratophrys, 13,36, 314, 334 comuta, 34,45, 55, 61,299, 300,334 cranwelli, 55 Chaophrys, 314,334 pierotci, 27,247,334 Chaparana, 329 Chaperina, 320 Chiasnwcleis, 320 Chirixalus, 186, 332 dmiae, 297 ezfingeri, 217,219, 332 idiootocus, 183 vittatus, 297 Chiroieptes, 56 Chiromanth, 227, 332 m&scens, 227,297 Chlorella, 255 Chlorolius, 3 12 Choerophryne, 172,320 Chrysemyspicta, 261 Chrysobatrachus, 312 Cochranella, 14,26, 304, 334 granulosa, 218, 334 gnfithsi, 299 uranoscopa, 334 Colostethus, 172, 173, 305, 334 albgularlr, 306 chaicopti, 173 pOtatov, 334 haydeeae, 297 nexipus, 220 nubicola, 13,51,57,299,305 semiguttata, 297 stephani, 173 subpunctatus, 57, 221,297 trinitatis, 57, 334 ii,hymperi, 112 Conraua, 14,51,329 Caphixalus, 172, 174, 320 parkeri, 174 Caphyla, 172, 174, 320 phyllo*la, 174 Cophylinae, 319 Copiula, 172, 320 fwtulans, 177 Corythomantis, 308 Crepzdophryne, 172,303 Crinia, 324 deserticola, 266 riparia, 269,270 signifera, 258,262,269,270, 271 ~rossod&lus, 111, 112,314 Cryptobatrahus, 112, 113, 172, 173.307 fih&nni, 173,179 Cryptobranchus, 198 alleganiensis, 196 Cryptothyhx, 312 Ctenuphryne, 321 Culiseta longiareolata, 275 Cycloramphus, 14,28, 172,228, 3 17,342 bwaceienk, 317 duseni, 107, 113 stejnegeri, 55,179 valae, 27, 317 Cyclorhamphus culeus, 56 Cyclorana, 310, 334 australis, 112 brevipes, 266,267 cultnpes, 202, 334 novaehollandiae, 266,267 platycephala, 56,202 Cynops spyrvhgas~er, 93 Dactylethra cape+ 55 Da.sypops, 321 Dendrobates, 13, 187, 305,334 auratus, 57,220, 221 histrioninrs, 219 pumilw, 219, 247 quinquivittntus, 334 speciosus, 2 19 tinctonus, 33, 57 Dendrobatidae, 305 Dendrophryniscus, 185, 303 minutus, 183 Dermatonotus, 32 1 muell* 57 Desmognathus ochrophaeus, 25 1, 282 Didynamipus, 172, 173,303 sjoestedti, 173 Dimiphognathus, 328 Dischidadtutylus, 172, 317 Discodeles, 172, 180, 329 opisthodan, 179 Discoglossidae, 306 Disc~lossus, 54, 59,62,63, 53, 70, 71, 72, 73, 74, 75, 112, 132,306,334 gabanoi, 269 pictus, 4, 55, 70,92, 111, 132, 135, 136, 137, 138, 142, 143,144,146,197,334 sardus, 55 Drosophila, 186 Duellmanohyla, 14,41, 308, 334 salvada, 334 uranochroa, 299, 300 Dyscophiinae, 320 Dyscophus, 320 Dytiicus, 25 1, 260 lapponicus, 262 verticalis, 260,262,263 Edulorhina, 315 Ehchistocieis, 321 ovalzs, 31 Ehchyglossa, 329 Eleuthero~lus,13, 51, 153, 171, 172, 173, 175, 178, 179, 180, 181, 182, 183, 184,317 abbotti, 173 antiliensis, 180 atkinsi, 173 audanti, 173 augusti, 173 cooki, 173 coqui, 16, 55, 171, 172, 174, 177, 178, 179, 180, 181, 186, 187, 188 comutus, 174 dimidiatus, 174 fwlen, 174 gossei, 174 gurnthen, 178, 180 hazelae, 179 hedvicki, 174, 179 imptatus, 174, 179 jasp* 13, 174, 183

