Estuaries Vol. 20, No. 3, p.

534-546 September 1997

Rate of Tube Building and Sediment Particle Size Selection During

Tube Construction by the Tanaid Crustacean, Leptochelia dubia
LYNNE D. KRASNOW1
Oregon State University
College of Oceanography
Corvallis, Oregon 97331

GARYL. TAGHON2
Rutgers, the State University of New Jersey
Institute qf Marine and Coastal Sciences
New Brunswick, New Jersey 08903-0231
ABSTRACT: In estuaries, organic coatings play an important role in the aggregation o f mineral particles. Particles
acquire adhesive surfaces through the activities of bacteria and microalgae in the sediment and water colunm. Eventually,
they may become incorporated into larger aggregates and structures, such as tubes, constructed by infannal benthic
animals. Where these structures are large enough, and the adhesive bonds between particles strong enough, individual
particles may remain in place at bed shear stresses otherwise strong enough to cause sediment transport. This study
examined the aggregation of particles during tube'building by the ubiquitous tanaid crustacean Leptochelia dubia. Particle
size selection and rates of tube building were determined as functions o f animal size, temperature, and the presence or
absence of bacteria and microalgae. These data were used to model seasonal patterns of sediment binding by a popu-
lation of L. dubia in Yaquina Bay, Oregon, taking into account seasonal changes in sizes and abundance of animals. Rates
of tube building (mass of sediment per day) increased with animal size, but temperature had no effect. The m o d e l
indicated that the field population of L dubia bound sediment into tubes at a gross rate of 350 g m ~ d -1, averaged over
a 2-yr period. Seasonally, gross rates of tube building were predicted to range from 70 g m 2 d 1 (during late winter-
early spring) to 600 g m -2 d -1 (during autmnn). When constructing tubes from sterile sediments, small animals selected
silt-sized particles while large animals discriminated against these particles. The presence of microbes in sediments
tended to reduce particle selectivity.

Introduction polychaete Owenia fusiformis. Rhoads et al. (1978)

An important component of the particle budget reported that the stability of sediment from Long
of an estuary is the degree to which particles set- Island Sound increased after seeding with another
tling through the benthic boundary layer are sta- tubicolous polychaete, Heteromastus filiformis, and
bilized against resuspension and erosion. The sta- suggested that the effect was due to binding of the
bilization of deposited particles is relevant to the sediment surface by mucous fihns associated with
shallowing of harbors and navigation channels enhanced microbial production. Eckman et al.
(Mehta 1986; Dyer 1989), to short-term fluctua- (1981) suggested that tubes promote microbial
tions in the amount of light reaching the micro- growth, and thus increase mucous binding of sed-
phytobenthos (de Jonge 1994), and to the rate of iment, by enhancing diffusional fluxes of nutrients
supply of food resources to benthic deposit-feeding through the benthic boundary layer. Tubes may
organisms (Nowell et al. 1981; Geesy 1982; Luck- also enhance ,microbial production by reducing
enbach et al. 1988; Miller and Sternberg 1988; Em- abrasion, which otherwise limits the distribution of
erson 1989). microflora to cracks, crevices, and other surface
Interest in the role of the tubes of infaunal ma- concavities (Meadows and Anderson 1967; De-
croinvertebrates in the stabilization of sediment Flaun and Mayer 1983; Miller et al. 1987; Miller
was stimulated by Fager's report (1964) that wave 1989). Because irrigated tubes provide an oxygen-
surge did not resuspend sediment in La Jolla Bay ated microhabitat below the depth of the chem-
colonized by dense aggregations of the tubicolous ocline (Meyers et al. 1987), the inner walls of tubes
are sites of intense microbial activity (Aller and
Ylngst 1978; Alongi 1985).
1Present address: National Marine Fisheries Service, 525 NE Where tubes form dense mats, tube building it-
Oregon St., Suite 500, Portland, Oregon 97232.
2Corresponding author, tele: 908/932-6555 x547; fax: 908/ self is a quantitatively important component of sed-
932-8578; email: taghon@ahab.rutgers.edu. iment stabilization. Polychaete worms and peracar-

9 1997 Estuafine Research Federation 534

 Tube Building by a Tanaid 535

id crustaceans (notably amphipods and tanaids) fide. Water temperature varies seasonally from 6~
are numerically the most important tube-building to 16~ Air temperature ranges from 0 ~ to 30~
macrofauna in most sedimentary habitats. In the (Davis 1981). Hours of sunlight range from 12.2
Bay of Fundy, maldanid and spionid polychaetes d -1 in June to 6.0 d -1 in December. Sediment at
incorporate up to 50% of the upper 5 cm of inter- the study site is a well sorted, very fine sand with
tidal sediment into tubes (Featherstone and Risk a median diameter of 96 p~m (G. Ditsworth, United
1977). Resin casts of intertidal sediment from the States Environmental Protection Agency, Newport,
Clyde Estuary, Scotland, reveal densely packed Oregon, unpublished data). Pennate diatoms dom-
tubes and burrows of the amphipod Corophium vol- inate the sediment microalgal community (Am-
utator and the polychaete Nereis diversicolor (Mead- spoker and Mclntire 1978). Numerically abundant
ows et al. 1990). Flume experiments showed high macroinfauna are tanaid (Leptochelia dubia) and
densities of juvenile Corophium volutator reduce amphipod (C0rophium salmonis and Eobrolgus spino-
bedload transport rates by binding sediment into sus) crustaceans, bivalve molluscs (Macoma balthi-
their burrow walls (Grant and Daborn 1994). Ag- ca), polychaete worms (Malacocerus glutaeus, Lum-
gregations of sabellariid worms create geological brineris sp., Pygospio elegans, and unidentified capi-
formations called "worm reefs" (Main and Nelson tellids) and unidentified naidid oligochaetes (Kras-
1988; Pawlik et al. 1991). The types and sizes of now 1992). Nematodes, harpacticoid copepods,
materials incorporated into tubes appear to de- and ostracods dominate the meiofaunal fraction of
pend on the availability of particles in the sur- the community.
rounding environment and on the selectivity and Tanaids used in experiments were females and
particle-handling capabilities of the species in- juveniles greater than 0.8 mm carapace length.
volved (Featherstone and Risk 1977; Lewis 1968; Male L. dubia were excluded because they have re-
Gr~mare 1988; Self and Jumars 1988). duced mouthparts, do not feed, and build tubes at
Tanaid crustaceans are distributed in benthic a slower rate than females or juveniles (Mendoza
habitats ranging from the deep sea to the intertidal 1982; Highsmith 1983). Furthermore, while sea-
(Sieg and Winn 1981; Holdich and Jones i983; sonally abundant, males are less than 15% of the
Sieg 1986). On the high-energy tidal flat of an Or- year-round population (Krasnow 1992). To stan-
egon estuary, the tube-building tanaid Leptochelia dardize the physiological condition of the experi-
dubia achieves year-round densities of approxi- mental animals, only tanaids with guts at least half
mately 150,000 m -2 (Krasnow 1992). By virtue of full were used.
its high abundance and extensive tube-building ac-
tivities, L. dubia is expected to affect the granulo- VISUAL OBSERVATIONS OF TUBE-BUILDING
metric and hydrodynamic properties of the sedi- BEHAVIOR AND TUBE CONSTRUCTION
ment in which it lives. In the present study, we ex- We observed tube-building behavior of animals
perimentally determined rates of tube building in sediment composed of crushed kryolite (a so-
and particle-size selection by L. dubia as functions dium, aluminum, fluoride mineral). Because of its
of animal size, temperature, and the presence or refractive index, kryolite in seawater is virtually
absence of bacteria and microalgae in the sedi- transparent, making it easy to observe the activities
ment. of animals below the surface. Particle size was gen-
erally similar to sand from the field site, and ani-
Materials and Methods mals readily built tubes with the kryolite particles.
Tube microstructure was studied using scanning
ORGANISMS AND STUDY SITE electron microscopy. Tubes built from clean sand
Tanaids used in the laboratory and field exper- (previously heated at 500~ were fixed in 0.2 Ixm-
iments were collected from a site on the intertidal filtered glutaraldehyde at a final concentration of
sandflat just east of the Oregon State University 2.5%. Preparation of fixed samples for SEM fol-
Hatfield Marine Science Center on Yaquina Bay, lowed standard procedures (Wardell 1988).
Oregon (44~ 124~ The study site en-
LABORATORY EXPERIMENTS
closed an area of 640 m 2 located at an elevation of
+1.0 m above mean lower low water. Yaquina Bay Experimental Conditions
is a seasonally well-mixed estuary (Kulm 1965). A 3 • 3 factorial experimental design was used
Freshwater inputs to the bay are low during sum- to estimate rates of tube building and particle-size
mer so that salinities of 30-33% are maintained (P. selection by L. dubia. Factors investigated were tem-
Henchman, Oregon State University, Hatfield Ma- perature (T1 = 5~ T2 = 10~ and T3 = 20~
rine Science Center, Newport, Oregon, unpub- and sediment treatment (A = sterile sediment, and
lished data). During winter, rainfall and increased B and C = two different cultures containing sedi-
river flow reduce salinity to as little as 6%0 at low ment-associated microbes).
536 L. D, Krasnow and G. L. Taghon

