Bivalve provinces in the Proto-Atlantic and along the southern margin of the Tethys in the Jurassic

PAL 0 ELSEVIER Palaeogeography, Palaeoclimatology, Palaeoecology 137 ( 1998 ) 127 151 Bivalve provinces in the Proto-Atlantic and along the southern margin of the Tethys in the Jurassic Chunlian Liu, Michael Heinze, Franz T. F a r s i c h * hTstitut /iTr Paldontologie der Universit6t Wiirzburg, Pleicherwall 1. D-97070 Wiir=burg. Germany Received 29 August 1996: accepted 5 June 1997 Abstract Using multivariate methods, the distribution pattern of Jurassic bivalve genera and subgenera within the ProtoAtlantic and along the southern margin of the Tethys is analysed based on an extensive literature survey and examination of field collections. Four bivalve provinces are recognised within the Proto-Atlantic: (1) the Boreal Bivalve Province in the north: (2) the Northern Transitional Province: (3) the Southern Transitional Province: and (4) the Mediterranean Bivalve Province. The former two provinces are assigned to the Boreal Faunal Realm, the latter two to the Tethyan Faunal Realm. The southwestern margin of the Tethys belongs to the Mediterranean Bivalve Province, the southeastern margin to the Ethiopian Bivalve Province. The latter is only poorly defined by bivalve genera/subgenera: in particular, its northern boundary is blurred, with the Arabian region occupying an intermediate position between the Mediterranean and the Ethiopian bivalve provinces. Province boundaries based on bivalves do not always coincide with those based on ammonites. The reason for this is probably different modes of life: ammonites, as a rule, being active swimmers and bivalves being substrate-related and subjected to passive dispersal during the larval stage. Diversity is not suitable to characterize bivalve provinces, due to differences in outcrop size, preservation potential and sampling history, and due to differences in the degree of environmental heterogeneity within and between regions. Migration patterns of Jurassic bivalves either reflect changes in the climatic condition (e.g. the northward spread of Tethyan taxa in the late Early Jurassic and Middle Jurassic in the course of amelioration of the climate) or prevailing current patterns. The origin of faunal provinciality is thought to be complex and a result of the interplay of several factors, the importance of which varied throughout the Jurassic. © 1998 Elsevier Science B.V. Keywords. palaeobiogeography; Jurassic: Bivalvia; Tethys; Proto-Atlantic; cluster analysis I. Introduction P r o v i n c i a l i t y in Jurassic m a r i n e f a u n a s is well k n o w n . First d e m o n s t r a t e d by the d i s t r i b u t i o n p a t t e r n o f a m m o n i t e s within N o r t h A t l a n t i c to M e d i t e r r a n e a n areas by N e u m a y r (1883), such * Corresponding author. Fax: +49 931 57705. E-mail: ffanz'fuersich@rzr°e'uni-wuerzburg'de 0031-0182/98/$19.00 ,<) 1998 Elsevier Science B.V. All rights reserved. PII S0031-0182(97)00096-5 p r o v i n c i a l i t y has since been recognised on a global scale, with the r e c o g n i t i o n o f several faunal realms a n d provinces (e.g. Arkell, 1956; H a l l a m , 1969, 1994; C a r i o u , 1973; Enay, 1980; W e s t e r m a n n , 1981: Doyle, 1987; Riccardi, 1991; see Table 1 for the P r o t o - A t l a n t i c ) . It was H a l l a m ( H a l l a m , 1977, 1983) who first e x p l o r e d in greater detail the extent benthic bivalves m i r r o r e d the d i s t r i b u t i o n p a t t e r n o f nektic a m m o n i t e s . H e recognized four bivalve 128 ( i L m ~'r a~ Pa/~t~'(<,~'o,,,,ra/d~. Po/ac'(~/imafolok, y. P,t/dco~.'~o/o.q,l' 137 / 19~,%'/ 127 151 Table I Faunal realms ~llld ptovillces in lhe Prolo-Atlantic based Oll ~.tll~nlOniles and bixal\cs Realms based on anlmonitcs Proxinces based on anlnlollites Provinces b~tsed on bi~alxc> Realms based on bikalvcs Boreal Bore~d Subboreal Boreal Northern Wvansitiomd Boreal Tcthynn Submcditevvancan Southern Transitional Mediterranean "l'eth~nn Mediterranean set, we demonstrate that the origin of faunal provincialily in the Jurassic cannot be related to just a single factor but is the result of a complex interplay of factors, the importance of which varied throughout the Jurassic. In addition, by extending Liu's ( Liu, 1995 ) study from the Proto-Atlantic to the southern margins of the Tethyan Ocean we unravel the relationship of Jurassic biwdves across this ocean and assess the importance of the Tethys as a physical barrier to benthic organisms. 