TAXONOMIC INDEX
johnstwi, 174, 175, 180 martinicensis, 137, 174, 178, 180,181 mtyeri, 179 nasutus, 174, 178, 180 nubtwla, 55, 172, 174, 178, 180 parvus 174 pahciae, 174 planiroh, 174, 178 pmtmicensis, 174,180 ricordi, 181, 182 ridens, 2 18 thorectes, 174 unistnigatus, 175 u d i , 179 variam, 174 varieyi, 174 wetmwei, 174 Epipedobates, 305 anthonyi, 57 boulengeri, 57 femmalis, 275, 276 tricoiar, 57 Ericabatmhus, 328 Euparkerella, 172, 3 17 Euphlyds, 329 yanophlyctis, 142,143,199, 247 hexadactylus, 44, 56, 143, 145 Eupsophus, 13, 172, 174, 317 roseus, 174, 179 taeniatus, 174 vittatus, 174, 179 Eurycea, 290,294 Flectonotus, 13, 113, 172, 173, 178,180, 184,307 jissilis, 31 Jitzgeraldi, 173, 184 goeldii, 56, 173, 184 ohausi, 173 ~gmaeus,173,175 F d u s , 172, 303 Gallus, 93 Gambusia, 260,263 ajinis, 263 Gasterosteus acuieatus, 260 Gashophym, 206,216,321,322 carolinensis, 22,28, 31, 32,47, 57,67,68, 202, 203, 204, 205,224,253, 267, 291, 299 usta, 57 Gashophrynoides, 172,321 Gasti-otheca, 13,22,34,73, 74, 75, 111, 112, 113, 171, 172, 173, 179, 180, 182, 186, 187, 188,290, 307, 340 ceratophyes, 173, 185 chmktiani, 173 emestoi, 173 espeleh, 56 gracilis, 56, 181 marsupiata, 56 wophylax, 56 myera, 173 peruana, 56 pseustes, 56 riobambae, 56, 171, 172, 175, 177,178,180,181 testudinae, 173 weinlandi, 173, 185 Genyophryne, 172,320 Genyophryninae, 320 Geobatmhus, 172, 3 17 Geocrinia, 172, 325 rosea, 185 victmiana, 132, 183, 185 Gephpmantis methueni, 297 Glyphoglossus, 321 Gynacuntha membranalis, 238 Hamptophyne, 321,334 boliviana, 57, 78, 334 Heieiopurus, 186,324,334 aushalincus, 334 eyrei, 168 Heieophyne, 14,27, 33, 34,41, 44,45, 53, 58, 61,63, 64, 88,306,324,334 natalensis, 55, 87, 112 purcelli, 32,42, 55, 63 regis, 334 Heleophrynidae, 306 Hemianaxpapuensis, 260 Hemicwdulia, 263 tau, 262, 263 Hemdacplium scutatum, 282 Hemiphractinae, 307 Hemiphractus, 13, 113, 172, 173, 187,307 scutatus, 185 Hemisotidae, 306 Hemisus, 28, 34, 37, 53, 220, 306,334 juttatum, 334 marmoratus, 220, 239 Heterixalus, 3 12 amulti, 297 betsiieo, 297 Hildebrantia, 329 Holoaden, 172,317 Hophbaz~achus, 329 rujuiasus, 42,44 ,?&&nus, 13,44, 56, 64, 93, 101, 142,209, 246, 247, 251,256 Hoplophym, 13,68,71,74,320 r* o, 27,29, 32,67, 228 uluguruensis, 228 Huia, 14,29,227,329 Hyalinobatrachium, 14, 305 euygnatha, 37 feischmanni, 87, 112 uranoswpa, 48 Hydroiaetare, 315 Hyla, 13, 14, 28,29, 37,41,44, 48,50,56, 112, 150, 153, 221, 225, 229, 238, 254, 268,269,271,272, 273, 276,277,297,308,334 abbreviata, 27 andwsonii, 248,263,273,276 arborea, 56, 107, 111, 112, 115, 140, 143, 225,238, 247,248,263,269,271 arborea savignyi, 133, 135, 137 armata, 218, 334 avivoca, 30,41, 268 bwbeba, 42 boans, 247 bogotensis, 34 bmmeliacia, 13,48,217, 299, 334 celata, 297 chrysoscelis, 30, 38, 41, 46, 224, 246, 253,254,261,262, 263, 264,266, 272,273, 274,276,277,286 c i m a , 30, 56, 158, 161,229, 254,258,268,282,284, 286 dendroscaea, 48, 112 ebraccata, 112 femoralis, 30,45, 223, 259, 260,272,276,285 geographica, 40, 132,231, 232, 238,247,261 jratiosa, 38, 39,223,250, 253,258,259,260,262, 272,273, 274,276,282, 284,285,286 iuncz@rmis, 56,78 iegleri, 88 hcophyllata, 13, 26,28, 31, 37,44,50,297 lindae, 299, 300 iaquax, 297 marmorata, 13,28, 37 meridionalis. 269 microcepha~, 13,37,56,217, 299,300 minuta, 41 nana, 37,56 parviceps, 13, 37,297 phlebodes, 112 puuoi, 71, 334 piees, 44,45 plaiydapla, 297 pseudopuma, 219,246, 247, 251,262,266 pulchella, 56,257 riimlaris, 299,300 rosenbergi, 56 rujtela, 112 sarayacuensis, 47, 112, 334 siopela, 297 squirella, 56,258,284 verstwiar, 30,224,253,263, 264,276 zeteki, 13, 71,247, 297 Hyiurana, 29 Hylidae, 306 Hylinae, 307 Hyiades, 315 Hylodinae, 3 14 Hyiamantis, 311 Hylophwbus, 172,319 Hylorina, 317 Hymenochirus, 63, 74, 75, 76, 216,219,326,334 boettgm; 13, 55, 61,63, 70, 71, 72, 84, 87, 88,90, 104, 113,125,127,247 curtipes, 334 Hyophym, 172,321 Hyperoliidae, 311 Hyperoliinae, 311 Hperolius, 312 marmoratus taeniatus, 252 picturatus, 297 vindzfimrc, 252 v. nitidulus, 252 v. m m a t o ~ u s 252 , Hpopachw, 322 aldmenter, 322 barbm; 57 varioiasw, 57 Indirana, 51,329 beddarei, 228 jundia, 13 ieithii, 228 Ingerana, 329 Insuetopbqmus, 3 17,334 urpicw, 334 Ischnomema, 172, 317 Iblophrynus, 172,297,322 pleurostiigma, 174 Kaloula, 230, 322 bwealis, 245 pulchra, 33, 158,230 mg@ra, 297 Kassina, 13, 28, 313, 334 maculata, 297 senegalensis, 28, 30,44297, 334 Kassininae, 3 13 I(asinula, 313 K w a n w . 172.174.324 Lunzarana, 329 Lutes calcarifer,262 Lutimeria, 290 Luurentophyne, 172,303 parkeri, 173 Lebiasina panamensis, 262 Lechriodus, 324 jktcheri, 13,247 Leiopelma, 52, 53,54, 59,63,89, 172, 173, 178, 183, 187, 290,294,3 13 archtyi, 13, 55,63, 173, 180 hamiltuni, 13,173,180 hochstetteri, 55, 173, 180 Leiopelmatidae, 313 Lcpuiobatr~us,31, 38,42,44, 219,296,314,334,340 W s , 25, 32, 55, 113, 127, 314 Uanensis, 33,55, 314, 334 Lepomis yanelus, 262,263 mamochirus, 262,263,264 Leptobrachella, 3 18,334 gracilii, 334 mjobergi, 297 Leptobrachium, 3 18 gradis, 297 hasselti, 55, 112, 297, 334 hendricksonii, 297 montanum, 135, 144 n&rops, 334 Leptodactyhdae, 313 Leptodactylinae, 315 Lqtodaqiadon, 13,27,301,334 venhmannoratus, 44 Leptodaqlus, 13, 30, 183,201, 231,255,315,334