For each cell in the 3 X 3 e x p e r i m e n t a l design, meiofauna, a n d silt particles) was p o u r e d o n t o the
40 tanaids were assigned at r a n d o m to individual sterilized f o u n d r y sand in acrylic culture dishes.
60-mm d i a m e t e r petri dishes, which were half-filled T h e culture m e d i u m (Guillard's f / 2 , diluted to
with e x p e r i m e n t a l sediment. T h e e x p e r i m e n t last- f / 2 0 to p r o m o t e bacterial adhesion) was d r a i n e d
ed 24 h. T h e t e m p e r a t u r e s e n c o m p a s s e d those t h r o u g h the 20-~m Nitex mesh o n the b o t t o m o f
m e a s u r e d at the study site d u r i n g a calendar year the culture dishes and r e f r e s h e d every o t h e r day.
(Davis 1981). Because only t h r e e temperature-con- At the same time that the culture m e d i u m was
trolled r o o m s were available, this factor could n o t changed, the sand was gently b u t t h o r o u g h l y
be replicated. Dishes were placed, in r a n d o m or- stirred to p r e v e n t the build-up o f anoxic condi-
der, u n d e r a rack o f f o u r alternating standard and tions. Particles smaller than 32 p.m d i a m e t e r were
wide-spectrum Sylvania Gro-Lux fluorescent bulbs. o m i t t e d f r o m the e x p e r i m e n t a l s e d i m e n t because
Light intensity u n d e r each rack averaged 110 + 18 some fraction o f t h e m would have b e e n lost
IxE m 2 s-1 (Li-Cor LI-1000 Data L o g g e r light me- t h r o u g h the mesh d u r i n g the drainage p r o c e d u r e .
ter, 4"tr m o d e l SPH Q u a n t u m sensor). P h o t o p e r i o d It was o u r i n t e n t that the microbial culture
was 12L:12D. r e a r e d at 17~ contain h i g h e r standing stocks o f
Tanaids were provided with o n e o f three exper- bacteria and microalgae than the culture r e a r e d at
imental sediments in which to build a tube: sterile 10~ Microalgal standing stocks were estimated as
sand, sand p r e c u l t u r e d at 17~ for 6.5 wk, and sand chlorophyll a content, m e a s u r e d by the m e t h o d o f
p r e c u l t u r e d at 10~ for 4 wk. Commercial f o u n d r y L o r e n z e n (1967). Bacterial standing stocks were es-
sand, p r e d o m i n a t e l y silica a n d free o f detrital or- timated by direct counts using epifluorescence mi-
ganic material, was used as the substrate. Sand was
croscopy. Samples were p r e p a r e d for c o u n t i n g fol-
separated into < 63 ~t, 63-125 p~, 125-250 Ix, 250-
lowing the p r o c e d u r e described by M o n t a g n a
500 ~t, and >500 Ixm fractions by dry sieving.
(1982).
Grains 32-63 ~zm in d i a m e t e r (coarse silt) were
separated f r o m the <63 ~ m fraction by b e a k e r de-
cantation (Wills 1988). (Grains less than 32 txm in Particle Size Measurements and Rates of Tube Building
d i a m e t e r were omitted f r o m the e x p e r i m e n t a l sed- At the e n d o f the e x p e r i m e n t , the contents o f
i m e n t for logistical reasons, cited below.) T h e sep- each dish were fixed with b u f f e r e d 0.2-ktm-filtered
arated size fractions were r e m i x e d in lots o f 600 g f o r m a l d e h y d e in seawater (final concentration,
to create a particle-size class distribution approxi- 5%). Evaporation caused some dishes to lose water,
mating that f o u n d in the field (G. Ditsworth, Unit- such that the sand was n o l o n g e r completely sub-
ed States E n v i r o n m e n t a l Protection Agency, New- merged. These dishes were discarded, r e d u c i n g
port, O r e g o n , u n p u b l i s h e d data). the sample size per s e d i m e n t - t e m p e r a t u r e treat-
T h e 600 g lots o f reconstituted sand were ster- m e n t f r o m 40 at the start o f the e x p e r i m e n t to 21-
ilized by heating at 500~ for 4 h, a p r o c e d u r e
36 at its conclusion. After fixation, samples were
which did n o t lead to any substrate toxicity as de-
rinsed o n a 300-l~m screen and e x a m i n e d u n d e r a
t e r m i n e d by a larval mussel d e v e l o p m e n t assay
dissection microscope. Tanaids picked f r o m the
(Krasnow 1992). Sediment-associated microalgae
samples were stored in 95% ethanol. Total cara-
and bacteria were cultured on sterilized sand to
pace length was m e a s u r e d to the nearest 0.01 m m
create the two nonsterile s e d i m e n t treatments. Mo-
using a Zeiss Videoplan II Image Analysis System.
bile, epipelic diatoms a n d their associated micro-
bial flora were isolated f r o m field sediments by a T u b e material was ashed at 500~ for 4 h and
modification o f the m e t h o d o f de J o n g e (1980). A weighed to the nearest 0.1 mg. After weighing, the
0.5-cm d e e p layer o f s e d i m e n t was placed in a mineral particles f r o m each tube were spread into
clean glass pan. T h e s e d i m e n t was covered with a casting resin o n a glass microscope slide. Diame-
2-mm layer o f sterile f o u n d r y sand. Two layers o f ters o f the first 100 particles e n c o u n t e r e d along a
lens tissue were gently pressed into the top o f the r a n d o m l y chosen transect were m e a s u r e d to the
sand, t h e n the c o n t a i n e r was covered with trans- nearest 1 I~m with the Zeiss image analyzer, con-
p a r e n t plastic wrap a n d placed u n d e r a rack o f flu- n e c t e d by video c a m e r a to a c o m p o u n d micro-
orescent bulbs (110 ~zE m -2 S-1, constant illumi- scope. Grains varied in shape f r o m spherical to
nation) at 20~ After 24 h, the lens tissue a n d ad- acicular a n d for many, there was n o obvious, single
h e r e n t material were gently agitated in a b e a k e r o f d i a m e t e r to be measured. T h e r e f o r e , the "project-
sterile f / 2 0 algal growth m e d i u m (Guillard 1975). e d d i a m e t e r " o f each grain was estimated as the
T h e suspension was p o u r e d t h r o u g h 37-1~m mesh d i a m e t e r o f a circle with the same area as that o f
Nitex | fabric to remove the lens tissue fibers, sand, the particle viewed normally to a plane surface o n
the macrofauna, and most o f the meiofauna. T h e which the particle is at rest in a stable position (Al-
r e m a i n i n g suspension (diatoms, bacteria, small len 1975).
 Tube Building by a Tanald 537