2. Data base provinces ill the Early Jurassic: ( 1 ) the European Province: (2) tile East Asian Province: (3) tile Southwest Pacific Province" and (4) the West American Province, supplemented by the Ethiopian Province from tile Mid Jurassic onwards. Subsequent work on the palaeobiogeography of Jurassic biwdves dealt largely with the Pacific or Arctic regions (e.g. Hayami. 1984. 1987, 1989: Crame, 1986, 1987, 1991: Damborenea, 1993) or discussed the distribution pattern of particular taxa such as Lithiotis ( Nauss and Smith. 1988) or buchiids (Sha and Ffirsich. 1994). Recently, Liu (1995) carried out an extensive survey of Jurassic bivalves along the margins o[" the Proto-Atlantic to study their palaeobiogeographic relationships. This was possible because a large new data base of the distribution pattern of Jurassic bivalves in time. space and different types of facies had become available in recent years through taxonomic or palaeoccological studies {e.g. Etirsich, 1977, 1982: Duff, 1978: Kelly, 1984, Kelly, 1992: Ftirsich and Werner. 1988. 1989; Oschmann. 1988; Clausen and Wignall, 1990: Heinze. 1991). Thus, one of the weaknesses of many palaeobiogeographic studies an insufficient data base is thought to have been at least partially overcome. The origin of limnal provinciality within the ,lurassic is still controversial: past explanations include, amongst other Factors, salinity w~riations and temperature gradients, physical barriers, and sea level changes (e.g. Arkell. 1956: Halhu-n. 1969. 1977, 1983: Gordon, 1975: FClrsich and Sykes, 1977: Riccardi, 1991). Using an improved data Many palaeobiogeographic studies are severely hampered by an inadequate data base. Fortunately, century-long palaeontological research in the European Jurassic has produced a high quality data base. Although, in the case of bivalves, genera yield considerably less detailed palaeogeographic information than species (Heinze, 1996), we have refrained from using species as the basic taxonomic unit because evaluation of published taxonomic information is critical for the quality of the data and cannot always be carried out at the desired species level. At the generic/subgeneric level, a taxonomic evaluation of published inl\~rmation was feasible and resulted in revised and accurate faunal lists for the respective localities. The Proto-Atlantic data matrix has been published (Liu, 1995, appendix 2). Data for tile southern Tethys has been deposited m the library of the Inslitul ./fir Paliiomolo~ie of Wtu'zburg University under the number PIW1996VI. It has been compiled from the literature on the southern margin of the Tethyan Ocean (mainly Sowerby, 1840; Stoliczka, 1865; Kitchin. 1903: Daqu& 1910; Holdhaus, 1913: Douville, 1916, 1925: Cox. 1925, 1935a,b, 1940, 1952, 1965; Stefanini, 1925. 1939; Weir, 1925.Weir, 1929; Barrabe, 1929: Dietrich, 1933: Jaboli. 1959: Desio et al., 1960: Freneix, 1965: Mongin. 1967: Hirsch. 1980: Parnes, 1981; Manivit et al., 1990; Jaitly et al, 1995; Holzapfel, 1997 ). These literature data were augmented by extensive collections of biwdves, particularly flom Morocco, Tunisia, western India, Jordan and the Sinai Peninsula, (It Liu ut aL . Pakwo,eeo~raplo', Pakwoclimalolo,zv, Palaeoecolo,.,,y 137/199,¥2 127 151 made in the last few years by members of the Wiirzburg Palaeontological Institute. The data matrix was subdivided into time slices corresponding to the eleven stages of the Jurassic {Table 2). This should allow us to trace changes in the palaeobiogeographic pattern through time. Furthermore, the substrate in which a bivalve occurred was noted in order to evaluate the importance of facies on the distribution pattern. As detailed information on the facies is often not available, only three major groups of facies have been distinguished: fine-grained siliciclastics, coarse-grained siliciclastics, and carbonates. Table 2 Geographic areas, time slices and facies types in which the data base has been grouped Proto-Atlantic Southern Tethyan Shelf 129 Despite this, a substrate-related analysis was possible only for some stages of the Jurassic. The geographic areas selected for the study range fl'om East Greenland to Morocco in a north-south direction and from Mexico/Texas to Kachchh (western India) in a northwest southeast direction {Table 2). Most palaeobiogeographic studies are beset by the problem that the available information is not evenly distributed among areas, time slices, and facies types. This is particularly true when the analysis is very detailed. As a result, the significance of the results may vary considerably. For this reason we restrict our analysis of the southern Tethys margin to the Bathonian and Callovian time intervals, for which an adequate data base is available. 