442
Leptoductylus (continued) albilabns, 66, 132,203,204 bolivianus, 2 19 bujbnius, 132 chaquensis, 42, 55 knudseni, 112,265 melamnotus, 334 mystaceus, 227 ocellatus, 55 pentndqlus, 55,93,217, 247,315,334 w a g d , 112 Leptolah, 318 gracilis, 48, 135, 144 Leptopelinae, 313 Leptopelis, 227,228, 313, 334 hykdes, 227,299 natalensis, 228, 334 Leptophryne, 303,334 borbunicll, 334 Lethocerus amencanus, 262 Leucwvhinea dubia, 262 Libellula herculea, 262 Limnodynastes, 272, 324 m t u s , 246, 247,257,258, 262,266,272 peronii, 55,209 tasmaniensis, 55,266 Limnodynastinae, 323 Limmmedusa, 317 Limmnectes, 329 cancrivms, 132,202,203, 210,213 kwalensis, 142 w o d o n , 297,298 mierodisca, 229, 298 paramacrodon, 298 Lithodytes, 315 Litwia, 29, 36,38,41,44,88, 267,310 alboguttata, 112,266 buolar, 266 mmlea, 266 citropa, 4 1 ewingi, 106, 197,256, 262 f a l h , 266 genimaculata, 248,264,269, 270,271 gracilenta, 266 inennis, 266, 272 nannohs, 248,269 nasuta, 266 nyakalemis, 297 rothi, 266 rubella, 202,247,266 subglandulasa, 37, 38,41 xanthomera, 264 Lysapsus 34, 327