Rate of Tube Breakdown tubes were separated and ashed for 4 h at 500~
A separate laboratory experiment was conducted Total carapace lengths of tanaids and the projected
to measure the rate at which tubes break down to diameters of sediment particles were measured as
constituent sediment particles under quiescent before.
conditions. A tube fragment at least 2 cm long was STATISTICALANALYSES
placed on each of one h u n d r e d twenty, 55-mm dia.
circles cut from 1.0-mm Nitex mesh. Each circle Statistical Software
had been previously rinsed, dried, weighed to the Data were analyzed using Statistix (Analytical
nearest 0.01 mg (tare weight) and set in the bot- Software, Tallahassee, Florida).
tom of a small glass petri dish. Tanaids were re-
moved from the tube fragments and after 5 min Particle Size Selection
air-drying, wet weights of tubes were measured. For the field experiment, selection for coarse silt
The petri dishes with the tube fragments were sub- (32-62 pan) and fine sand (125-249 Ixm) was ex-
merged in a laboratory water table flushed contin- amined by plotting the arcsine-transformed pro-
ually with unfiltered seawater (12~ which intro- portion of each size class in tubes against tanaid
duced both microbes and meiofauna. Illumination length. We did not use this method for data from
was provided by a 75-W General Electric "Plant the laboratory experiment because we discovered,
Gro-and-Sho" fluorescent bulb with photoperiod after the fact, that the proportions of silt and sand
12L:12D. At each of three time intervals, 7d, 14d, in the different sediment treatments were not con-
and 28d, 30 randomly selected dishes were re- sistent due to a defect in a sieve (see Results).
moved from the water table. Tube fragments that Therefore, particle selection for tubes constructed
remained on the Nitex mesh were ashed for 4 h in the laboratory experiments was examined using
at 500~ Percent loss of material from a tube at Vanderploeg and Scavia's Relativized Index of
each sampling interval (t = x) was calculated as Electivity, E* (Lechowicz 1982). E* measures an in-
dividual animal's likelihood of choosing a particle
as a function of both its abundance and the abun-
percent lossvx_ = (1 ~jAFDWt-x~]00
dances of the other particles present. E* ranges
between +1 (maximum preference) and - 1 (total
Thirty tubes were set aside at t = 0 for the deter- avoidance). A value of 0 indicates random selec-
mination of the ratio of wet weight to ash-free dry tion for a particle size class. Electivity indices were
weight. This relationship was used to estimate the plotted against tanaid length. Because E* is not
initial ash-free dry weight (t = 0) of the tubes. amenable to parametric regression analysis (Le-
chowicz 1982), smoothed scatterplots of E* against
Particle Size Selection and Rate of Tube Building in the tanaid length were created using LOWESS, a ro-
Field bust, locally weighted regression algorithm (Cleve-
A method was devised for determining rates of land 1979).
tube building by tanaids in the field, to allow com-
parison with the laboratory results. Approximately Rate of Tube Building
300 tanaids were placed on sterile sand inside plas- For each cell in the 3 • 3 factorial experiment,
tic rings (1 cm high • 2.6 cm wide), giving a den- a least-squares linear regression equation which
sity of 550,000 animals m -2. Tanaids built tubes best fit the relationship between the square-root
overnight in this s e d i m e n t in the l a b o r a t o r y transformed ash-free dry weights of tubes and an-
(10~ The light-colored foundry sand incorpo- imal lengths was calculated. Analysis of residuals
rated into the tube served as a marker for its ensured that the data conformed to the assump-
length at the time of its transplantation to the field, tions of parametric testing. Next, a multiple linear
t = 0. The following day at low tide, the rings and regression with indicator variables to identify tem-
their contents were transferred to the study site. perature and sediment treatments was used to
The contents of each ring were gently shaken into compare the regression functions for the separate
a 1-cm-deep hole in the sediment. A thin layer of cells. Preliminary experiments showed that rate of
natural sediment was transported over the foundry tube building was a linear function of tanaid
sand plugs during the succeeding flood tide. After length. Thus, length was used as a covariate. The
24 h, 8-cm diameter cores, centered on each plug, three-way interaction was significant so the main
were taken and fixed with 10% buffered formalin treatment effects could not be tested. However, be-
in seawater. Fixed samples were gently washed on cause there were no replicate measurements of the
a 300-wm sieve. Tanaids were removed from 50 rate of tube building for a tanaid of a given size in
tubes with segments of both foundry sand and a given temperature-sediment treatment cell, the
field sediment. The field sediment portions of error term in the multiple regression model may
538 L.D. Krasnow and G. L. Taghon

have been inappropriate. As an alternative, analy- Results

ses of variance for the slopes and the intercepts of VISUAL OBSERVATIONS OF TUBE-BUILDING
the nine regression equations were performed. In BEHAVIOR AND TUBE CONSTRUCTION
these tests, the n = 21 to 36 observations in each
t e m p e r a t u r e - s e d i m e n t t r e a t m e n t cell provided As described by Mendoza (1982), L. dubia bur-
rows head-first, excavating sediment with its second
replication.
antennae and gnathopods. Gnathopods grasp and
move particles or, when closed, scrape material to-
Modeling Rates of Tube Building by the ward the body. The tanaid glues particles together
Field Population with mucous threads extruded from pores in the
dactyles of the first pair of pereopods. Scanning
In the field, the rate at which tanaids bind sed-
electron micrographs of tubes built in sterile foun-
iment into tubes will depend on the total abun- dry sand show that the threads capture sediment
dance and size distribution of the animals, among particles as though in an adhesive net (Fig. 1).
other factors. Using the data from the laboratory Once wrapped in mucus, material passes down the
experiments on rate of tube building by individuals rows of pereopods until it collects in a hook
and data on the population dynamics of L. dubia formed by a ventral folding of the posterior end
at the study site (Krasnow 1992), a model was de- of the body. The tube is built from the posterior
veloped to calculate the bulk amount of sediment end of the animal forward. The tanaid rotates dor-
bound into tubes by the field population. The soventrally, completing the circumference of the
model was written in SAS (Statistical Institute, Inc., tube.
Carey, North Carolina), version 6.03. Four param- In natural sediments, tube building and feeding
eters were input to the model: abundance of the activities appear to be closely related. Particles are
field population (inds m-Z), obtained from sam- grasped initially by the gnathopods, passed to the
pling the study site every 2 wks from January 1986 maxillipeds, sorted and then moved either to the
through December 1987 (Krasnow 1992); size-fre- mouth (food) or to the pereopods (tube material).
quency distribution, determined as the percentage Tube building appears to be a continuous process,
of tanaids in each of nineteen 0.2-mm size-class in- permitting the animal to extend its feeding range
tervals for a minimum of three replicate cores per while staying anchored to the substratum. Organic-
sampling date; dates of field sampling, used to mineral aggregates (small inorganic particles, de-
match replicate estimates of population density to tritus, bacteria, and microalgae), abundant in sur-
estimates of the proportion of the population in face sediments, are woven into the structure of the
each size-class interval on a given date; and the tube.
regression equation for the rate of tube building
LABORATORY EXPERIMENT
for an individual tanaid as a function of size. Using
these inputs, the model calculated the rate of tube After the lots of sediment were prepared, a crack
building for an individual juvenile or female tanaid was discovered at the edge of the 125-p~m screen
of a given size, using the empirically derived equa- which was responsible for the actual size distribu-
tion determined from the laboratory experiments tions of the sediments differing between treat-
for the rate of tube building, then calculated the ments and from the sediment at the study site (Ta-
rate of tube building by all individuals in the field ble 1). Because of the time and effort involved in
population for each sampling date. Because the setting up and sampling the experiment, we elect-
rate of tube building is a computed quantity and ed not to repeat it in its entirety. A second, smaller
experiment (discussed below) was conducted to
the uncertainty associated with its estimate is a
verify trends found in the main experiment.
composite of the uncertainties associated with the
The two sediment cultures differed in that the
component variables (i.e., tanaid length, rate of
17~ culture (culture I sediment) had a signifi-
tube building by individual animals, population cantly higher concentration of chlorophyll a but a
size and the proportion of the population in a giv- significantly lower bacterial abundance than the
en size class), the variance associated with each es- 10~ culture (culture II sediment) (6.1 mg chlo-
timate of rate of tube building was calculated by rophyll g-X sediment versus 2.7, p < 0.001 and 3.7
propagation of error (Gore 1952). Output from • 107 bacteria g-1 sediment versus 4.9 • 107, p <
the model was graphed as the bulk rate of tube 0.01, respectively). Culture I sediment was domi-
building (g m -2 d -1) by the tanaid population on nated by small pennate diatoms (Stauroneis constric-
each biweekly sampling date. The error associated ta and Navicula complanatula), although cyanobac-
with each estimate was plotted as one standard de- teria (Phoridium spp. and Spirulina sp.) were also
viation from the mean. abundant. The surface of culture I sediment was
 Tube Building by a Tanaid ~39