3. Numerical methods Geographic units East Greenland Scotland/northern England Southern England Northern France Southwest France Spain Portugal Morocco Texas Mexico Tunisia Arabia.'Near East East Africa West India (Kachchh) Time .v/ices Tithonian Kimmeridgian Oxfordian Callovian Bathonian Bajocian Aalenian Toarcian Pliensbachian Sinemurian Henangian Callovian Bathonian Although palaeobiogeographic studies should be based on abundance data, such data are hardly ever available and must be substituted by presence absence data ( binary data), which are usually sufficient for large-scale palaeobiogeographic studies (e.g. Digby and Kempton, 1987; Shi, 1993). For the present study, presence absence data were used and treated by multivariate analysis. A comparison of various analytical approaches by Liu (1995) showed that the Q-mode cluster analysis with the weighted pair-group method using arithmetic averages ( W P G M A ) and based on the Simpson Similarity Coefficient or, in the case of the Pliensbachian, oll the Simple Matching Coefficient (e.g. Simpson, 1960; Newton, 1990; Shi. 1993 ) yielded the biologically most meaningful results. (A detailed discussion of the methodology is found in Liu, 1995.) 4. Results fine-grained siliciclastics (clay. silt): T coarse-grained siliciclastics (sand): S carbonates: C 4.1. Bivalve prot'hlces m the Jurassic PromA l/antic Based on the multivariate analysis discussed above and based on characteristic taxa, four bivalve provinces can be distinguished within the l "~0 ('. Liu el a[. " Palaeo,.,eoyraldly, PalauoclhJlaloh)gy, Pa[auo¢,co/o~l' 137 (199~'~') 127 151 Jurassic Proto-Atlantic (Liu, 1995; table I). They are, from north to south: the Boreal Bivalve Province, the Northern Transitional Province, the Southern Transitional Province and the Mediterranean Bivalve Province. To avoid confusion, Liu (1995) introduced the terms "Northern Transitional Province" and "Southern Transitional Province" because their boundaries only partly coincide with the ammonoid-defined Subboreal and Submediterranean provinces. Not all provinces can be recognized for each time interval considered, and the boundaries of the provinces change throughout the Jurassic. The Boreal Bivalve Province and the Northern Transitional Province are grouped in the Boreal Realm, whereas the Southern Transitional Province and the Mediterranean Bivalve Province belong to the Tethyan Realm (Table 1). This pattern corresponds to that of ammonite provinces and realms within the Proto-Atlantic (e.g. Hallam, 1969, 1971, 1975). Apart from forming a cluster, each province is characterized by several taxa which are either restricted to it within the ProtoAtlantic or are at least conspicuous elements of the fauna, occurring rarely outside that province during the time interval under consideration. Details of the cluster analyses and the resulting bivalve provinces including their characteristic taxa are given in Liu (1995). Four of her maps detailing extent and boundaries of these provinces during the Pliensbachian, Oxfordian, Kimmeridgian and Tithonian are reproduced here in a modified way ( Figs. 1 and 4 -6) as they are the data base l\~r the discussion. In addition, maps of the bivalve provinces during the Bathonian and Callovian that include parts of the southeastern Tethys are given in Figs. 2 and 3. These six time interw~ls are best represented by data. Only for the Callovian can all four provinces be distinguished. During the other stages either the Northern o1" the Southern Transitional Province is missing. 