TAXONOMIC INDEX
bicalcaratus, 297 brevipalmatus, 297 curtus, 297 femoralis, 334 guttulatus, 38, 51 lugubris, 44,45, 51,219,332, 334 opiparus, 13,334 pulcher, 228 ulcerosw, 297 Mantophrp, 172,319 Megmlosiu, 315 goeldii, 42 Megalopepus caerulatus, 262 Megktolahs, 324 lignaks, 112 Megophtyidae, 318 Megophrys, 13, 36,51,61, 171, 172,319,334 montana, 55,222,239,299, 300 nasuta, 55, 135, 144 pawa, 55 robusta, 55 stejngeri; 55 Melanobatrachinae, 320 Mehbatrachus, 172,320 Melanophryniscus, 303 Menstogenys, 14, 29,227,329, 330 orphmcnemis, 45 phaeomms, 298 poecilw, 298 Mertensophyne, 13.27,303 Metacrinia, 325 Metaphrynella, 31, 322 Metaphryniscus, 172, 303 Mimzalus, 330 Mimbatrachella. 297. 328 berdmorei, 112 bomeensis, 297 heymonsi, 13, 31, 37, 51, 112, 222,334 omata, 57,263 pulchra, 57, 67,68 Microhylidae, 319 Microhylinae, 321 Mieropterus salmoides, 260 Minyobates, 305 Mixophyes, 324 fasc~latus, 5 5 schhlli, 248, 269, 270 Myersiella, 172, 174, 322 microps, 174 Myobatrachidae, 323 Myobatrachinae, 324 Mvobatrachw, 172, 175, 179, tomieri, 13,173, 178, 179, 180 viviparus, 13 Nelsonophryne, 38,51,319, 322, 334 atewimus, 334 Neobatvachus, 324 Nephalobates, 306 Nesirmdus, 313 Nesomantis, 172, 333 Nimbaphynoides, 172,173,303 occidentalis, 13, 140, 142, 144, 173, 179, 181, 182, 184, 303 Notaden, 324 Nothophryne, 328 Notonecta glauca, 262 indica, 262 unulata, 262 Notophthalmus, 277 videscem, 254, 261, 263,274, 275,276,291 videscens dorsalis, 264, 273, 277 ~vctibatrachus, 330 pygmaeus, 297 Nyctimantls, 308 Nyctimfles, 310 cheesmanae. 27. 297 montana, 297 Ny&alus, 13, 333 Occidozyga, 13,28, 37,44, 51, 229,330,331 bis, 56,127 lima, 127,299 Odontupblynus, 73, 314 achalensis, 56 oceidentalis, 40, 112 Ola[rgonabbrepiata, 27 Oncorhynchusmykiss, 25 1 Ophryophryne, 319 Opisthothyh, 313 Oueophryne, 172,174,320 annulata, 174 Oreaphrynella, 172,303 n&a, 173 Ortbetwm caledonuum, 260,265 Osmnophyne, 172,303 Osteocephalw, 297,308,310 lepneuni, 2 17 oophagus, 217,219,220,239, 247 pybumi,9, 14,31,33,37,50, 57,218,219,297,298,334 robusta, 57,297 Oxyglssus, 56