,'~ , , , ,

",5' "" , , " , , ' , ' ', ," " " ,, '" " " " " ' ~ ~,,~!)~'

Fig. 1. Scanning electron micrograph of clean sand grains b o u n d in a web of m u c u s by Leptocheha dubia. Scale bar represents 100 p~m.

golden brown. In contrast, the surface of culture (250-500 p~m) composed an average of less than
II sediment was bright green and the flora were 3% of the particles in tubes. Tanaids did not use
more evenly divided between cyanobacteria and grains larger than 500 p~m.
the diatoms Stauroneis, Navicula, and Nitzschia soci- When given sterile sediment to construct tubes,
alis. More importantly, the two cultures produced small tanaids showed positive selection for silt-size
morphologically different types of mucus. Culture particles at both 5~ and 20~ (Fig. 2). Small tan-
I, dominated by diatoms, produced mucus in aids either avoided using sterile fine sand (at 5~
sheets and "wads" embedded with sediment. Cul- or showed random selection (at 20~ Conversely,
ture II produced mucus in strings. It was some- the largest tanaids avoided using sterile silt but
times difficult to distinguish sediment particles showed positive selection for sterile fine sand (Fig.
bound by microbes in culture II during the prein- 2). The presence of a microbial community in the
cubation period from material bound by tanaids sediment resulted in some shifts in these size-selec-
during the experiment. For this reason, rates of tion patterns. Small tanaids showed less positive se-
tube building may have been overestimated for lection (at 5~ or slightly negative selection (at
tanaids in culture II sediment. Particle-size selec- 20~ for silt-size particles in culture I sediment.
tion was not analyzed for tanaids that built tubes Fine sand with a microbial coating was not as
in culture II sediment. strongly avoided. Large tanaids also altered some-
Particle-size selection was examined in four of what their selectivity patterns on culture I sedi-
the nine treatment cells: 5~ sterile sediment; 5~ ment, showing less avoidance of silt. Selection of
culture I sediment; 20~ sterile sediment; and fine sand was not notably different.
20~ culture I sediment. The bulk of the particles
incorporated into tubes were in the size class "very C O R R O B O R A T I O N OF PARTICLE SELECTION
fine sand" (63-125 ~m) with different patterns in EXPERIMENT
the use of silt (32-63 ~m) and fine sand (125-250 Although E* corrects mathematically for the dif-
p~m) by small and large tanaids. Medium sand ferences in particle availability inadvertently

TABLE 1. Particle-size c o m p o s i t i o n ( m e a n % d r y weight _+ s t a n d a r d error) o f s e d i m e n t s u s e d in laboratory a n d field e x p e r i m e n t s .

Type <63 ~m 63-125 ~m 125-250 ~m 250-500 ~m

Laboratory, sterile 10.6 -+ 0.8 19.3 _+ 0.6 61.2 _+ 0.4 5.2 + 0.5
Laboratory, culture I 21.6 -+ 0.7 34.5 -+ 0.3 40.0 -+ 0.7 1.3 + 0.3
Field 38.2 -4- 0.3 28.8 -+ 0.1 28.8 + 0.4 2.5 -+ 0.4
540 L . D . Krasnow and G L Taghon

Silt Fine sand

0.4
04--

0.0
5oc 0.0
-0.4 sterile
9 coo i
-0.8 -04

t I I t I 9 I I F I I I I

0.4
0,4 --

9 9 9 ~e
0.0 20~
0.0
sterile
x -0.4
' (1)
0 8 9
"" -0.8 -0.4

i i i i i ~ I I I I I I

0.4
__.m 0.4--
LU ee

0.0 5~
culture I 0.0
-0.4 e e

-0.8 -0.4

e l l l J I I

0.4
0,4 --

0.0 20~ 0.0

culture I
-0.4

-0 8 -0 4

I I I I I t I I I I I I

5 10 1.5 2 0 2 5 3 0 3 5 4.0 0 5 1.0 1.5 2.0 2.5 3.0 35~4

Carapace length, mm
Fig. 2. Electivity index for coarse silt-size and fine sand-size particles used for tube building by Laptocheha dubia in the laboratory
experiment. Solid line represents LOWESS-smoothed fit using an f v a l u e o f 0.7. Dashed horizontal line indicates r a n d o m selectivity.

caused by the defective sieve, results may still be out o f nine large tanaids in sterile sediment, the
biased by differences in a tanaid's behavioral re- p r o p o r t i o n o f silt in tubes was 4% or less. In pre-
sponse. A s e c o n d e x p e r i m e n t was c o n d u c t e d to cultured sediment, the a m o u n t o f silt in tubes o f
verify trends in grain-size selection. Twelve tanaids, large tanaids varied f r o m 5% to 23%. Results for
o n e per dish, were selected for each o f four sedi- these two s e d i m e n t treatments were significantly
m e n t t r e a t m e n t - t a n a i d size class c o m b i n a t i o n s different (Rank Sum Two-Sample Test, p = 0.02,
(sterile sediment, small tanaids; sterile sediment, n = 9). Thus, the p r e s e n c e o f microbes a p p e a r e d
large tanaids; p r e c u l t u r e d sediment, small tanaids; to increase the rate at which large tanaids incor-
p r e c u l t u r e d sediment, large tanaids). In this case, p o r a t e d silt into tubes. Silt c o m p o s e d an average
care was taken that the p r o p o r t i o n s o f silt and sand o f 45% (range 34-56%) o f the grains in the tubes
were the same in the sterile a n d p r e c u l t u r e d sedi- o f small tanaids in sterile s e d i m e n t and 22% -+ 8%
m e n t mixtures. "Small" tanaids r a n g e d f r o m 0.8 in tubes built in s e d i m e n t with microbes. We did
m m to 1.0 m m carapace length. " L a r g e " tanaids n o t c o m p a r e these m e a n s statistically because only
were 3.4 4.2 mm. Tanaids built tubes for 24 h at two small tanaids built tubes in sterile sediment.
18~ Fine sand c o m p o s e d 12% -+ 1% o f the particles
T h e r e was n o significant difference in the per- available in the second e x p e r i m e n t , with n o signif-
cent composition o f silt-size particles ( m e a n = 18.2 icant difference in the availability o f fine sand be-
-+ 0.2% SE) between the sterile and p r e c u l t u r e d tween the sterile a n d p r e c u l t u r e d s e d i m e n t mixes
s e d i m e n t mixes (Rank Sum Two-Sample Test, p = (Rank Sum Two-Sample Test, p = 0.25, n = 9).
0.44, n = 4 slides, 100 grains p e r slide). For seven T h e overall m e a n p e r c e n t fine sand in tubes built
 Tube Building by a Tanaid 54.1

5~ sediment 10~ sediment 20~ sediment

02
Y=OO22X - 0.015 (p < 0 001) Y = 0 . 0 0 6 3 X + 0,028 Y = O 0 3 4 X -0034(p<0.001)
r2 = 0.46 r 2= O 047 r2= 0.70

0.1
,5

"O
E o.o
E
-" 0- 0.2
5~ I sediment 10~ I sediment 20~ I sediment
Y = O 0 2 2 X - 0 . 0 0 0 8 9 (p < 0,001} Y = 0.022 X + 0.0045 (p < 0.001) Y = 0 0 1 8 X + 0 , 0 0 7 6 (p < 0 0 0 1 )
r2= 0 73 r2= 0.29 r2= 0 26

o II
o; ol e%
._

:3
..Q

...Q oo I I I I I I -I~ J I I I k I i i i i i i i

5~ II s e d i m e n t 10~ II s e d i m e n t 20~ II s e d i m e n t
02
o
Y = 0 , 0 2 9 X - 0 . 0 0 6 2 (p < 0 0 0 1 } Y = 0,026 X + O 0041 (p < O 001}
r2= 0 56 9 r2= 0 83
rr 9 oi,9
% 9
9 9 e8
01 Q9
o0 9 9
9 8
Y = 0 0031 X + 0 095
r2= 0 013 #
00 [ I I ] I I I I I J I [ I F I I ~ I I I I
5 1.5 2.5 3.5 4.5 0.5 1.5 25 3.5 4 5 0.5 1.5 2.5 3.5 4,5
Carapace length, mm
Fig. 3. Rates of tube building as linear functions of tanaid length in the laboratory experiment. Rate of tube building is expressed
in units of grams ash-free dry weight sediment d -l. A square-root transformation of rates of tube building was used to satisfy the
assumptions of parametric least squares regression testing. Significance levels are for slope of regression.

by large tanaids was 9% -+ 2%, with no significant lection by small tanaids was not analyzed statisti-
difference between sediment treatments. Thus, cally because only two small tanaids built tubes in
large tanaids used fine sand at a rate approximately sterile sediment.
equal to its availability. The mean percent fine sand
in the tubes of small tanaids in sterile sediment was RATE OF TUBE BUILDING
5% (range 0-10%), compared to 17% + 11% in
precultured sediment. As before, particle size se- The square root-transformed rate of tube build-
ing was a positive function of tanaid length in sev-
en out of nine cells in the experiment (Fig. 3).
TABLE 2. Analysis of variance comparison of (A) the slopes
and (B) the intercepts (at X = 2.8 mm carapace length) of the Across all seven temperature and sediment treat-
regression equations derived from the laboratory rate of tube ments, larger tanaids bound more sediment into
building experiment. tubes over 24 h than small tanaids. In the experi-
A. Slopes mental cells 5~ culture II sediment and 10~
Source df SS MS F P sterile sediment, no statistically significant relation-
Main effects ship between tanaid size and rate of tube building
Temperature 2 1.718E-04 8.588E-05 0.54 0.62 could be detected. In several of the treatment cells
Sediment 2 3.509E-06 1.754E-06 0.01 0.99 where a significant linear relationship did occur
Treatment (e.g., 10~ culture I sediment and 20~ culture I
Residual 4 6.351E-04 1.588E-04
Total 8 8.103E-04 sediment), scatter around the line indicated a high
B Intercepts
degree of individual variation.
Source df SS MS F P Analyses of Variance were performed for the
Main effects slopes and intercepts of the nine regression equa-
Temperature 2 7.268E-04 3.634E-05 0.22 0.81 tions. In these analyses, the 21 to 36 observations
Sediment 2 1.950E-03 9.75lE-04 5.88 0.06 of each temperature-sediment treatment cell pro-
Treatment vided replication. There was no significant effect
Residual 4 6.670E-04 1.668E-04 of temperature or sediment treatment on the
Total 8 2.690E-03
slopes of the nine regression equations (Table 2).
542 L D. Krasnow and G, L. Taghon

020
T h e overall m e a n slope was 0.020 g m m -1 d -1 (SE
= 0.003). u2.
o
Before c o n d u c t i n g the ANOVA for the nine in- "7" o Q

tercepts, the estimated rate o f tube building was Y=O030X+O0068(p<0001) 9

adjusted for the m e d i a n size o f tanaid used in the o15 r2=025
E
e x p e r i m e n t (2.8 m m carapace length). Analysis o f -cl
Q2
variance o f the intercepts at X = 2.8 m m detected 03

n o effect o f t e m p e r a t u r e , although the effect o f

s e d i m e n t t r e a t m e n t o n the intercepts a p p r o a c h e d .~_ 010 -- 9

significance (Table 2). L i n e a r contrasts ( S n e d e c o r

and C o c h r a n 1980) d e m o n s t r a t e d that the average
c33
effect o f microbes in s e d i m e n t o n the intercepts ..Q

was significant (p < 0.005). In addition, the effect 005 --

o f culture II was greater than that o f culture I (p ~ "..

< 0.005).
n-
RATE OF BREAKDOWN OF TUBES
000 I I I I I 1 l
T h e relationship between the wet weight and 5 10 15 20 25 30 35 40 45
ash-free dry weight o f tubes at t = 0 was estimated Carapace length, mm
by the linear regression function:
Fig. 4. Rates of tube building as linear functions of tanaid
AFDW = 0.00262 + 0.515 (wet weight)' length in the field experiment. Rate of tube building is ex-
pressed in units of grains ash-free dry weight sediment d -1. A
(p < 0.001, r 2 = 0.98). Within 7 d, tubes lost an square-root transformation of rates of tube building was used
average of 71.7% (+ 5.4%, 95% c o n f i d e n c e inter- to satisfy the assumptions of parametric least squares regression
val) o f their estimated initial ash-free dry weight. testing. Significance levels are for slope of regression.
After 14 d, tubes a p p e a r e d to be intact b u t disin-
tegrated w h e n rinsed o n the sieve. Much o f the
organic matrix that h e l d the mineral grains togeth- and rate of tube building, the slope o f the new
er h a d d e c o m p o s e d . After 28 d, the tube fragments m o d e l was estimated f r o m the g r a n d m e a n o f the
r e m a i n i n g in the dishes could n o t be distinguished slopes o f the regression equations (Fig. 3). We did
f r o m aggregates that had c o m e in t h r o u g h the sea- observe a significant effect o f microbial culture o n
water line. the intercepts o f the regression equations, howev-
er. T h e r e f o r e , we estimated the i n t e r c e p t o f the
PARTICLE SIZE SELECTION AND RATE OF new m o d e l by the g r a n d m e a n o f the intercepts (at
TUBE BUILDING IN THE FIELD X = 0) for the six regression functions in the cells
N e i t h e r the p r o p o r t i o n o f silt or fine sand in containing microbial s e d i m e n t cultures:
tubes built d u r i n g the August 1989 field experi-
Y = 0.017 + 0.020X
m e n t was a function o f tanaid length, n o r did the
p r o p o r t i o n s in tubes differ notably f r o m the pro- where Y = square r o o t o f rate o f tube building (g
portions in natural sediment. This result is expect- AFDW d -a) a n d X = tanaid carapace length (ram).
ed if, as observed in the laboratory e x p e r i m e n t , In five o f these six cases, the i n t e r c e p t at X = 0
electivity b e c o m e s m o r e r a n d o m when a s e d i m e n t was not significantly different f r o m zero. Thus, the
microbial c o m m u n i t y is present. i n t e r c e p t was omitted f r o m the model.
T h e r e was a positive linear relationship between T h e estimated gross rate o f tube building aver-
rate o f tube building in the field and tanaid length aged approximately 350 g m -2 d -1 between J a n u a r y
(Fig. 4). Given the d e g r e e o f individual variation 1986 and D e c e m b e r 1987 (Fig. 5). In general, rates
exhibited, rates o f tube building in the laboratory o f tube building were h i g h e r after the fall repro-
(Fig. 3) a p p e a r to reasonably a p p r o x i m a t e those ductive burst than d u r i n g spring, when smaller
observed in the field. n u m b e r s o f juveniles r e c r u i t e d into the p o p u l a t i o n
(juveniles recruit into the p o p u l a t i o n during b o t h
MODELING RATES OF TUBE BUILDING BY THE spring and tall (Krasnow 1992). T h e estimated
FIELD POPULATION standard deviation, d e t e r m i n e d by p r o p a g a t i o n o f
We used a new regression function, based o n the error, was approximately equal to the mean, and
laboratory e x p e r i m e n t a l results, to m o d e l rates o f was most sensitive to tanaid length. A doubling o f
tube building in the field. Because there were n o tanaid length, f r o m 2 m m to 4 ram, increased the
significant effects o f t e m p e r a t u r e or s e d i m e n t e r r o r associated with an estimate o f the gross rate
t r e a t m e n t o n the relationship between animal size o f tube building by 12 times.
 Tube Building by a Tanald 543