Pna, 330
Padymedusa, 311 dunzicolar, 14, 87 Palmatwappiu, 172,330 Pandanus, 228 Pantalafivescens, 260, 26 1, 262 Parmassina, 313 Paracriniu, 325 Paradoxophyla, 323 palmata, 14, 31 Paratelmatobiw, 315 lutzii, 112,297 Parhopiupbryne, 172,320 Pedostipes, 303 Pelabates, 13, 54,61, 67,68, 70, 71, 72, 73, 74, 75,81, 111, 112,269,325 cultripes, 246 jscw, 4, 53, 55, 71, 75,80, 84, 107, 112, 115,135, 136,138,248,269 syriacus, 55, 68, 269 Pelobatidae, 325 Pelodryadinae, 310 Peladryas, 310 Peladytes, 53, 58, 71, 75, 76,77, 267,326,334 punctatw, 55,107,246,334 Pelodytidae, 326 Pelophryne, 13, 172,304 bvevipes, 173, 179 Peltophryne, 304 Petropedetes, 14,28,228,328 camunemis, 228 natatm, 328 parkeri, 228,239 Petropedetinae, 327 Phmahvla. 3 11.334

Nannophrys, 28,51,330 ceylanensis, 14, 34, 228 Nanmrana, 330 Natalabatrachus, 328 Nectophyne, 172,173,303 afia, 173,231 Nectaphrynoides, 171, 172, 173, 179, 180, 181, 183, 184, 303

gryllw, 123,125,130 hosii, 175 lissobrachiw, 175 schmacheri, 175 vanabilis, 57,297 Philana, 13, 172, 174, 179, 180, 183, 184,267,324 fiosti, 174 Phlyctimantis, 313 leonardi, 297 Phvynella, 172,322 Phvynobatrachus, 37,328 natalensis, 143 Phrynodon, 13,172,184,328 Phrynoglossus, 229,330 Phrynohyas, 13,40,308,310 resinijifirrclc,217,218 Phrynomantis, 67,84,86,323 annectans, 57,67, 84, 86 mierops, 261 Phvynomedusa, 311 Jimbnata, 13