1600 - -
wide enough that silt-size grains slip through easily.
When microbes are present, animals manipulate
1200
the organic film in which small particles are em-
bedded. Small tanaids showed an increased ten-
dency to use fine sand-size particles when these

I I:i tIttI
800 were cultured. Presumably, small tanaids can ma-
nipulate fine sands whether they are discrete par-
ticles or part of an organic-mineral aggregate. Se-
"1~ 400 lection of sands with microbes may reflect their
increased 'value' to small tanaids as food particles
~2b
or tube material.
Leptochelia dubia does appear to actively sort par-
Jan Feb Mar Apr May Jun Jul Aug $ep Oct Nov Dec Jan
ticles. Particles grasped by the gnathopods are
1986
passed to the maxillipeds, sorted, and then moved
1600
to the mouth (food) or to the pereopods (tube
material). Miller (1984) provides a detailed de-
1200
scription of post-capture particle sorting for am-
o~ phipods of the genus Corophium. The first gnatho-
or"
pods transfer particles to the mouth where they are
800 sorted by rapid motions of the maxillipeds and first
and second maxillae and slower movements of the

00 IIIIiIi, IIiII mandibles. Rejected particles are passed forward

and brushed into the pleopodal current or are
swept posteriorly between the second gnathopods.
Miller (1984) notes that post-capture sorting allows
0 r J r r r t ~ F I I i contact between sediment particles and potential
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan
chemosensory organs, permitting active selection
1987
on the basis of characteristics such as adsorbed sub-
Fig. 5. Estimated mean (+ SD) rates of tube building by the stances and surface-attached microflora.
Leptocheha dubia field population at the study site in Yaquina Bay, These explanations are a posteriori and require
1986--1987. Estimates were derived using the model described
in the text. further experimentation to test them explicitly.
Whatever the mechanism, the p h e n o m e n o n of
particle selection is relevant for several reasons.
Discussion
Many organic and inorganic compounds prefer-
endally associate with smaller particles. If such
PARTICLE SIZE SELECTION DURING TUBE BUILDING compounds are pollutants (e.g., Boehm and Far-
Experimental studies of particle selection by rington 1984), then incorporation into tubes of
benthic infauna, whether particles are ingested or benthic animals may be one mechanism for mag-
used in tube or burrow construction, often use ex- nifying and maintaining their concentrations in
otic particles (e.g., Taghon 1982; Self and Jumars sediment. Alternatively, since tubes and burrows
1988; Carey 1989). Since exotic particles often lack are often localized sites of greatly enhanced micro-
a natural microbial community, concerns about ar- bial activity, the degradation of these compounds
tifacts must always temper conclusions from such may be facilitated. Furthermore, patterns of size
experiments. The present data suggest that the selectivity during tube construction will influence
presence or absence of microbes influenced par- transport rates of water into sediments, since the
ticle selectivity by L. dubia during tube building. sizes of particles incorporated into tubes will affect
Although the mechanisms of selection were not the permeability of the tube wall. Irrigation of sed-
unambiguously identified, there appear to be me- iments by tube-dwelling fauna is an important pro-
chanical and behavioral c o m p o n e n t s involved. cess affecting chemical reactions in sediments (All-
There may be behavioral or chemosensory com- er 1983). Overlying all of these processes is the
ponents involved in particle selection as well. temporal variation in sizes of particles in tubes, as
Large tanaids showed an increased tendency to use cohorts recruit and animals grow (Fig. 2).
silt-size particles in their tubes when these particles
had been cultured with bacteria and benthic mi- RATES OF TUBE BUILDING
croalgae. Direct observations suggest that it is dif- Because we wanted to model the total mass of
ficult for large tanaids to collect small particles. sediment bound by the field population, we mea-
The central gap in the chela of a large tanaid is sured tube production in units of mass, rather than
544 L.D. Krasnow and G. L. Taghon

length. O t h e r authors have r e p o r t e d rates of tube was a surprising result. T h e local p o p u l a t i o n o f L.

building by tanaids in units o f length. Rates re- dubia may be acclimated to a wide range o f tem-
p o r t e d by Richards (1969:1.3-4.0 cm d -1) and My- peratures. O n a cloudless day, tanaids may experi-
ers ( 1 9 7 7 : 3 . 0 cm d -1) are c o m p a r a b l e to those ence a wide fluctuation in t e m p e r a t u r e between
observed in this e x p e r i m e n t (personal observa- high a n d low tide conditions. For example, d u r i n g
tions). Mendoza's (1982) estimate (15.8 cm d - l ) , low tide o n May 9, 1989, the t e m p e r a t u r e at the
extrapolated f r o m results o b t a i n e d d u r i n g the first study site averaged 20.1~ at the sediment surface
h o u r o f an e x p e r i m e n t , is m u c h higher. This dis- a n d 19.7~ at a d e p t h o f 2 cm below the s e d i m e n t
parity may indicate that the short-term, initial rate surface. T h e t e m p e r a t u r e o f the water in the bay,
o f tube building is disproportionately high. which would immerse the study site o n the ensuing
Although there are n o c o m p a r a b l e studies o n high tide, was 12.7~
rates o f tube building by o t h e r chelate macro- O u r m o d e l o f rate o f tube building in the field
fauna, a n u m b e r o f field and laboratory studies has two major limitations. First, it does n o t include
have m e a s u r e d sediment-processing rates by de- any d e n s i t y - d e p e n d e n t effects. T u b e building rates
posit-feeding polychaetes and holothurians (e.g., o f individual animals may well be affected by the
H o b s o n 1967; Hauksson 1979; Rice et al. 1986). In a b u n d a n c e o f o t h e r tanaids or o t h e r species o f in-
these cases, s e d i m e n t reworking represents the fauna. At least over a 24-h period, however, rates
rate at which animals ingest and defecate sediment o f tube building were quite similar at an abun-
and thus differs intrinsically f r o m tube building by dance o f o n e animal p e r 28 cm 2 in the lab exper-
L. dubia. S e d i m e n t reworking d u r i n g deposit feed- i m e n t (Fig. 3), c o r r e s p o n d i n g to an a b u n d a n c e o f
ing may destabilize sediments by increasing poros- 350 m -2, and an a b u n d a n c e o f 550,000 m -2 in the
ity. Flume e x p e r i m e n t s show that tube and b u r r o w field e x p e r i m e n t (Fig. 4). Second, in its present
construction by tanaids and a m p h i p o d s decrease f o r m the m o d e l c a n n o t be used to predict the mass
b e d l o a d transport rates (Krasnow and T a g h o n o f s e d i m e n t b o u n d into tubes at any given time. In
1992; G r a n t a n d D a b o r n 1994). the field, the a m o u n t o f s e d i m e n t in tubes reflects
Rates o f tube building increased with tanaid size. a balance between rate o f tube construction and
T h e e r r o r o f a derived estimate o f the bulk rate o f rate o f breakdown. T h e laboratory e x p e r i m e n t o n
tube building by the field p o p u l a t i o n was also high- rate o f b r e a k d o w n indicated that tubes lost an av-
ly sensitive to tanaid size. Thus, the size-class dis- erage o f 72% o f their initial mass within 7 d. O n
tribution o f the p o p u l a t i o n was an i m p o r t a n t com- average, the field p o p u l a t i o n p r o d u c e d 2,450 g
p o n e n t o f the model. Typically, estimates o f sedi- m -2 o f tubes per week (i.e., 350 g m -2 d -1 • 7 d).
m e n t reworking by animals in the field are based If 72% o f this p r o d u c t i o n was lost over the same
on simply multiplying individual rates by observed period, the tanaids created a n e t p r o d u c t i o n o f 686
a b u n d a n c e s in the field. O u r results indicate that g AFDW m -2 wk -1. However, the rate o f breakdown
the errors i n h e r e n t in such an a p p r o a c h can be o f tubes in the field is unknown. Physical stresses
considerable (Fig. 5) a n d deserve f u r t h e r atten- f r o m waves a n d currents can contribute to break-
tion. We also a t t e m p t e d to i n c o r p o r a t e the effects down o f tubes. Bacterial activity, presumably re-
o f t e m p e r a t u r e a n d sediment-associated microbes sponsible for the degradation o f m u c u s used to
into the model. Due to the significant interaction b i n d sediment particles together, can also be stim-
between t e m p e r a t u r e a n d s e d i m e n t t r e a t m e n t in ulated u n d e r field conditions (e.g., flowing water).
the e x p e r i m e n t , we were unable to detect a main Alternatively, tubes a n d burrows o f some benthic
effect due to t e m p e r a t u r e or s e d i m e n t t r e a t m e n t infauna contain c o m p o u n d s that a p p e a r to de-
o n rates o f tube building. T h e r e is n o clear expla- crease the susceptibility o f the m u c u s to microbial
nation for the interaction between t e m p e r a t u r e attack (King 1986). F u r t h e r m o r e , tanaids may con-
and s e d i m e n t treatment. T h e interaction may have tinually "recycle" tubes. Tanaids were observed to
b e e n an artifact, caused by differences in the phys- grab bits o f the walls o f adjacent tubes and incor-
iological condition o f animals between e x p e r i m e n - p o r a t e t h e m into their own. In this way, some tube
tal cells. Although we were careful to use only ju- material appears to be r e i n f o r c e d or p r o t e c t e d
veniles a n d female tanaids without oostegites or f r o m breakdown. In the field, it appears that m u c h
broods, there may have b e e n o t h e r physiological o f the u p p e r 2 cm o f s e d i m e n t is i n c o r p o r a t e d into
or behavioral attributes o f individual condition tubes at any o n e time (Fig. 6).
that biased activity. T h e significance o f tube building by L. dubia lies,
T h e ANOVA comparisons o f the slopes and in- in large part, in the ubiquitous, worldwide distri-
tercepts o f the regression equations derived f r o m b u t i o n o f this species a n d the densities it achieves
the laboratory e x p e r i m e n t o f rate o f tube building in the field. Each individual binds only 1-20 m g
(Table 2) indicate that the effect o f t e m p e r a t u r e s e d i m e n t d -1, b u t d e n s i t i e s typically r e a c h e d
o n rates o f tube building was n o t significant. This 100,000 individuals m -2 at the study site in Yaquina
 Tube Building by a Tanaid 545

9 ,', 9

:?

9 : ,~,.,~,~

,, ~, ,

,, ~:,. .. ,~:, '. ~ 7 .

"7

Fig. 6. Tubes of Leptochelia du&a at the edge of a tidal creek near the study site. The pen points to the free ends of the tubes,
exposed where sediment has been transported by the flow in the creek.

Bay. At the site examined, L. dubia, in concert with ments in the Georges Bank region. Environmental Saence and
the local microbial assemblage, may control the re- Technology 18:840-845.
CARE7, D. A. 1989. Fluorometric detection of tracer particles
suspension of particles through the benthic used to study animal-particle dynamics. Limnology and Ocean-
boundary layer. The bulk of the particles reaching ography 34:630-635,
the bottom appear to be bound into aggregates by CLEVELAND, W. S. 1979. Robust locally weighted regression and
microbial adhesives, woven into tubes and stabi- smoothing scatterplots. Journal of the American Statistical Asso-
lized against erosion (Krasnow and Taghon 1992). aation 74:829-836.
DAWS, M. W. 1981. Production dynamics of sediment-associated
algae in two Oregon estuaries. Ph.D. Thesis, Oregon State
ACKNOWLEDGMENTS University, Corvallis, Oregon.
DEFLAuN, M. E AND L. M. MAYER. 1983. Relationships between
We thank J o h n Chapman for assistance with the computer
bacteria and grain surfaces in intertidal sediments. Limnology
model, Michael Davis for identification of the benthic algae,
Anja Robinson for performance of the larval mussel bioassay, and Oceanography 28:873-881.
and J o h n Chapman and Cynthia Trowbridge for comments on DE JONGE, V. N. 1980. Fluctuations in the organic carbon to
the manuscript. This research was supported by a grant from chlorophyll a ratios for estuarine benthic diatom populations.
the National Science Foundation #OCE 8800642. Manne Ecology Progress Series 2:345-353.
DEJONGE, V. N. 1994. Wind-driven tidal and annual gross trans-
port of m u d and m i c r o p h ~ o b e n t h o s in the Ems Estuary, and
LITERATURE CITED its importance for the ecosystem, p. 29-40. In IC R. Dyer and
ALLEN, T. 1975. Particle Size Measurement. J o h n Wiley and R.J. Orth (eds.), Changes in Fluxes in Estuaries. Olsen and
Sons, New York. Olsen, Fredensborg.
ALLER, R. C. 1983. The importance of the diffusive permeability DYER, tC R. 1989. Sediment processes in estuaries: Future re-
of animal burrow linings in determining marine sediment search requirements. Journal of GeophysicalResearch 94:14327-
chemistry. Journal of Marine Research 41:299-322. 14339.
ALLER, R, C. AND J. Y. YINGST. 1978. Biogeochemistry of tube- ECKMAN,J. E., A. R. M. NOWELL, AND E A. JUMARS. 198l. Sedi-
dwellings: A study of the sedentary polychaete Amphitrite or- ment destabilization by animal tubes. Journal of Manne Re-
nata (Leidy). Journal of Marzne Research 36:201-254. search 39:361-374.
ALONG[, D. M. 1985. Microbes, meiofauna, and bacterial pro- EMERSON, C. W. 1989. Wind stress limitation of benthic sec-
ductivity on tubes constructed by the polychaete Capitella cap- ondary production in shallow, soft-sediment communities,
itata. Mar*ne Ecology Progress Series 23:207-208. Marine Ecology Progress Series 53:65-77.
AMSPOKER, M. C. AND C. D. MCINTIRE. 1978. Distribution of FAGER, E. W. 1964. Marine sediments: Effect of a tube-building
intertidal diatoms associated with sediments in Yaquina Bay, polychaete. Science 143:356-359.
Oregon. Journal of Phycology 14:387-395. FEATHERSTONE, R. E AND M.J. RISK. 19779 Effect of tube-build-
BOEHM, P. D. ANDJ. W. FARRINGTON. 1984. Aspects of the poly- ing polychaetes on intertidal sediments of the Minas Basin,
cyclic aromatic hydrocarbon geochemistry of recent sedi- Bay of Fundy. Journal of Sedimentary Petrology 47:446-450.
546 L.D. Krasnow and G. L. Taghon