TAXONOMIC INDEX
Phrynomerinae, 323 Phlynomerus, 57,67 annectens, 67 Phynqus, 172,317 Phylbbates, 306 bicolov, 57 IugubnS, 34 Phyllodrtes, 308 Phyllomedusa, 14,29, 111, 231, 261,311,334 boliviana, 56 sauvagii, 56, 132, 140, 144 tarsius, 238 tomoptema, 275, 276 trinitatis, 56, 84, 261 vaillanti, 232, 247, 334 Phyiiomedusinae, 3 11 Pihyllonastes, 172, 317 Physaiuemw, 29, 261,262, 315 pctersi, 316 pustulosus, 56, 261,262,291 Phyzelaphyze, 172,317 Pipa, 13, 14,63, 71, 72, 73, 74, 75, 79,90, 113, 172, 175, 178,180,187,326 americana, 55 awabali, 266 cawalhoi, 55, 61, 71, 132 monstrosa, 55 pawa, 55,71 p$a,55,71,74,75,175, 179, 180,187 Pipidae, 326 Pipinae, 326 Pitangw sulphuratus, 262 Platymantis, 13, 172, 175, 178, 180,331 guentheri, 175,178 hazaiue, 175 mqeri, 175 pelewensis, 181 Phtypelis, 172, 174, 320 grandi, 174 Plectrohyla, 14,309 ixil, 44, 45 matudai, 44 teuchestes, 297 Pietbodontohyla, 172, 174, 184, 320 inpinalts, 177 notostida, 174 Pleurodema, 34, 316 bibronii, 56 burellii, 56, 262 cinerea, 112 Polypedates, 333, 334 ieucomystm, 147,158 mamotis, 298,333 mulatus, 245 megacephala, 334 otibphus, 333 Pqvntonia, 328 paludicola, 28,29 Probrepiceps, 172, 319 Proceratophys, 3 14 Prototheca, 253 richardri, 255 Pseudacti, 12, 13,29,206, 229, 271,273,276,309 hacLyphona, 297 crtrcifer, 30,41,56,61, 207, 223, 233, 234,235, 236, 244,249,250,252,253, 254, 259, 260, 261, 262, 263,264, 268, 271,273, 274,275,276,282,283, 286 oculart, 242,268, 269 ornata, 22,204,205, 242, 268 regilla, 34, 56, 203,207,230, 231,232,233,235,236 meckeri, 221 mkeriata, 41, 202, 203,204, 205,206, 221,223, 224, 252,259,260,263,264, 274,285 triseriataferiarum, 246 rrism'ata m u l a t a , 224 Pseudhymenochirus, 37, 75, 326 Pseudidae, 327 Pseudis, 3, 30.67, 327,334 229,230, 231, 232, 233, 234,235,236,237,259 catesbeiana, 9, 31,38, 53, 56, 58,63, 65,67, 68,69, 70, 71, 72, 74, 75, 80, 81, 82, 83, 84, 87, 93,96, 97, 100, 101, 103, 106, 107, 111, 125, 131, 132, 136, 138, 140,144,146,153,154, 155, 158, 161,162,189, 190, 191, 193, 194, 195, 197, 198,200,201, 202, 203, 204, 205, 206,207, 208,209, 210, 211, 212, 223, 224, 225, 226,231, 238,239,244,245,247, 248, 250,251,256, 261, 263,271,272,275,292 chalconota, 30 chiriuthuensis, 28 clamitans, 56, 107, 111, 140, 203,204,205,206,208, 209, 211, 212, 214,216, 223,225,226,230, 238, 239,244,245,246,247, 251,253,255,260,262, 263,274,275,293 mrtipes, 56 dalmatina, 56,92, 107, 133, 135, 137, 138, 143,225, 238,251,263,269 esculenta, 56,93,96, 100, 101, 103, 104, 106, 111, 115, 133, 135,136,137, 138, 140, 142, 143, 145, 147, 153, 158, 160, 163, 168, 263,269,271,272 fmvm; 297 fwciguh, 67,84 ghndulma, 31 grayi, 183 heckcchwi, 30, 31,201, 231, 239 hosii, 297 japonica, 104, 107, 140, 144, 146,297 latastei, 140 lessonae, 111, 263,271,272 Luctuosa, 298 macrodon, 297,298 maculata, 298 mimodisca pahvanensis, 298 montezumae, 147 muscosa, 194, 203, 204, 207, 21 1,212,263 nicobariemis, 298 nkromaculata, 131, 137, 140, 142,144 occidentalis, 101 palmipes, 14,299, 300 palustni, 266,270,274,275 paradoxa, 3 perezi, 267, 269 pzpiens, 8,21,31,41,44,56, 92, 95,96, 97, 100, 101, 102, 103, 105, 125, 133, 135, 136, 137, 140, 143, 144, 145, 146, 153, 157, 158, 159, 160, 161, 166, 167, 168, 194, 196,203, 204,205,206,207,208,