GEESEV,G. G. 1982. Microbial exopolymers: Ecological and eco- chelia dubia (Kroyer, 1842) (Tanaidacea). Crustaceana 43:225-
nomic considerations. American Society of Microbiology, News 48: 240.
9-14. MEYERS, M. B., H. FOSSING, AND E. N. POWELL. 1987. Microd-
GORE, W. L. 1952. Statistical Methods for Chemical Experimen- istribution of interstitial meiofauna, oxygen and sulfide gra-
tation. Interscience Publishers, New York. dients, and the tubes of macro-infauna. Marine EcologyProgress
GRANT,J. AND G. DABORN. 1994. The effects ofbioturbation on Series 35:223-241.
sediment transport on an intertidal mudflat. NetherlandsJour- MILLER, A. R., R. L. LOWE, ANDJ. T. ROTTENBERRY. 1987. Suc-
nal of Sea Research 32:63-72. cession of diatom communities on sand grains. Journal of Ecol-
GR.gMARE,A. 1988. Feeding, tube building and particle-size se- ogy 75:693-709.
lection in the terebellid polychaete Eupolymnia nebulosa. Ma- MILLER, D. C. 1984. Mechanical post-capture particle selection
rine Biology 97:243-252. by suspension- and deposit-feeding Corophium. Journal of Ex-
GUILLARD, R. R. L. 1975. Culture ofphytoplankton for feeding perimental Marine Biology and Ecology 82:59-76.
marine invertebrates, p. 29-59. In W. L. Smith and M. H. MILLER, D. C. 1989. Abrasion effects on microbes in sandy sed-
Chanley (eds.), Culture of Marine Invertebrate Animals. Ple- iments. Marine Ecology Progress Series 55:73-82.
num, New York. MILLER, D. C. AND R. W. STERNBERG. 1988. Field measurements
HAUKSSON, E. 1979. Feeding biology of Stichopus tremulus, a de- of the fluid and sediment-dynamic environment of a benthic
posit-feeding holothurian. Sars~a 64:155-160. deposit feeder. Journal of Marine Research 46:771-796.
H1GHSMITH, R. C. 1983. Sex reversal and fighting behavior: Coe- MONTAGNA, P. A. 1982. Sampling design and enumeration sta-
volved phenomena in a tanaid crustacean. Ecology64:719-726. tistics for bacteria extracted from marine sediments. Apphed
HOBSON, K. D. 1967. The feeding and ecology of two North and Environmental Microbiology 43:1366-1372.
Pacific Abarenicola species (Arenicolidae, Polychaeta). Biologz- MYERS, A. C. 1977. Sediment processing in a marine subtidal
cal Bulletin 133:343-354. sandy bottom community: I. Physical aspects. Journal of Marine
HOLDmH, D. M. ANDJ. A. JONES. 1983. Tanaids. Keys and notes Research 35:609-632.
for the identification of species. Synopses of the British fauna, NOWELL, A. R. M., P. A.JUMARS,ANDJ. E. ECKMAN. 1981. Effects
No. 27. Cambridge University Press, Cambridge. of biological activity on the entrainment of marine sediments.
KING, G. M. 1986. Inhibition of microbial activity in marine Marine Geology42:133-153.
sediments by a bromophenol from a hemichordate. Nature PAWLIK,J. R., C. A. BUTMAN,AND V. R. STARCZAK. 1991. Hydro-
323:257-259. dynamic facilitation of gregarious settlement of a reef-build-
KRASNOW, L. D. 1992. Particle selection for tube building and ing tube worm. Science 251:421-424.
the stabilization of sediments by Leptochelia dutna (Crustacea, RHOADS, D. C., J. u YINGST, AND W. J. ULLMAN. 1978. Seafloor
Tanaidacea): Interactions with bacteria and microalgae. Ph.D. stability in central Long Island Sound: Part 1. Temporal
Dissertation, Oregon State University, Corvallis, Oregon. changes in erodibility of fine-grained sediment, p. 221-244.
KRASNOW, C. D. AND G. L TAGHON. 1992. Influence of macro- In M. L. Wiley (ed.), Estuarine Interactions. Academic Press,
and microorganisms on bedload transport in an estuary, p. New York.
260-272. In M. E. Jones and A. Laenen (eds.), Interdisciplin- RICE, D. L., T. S. BIANCHI,AND E. H. ROPER. 1986. Experimental
ary Approaches in Hydrology and Hydrogeology. American studies of sediment reworking and growth of Scoloplos spp.
Institute of Hydrology, Minneapolis, Minnesota. (Orbiniidae: Polychaeta). Martne EcologyProgressSeries30:9-19.
KULM, L. D. 1965. Sediments of Yaquina Bay, Oregon. Ph.D. RICHARDS, L 1969. Tanaidacea (Crustacea: Peracarida) of San
Dissertation, Oregon State University, Corvallis, Oregon. Juan Island. Zoology Report 533b, Friday Harbor l,aborato-
LECHOWmZ, M.J. 1982. The sampling characteristics of electiv- ties, University of Washington, Friday Harbor, Washington.
ity indices. Oecologia52:22-30. SELF, R. F. L AND R A. JUMARS. 1988. Cross-phyletic patterns of
LEWIS, D. B. 1968. Feeding and tube building in the Fabriciinae particle selection by deposit feeders. Journal of Manne Research
(Annelida, Polychaeta). Proeeedzngsof the Linnean Soaety of Lon- 46:119-143.
don 179:37-49. SIEG, J. 1986. Distribution of the Tanaidacea: Synopsis of the
LOP,ENZEN, C.J. 1967. Determination of chlorophyll andphaeo- known data and suggestions on possible distribution patterns,
p. 165-194. In R. H. Gore and K. L. Heck (eds.), Crustacean
pigments: Spectrophotometric equations. Limnology and
Biology. Balkema, Rotterdam.
Oceanography 12:243-246.
SIEG,J. AND R. N. WINN. 1981. The Tanaidae (Crustacea; Tan-
LUCKENBACH, M. W., D. g. HUGGETT, AND E. C. ZOBRIST. 1988.
aidacea) of California, with a key to the world genera. Pro-
Sediment transport, biotic modifications and selection of
ceedings of the Biological Society of Washington 94:315-343.
grain size in a surface deposit-feeder. Estuaries 11:134-139.
SNEDECOR, G. W. AND W. G. COCHRAN. 1980. Statistical Meth-
M~aN, M. B. AND W. G. NELSON. 1988. Sedimentary character-
ods. Iowa State University Press, Ames, Iowa.
istics of Sabellariid worm reefs (Phragrnatopoma lapidosa Kin- TAGHON, G. L. 1982. Optimal foraging by deposit-feeding in-
berg). Estuarine, Coastal and Shelf Saence 26:105-109. vertebrates: Roles of particle size and organic coating. Oecol-
MEADOWS, E S. ANDJ. G. ANDERSON. 1967. Micro-organisms at- ogia 52:295-304.
tached to marine sand grains. Journal of the Marzne Bwlogtcal WARt)ELL,J. N. 1988. Methods for the study of bacterial attach-
Assoaation United Kingdom 48:161-175. ment, p. 389-415. In B. Austin (ed.), Methods in Aquatic Bac-
MEADOWS, P. S., J. T,~T, AND S. A. HUSSAIN. 1990. Effects of teriology. John Wiley and Sons, New York.
estuarine infauna on sediment stability and particle sedimen- WILLS, B. A. 1988. Mineral processing technology: An intro-
tation. Hydrobiologia 190:263-266. duction to practical aspects of ore treatment and mineral re-
MEHTA, A.J. 1986. Introduction, p. 251-269. In A. J. Mehta covery. Pergamon, New York.
(ed.), Estuarine Cohesive Sediment Dynamics, Volume 14.
Springer, New York. Received for consideration,June 10, 1996
MENDOZA,J. A. 1982. Some aspects of the autecology of Lepto- Acceptedfor pubhcation, March 17, 1997