443
209, 211, 214, 223,224, 231, 233, 234,235, 236, 237,239,244,247,251, 254, 255,256, 263, 270, 271,272,273,292, 297 pleshi, 298 pomsa, 144 pretiosa, 232,233, 234,235, 236,252,270 pwtlrlosw, 14 ridibuma, 93,97, 101, 106, 123, 125, 127, 129, 136, 137,140,142,144,145, 160,255,269,271 rugosa, 140, 144 septenhionaiis, 225, 226,239 sphenocephaia, 28,42,45, 193, 194, 199,200,203,205, 206,207,209,223,232, 242,244,247,253,255,
sylvatica, 4,8, 87, 88, 107, 111, 135, 137, 145, 202, 203,204, 205, 208, 209, 211,216,217,219,223, 224, 230, 233, 234,235, 236,239,244,247,249, 251,252, 253, 254, 255, 256,257,260,262, 263, 271,272,273,274,281, 284,286,287,288,291 temporaria, 53, 56, 57, 58, 59,61, 64,65, 67,68,69, 70,72,73, 74,75,77, 80, 93, 94, 101, 104, 105, 106, 107, 108, 109, 111,

australis, 55 bibronii, 55, 112, 194,262, 270 semimammata, 270 Psyllophv, 172,302 Pternohyla, 309 fodiens, 297 Pqchadena, 331, 334 anchietae, 297 oxyihynchus, 297 Ptycbohyla, 14,111,112,309 leonhardschulizei, 11 112 1, Pyrrhulina, 262 Pyxicephalus, 231,331 adspersus, 219,231,247

331' agilis, 56 albohbris, 297 alticoh, 30 amurensis, 246 areolata, 262 awalis, 115,259,262 aumva, 229,230,232,233, 234, 235,236,248,251, 252 berlandieri, 41, 192, 193, 194, 195, 198,209,223,224, 226,261 boylii, 88, 207, 23 1 wcadae, 203,205,206,207,

128; 132; 135; 136; 137; 138, 140, 143, 158, 161, 196,197,211,243,244, 246,247,251,266,267, 271 urnbraculata, 298 vaillianti, 30 vertehali, 29,298 warschm~chii, 262 whiteheadi, 56 Ranidae, 327 Raninae, 328 Relictovomev, 322 Rhacophoridae, 331 Rhacophorinae, 332 Rhacophorus, 28,180,298,333 arborew, 140, 142, 144,245 bimaculatus, 298 gauni, 184,298 lewomystm, 57,298 sp. B, 298 viridi, 246 Rhamphophyne, 172,304 Rhantus exoletus, 262 guticollis, 262 Rheobatvachus, 13, 172, 175, 324,340 silus, 142, 175 Rhinoarma, 13,53,172,175, 179,180, 181, 183, 187, 333,340,343

444
Rhinodemza (contznued) da+nii, 56, 131, 175, 184 rufum, 131 Rhinodermatidae, 333 Rhinophrynidae, 333 Rhinaphlynus, 14, 33,38,59, 61,63,71, 75,90, 108, 109, 112, 115,333, 334, 340 domalis, 33,47,53,55,63,88, 112,201,216, 217,231, 232,247,298,334 Rhombophyne, 172,174,320 testudo, 174 Salmo clarbii, 260 Scaphiophyne, 33,36,38,44, 51, 107, 111, 112, 123, 179, 319,323,334 Scaphiophryninae,323 Scaphiopus, 13, 55,67,68, 72, 73, 206, 221, 229,231, 265,271,273,274,277, 288,297,325,334 couchii, 132,202,203,244, 252,258,259,263,271, 281,286,334 holbrookii, 32, 38, 55,87, 158, 201,203,204,247,250, 256,267,273,274,277 Scarthyiu, 36,297,309,334 omwdactyiu, 334 Schismaderma, 231, 304,334 carens, 27, 32, 334 Scinax, 13, 28, 309, 334 acuminata, 56 elaeochroa, 309, 334 nebulosa, 299 rostrata, 36, 50,297, 309 wbra, 28,50, 56,238, 309 stauffm; 217,246,309,334 sugiliuta, 309, 334 Scotobeps, 301 Scutuq 14,319 mammata, 298 S@brophys, 317

TAXONOMIC INDEX
Semnodactylus, 313 Silurana, 14,63,90,326 tvopicalis, 55, 61, 70, 71, 74 Szren, 36, 198,201,291 intermedia, 275 lace&'na, 193 Smilzsca, 111, 262, 309 phaeota, 262, 266 Somuncuria, 3 17 Sooglossidae, 333 Sooglossus, 172, 175, 183, 333 gardineri, 13, 175, 185, 186 secheensk, 13, 187, 229 Spea, 13,27,40,41,44, 55,61, 73,206,231,246,247, 297,325,326 bombtfins, 40,42,55,73,75, 78, 87, 88, 223, 247 hammondii, 203,210,247 intemntana, 230,232,234 multiplpliutta, 132, 230,246, 247,250,263,271 Speiaeaphqwe, 172,319 Sphammhynchus, 310 Sphagnum, 213 Sphenophyne, 172,320 Spicospina, 325 SpPzmphrynoUies, 304 Staurois, 36,41, 331 iutopalmatus, 298 natator, 29,227, 298 Stefania, 113, 172, 173, 307 goini, 179 sculue, 173 Stephopaedes, 28,303,304,334 anotis, 13,299, 334 Stweocyclops, 28, 31, 33, 319, 322 Stumpjh, 171,172,179,320 grandis, 174 Sympetwm n&rocentrum, 262 Synapturanus, 13, 172, 174, 323 salseri, 174 Syncope, 172, 174,323 a t* ne 174
Tachycneminae, 313 Tachycnemis, 3 13 sqchellensis, 183 T a u ~ 1 u s14,41, 325 , Taylwana, 172, 331 Telmatobiinae, 316 Telmatobius, 318 bolivianus, 56 ceiorum, 56 C U ~ U S 56 , jehkii, 56 iuticeps, 56 maniwratus, 56 pisanoz, 56 Telmatobuf, 34,318 australis, 227 Tepuzhyiu, 31 O Thamnapbis, 219 Theloa'mma, 13,333 wrinensis, 29 Thonpa, 14, 28, 111, 112,228, 318,334,342 miliaris, 27, 87, 107, 113, 218,334

Wernevia, 304,334 pvewsz, 334 Woltento@na, 172,304 Xenobatvachus, 172,319 Xenohyla, 30 Xenopodinae, 326 Xenapus, 8, 14, 16, 24,25,28, 29, 31, 34, 38, 54,63,70, 71, 72, 73, 74, 75, 77, 79, 80, 84, 86, 87, 88,90,92, 103, 104, 107, 112, 113, 115, 117, 119, 121, 123, 125, 127, 131, 132, 136, 137,138,140,144,145, 146, 150, 153, 154, 155, 157, 159,163,164,165, 168,216,221,224,225,

Tmnoptema, 331 TracLycephalus,40, 308, 310 Trama, 274 carolina, 274 lucerta, 253,261,262,264 Trichobatvacbus,27,30 1, 334 robustus, 14,44, 334 Triprion, 310 Triturusaulgan, 262 Truebeliu, 172, 304 Uperodon, 31, 323 systoma, 57,142 Upwoieia, 13,267,325 lithomoda, 55,266 mimuiu, 266 Urtka, 22

The Universi!

P Chicago F k

I S B N 0-22b-SS7b2-