4.2. Bivalve dLs'tribulio, patterns aloHg tile South Tethvan margi17 Figs. 7 and 8 provide dendrograms derived fl-om the cluster analysis in which data from the ProtoAtlantic are combined with data from the southern margin of the Tethys (Morocco, Tunisia, Arabia and Middle East, East Africa and Kachchh) for the Bathonian and Callovian. The analysis ( W P G M A ) has been run twice: once with the geographic areas defined in Table 2 as the basic units and once with data from the geographic areas divided up according to facies types as defined in Table 2. For the Bathonian, the den&ogram based on areas ( Fig. 7A) shows four clusters: (1) East Greenland, (2) southern and northern England, (3) Tunisia, Mexico and Kachchh, and (4) northern and southwestern France, Portugal, Morocco, Arabia, and East Africa. While East Greenland can be interpreted as representing the Boreal Bivalve Province and northern and southern England as representing the Southern Transitional Province there is little meaning in differentiating between the remaining two clusters, as adjacent data points often belong to geographically widely separated areas. Together they could be regarded as representing the Mediterranean Bivalve Province. However, when differences in facies are taken into account, the cluster analysis results in four clusters with East Africa and Kachchh forming a distinct grouping that can be regarded as representing an Ethiopian Bivalve Province. Surprisingly, Arabia does not cluster with East Africa and Kachchh, but forms an independent cluster related to northern Africa, southern European areas, and northern France (Fig. 7B). An analysis of lists of genera from the various areas reveals that it is the presence of taxa such as Fimbria, Pseudis'ocardht, Pterocardia, Prorokhl, bTtegricardium and Gryphaeligmus in Arabia, compared to areas further south, that led to the separation of the two areas in the dendrogram. However, the local abundance of Eli gmus in Arabia as well as in Kachchh and East Africa suggests that the three areas should be combined in a single province. Additional arguments for this come fi-om an analysis of the distribution patterns of some species. Thus, Aji'icogryplmea costelhtta is known from Arabia as well as East Africa: and also occurs rarely in Tunisia. Chhmlys curvivc,'ians is another species restricted in the Bathonian to Arabia and East Africa/Kachchh (see also Hallam, 1977: Heinze, 1996). Clearly, in the Bathonian, Arabia seems to occupy an intermediate position and could be placed with either East Africa and Kachchh, to represent the Ethiopian Bivalve C. Liu et al. / Palaeogeography Palaeoclimatoh~gy, Palaeoecology 137 (1998) 127 151 131 60 ° iiiiii!iiiiiiiiii::::::: !if:: iii: 30 ° • @ • • • -~] Weyla (Weyla) Lithiotis Opisoma Pachymytilus Pachyrisma (Durga) Hipppopodium ~-..~i ::::!!iiiiiiiiiiii!i/ .?:iiiii!!iii!!iii ...... iii!iiiii x~ 0° Boreal BivalveProvince Mediterranean BivalveProvince [~ Southern Transitional Province Pliensbachian Fig. 1. Pliensbachian bivalve provinces and occurrence of some characteristic taxa in the Proto-Atlantic. Base m a p after Barron et al. ( 1981 ), modified after Ziegler (1988). Dashed line: realm boundary based on ammonites. Province, or with North Africa and southern Europe to represent the Mediterranean Bivalve Province. For the Callovian, the dendrogram based on areas (Fig. 8A) shows the Mediterranean Bivalve Province discussed in the context of the ProtoAtlantic to extend eastwards to include Tunisia, whereas Arabia, East Africa and Kachchh belong to a separate cluster which, again, can be interpreted as representative of the Ethiopian Bivalve Province. As an anomaly, Portugal is also grouped with this cluster. Based on facies types (Fig. 8B), the same grouping can be recognised, except that what we regard as the Ethiopian Bivalve Province is split into two subclusters, largely on the basis of facies (carbonates versus siliciclastics). As an oddity, southwestern France groups with Arabia, East Africa and Kachchh. Thus, an Ethiopian Bivalve Province is discernible for both the Bathonian and Callovian. it is more difficult to define the northern boundary of this province and to define the province by genera or subgenera (see discussion below). 5. Discussion 5.1. Acuity o f resolution. genera versus species as data base There can be little doubt that a palaeobiogeographic analysis carried out at the species level C Lit~ ~'t dl. 132 Pa/a~'o'..,~'~,,..,r~qdO' Pa/acoc/iHmlo/~yl'. Pa/ae~J~'co/o.~,l' 137 .~ 199~, ', 127 151 6 0 ,~ .... iiiiiiiiiiiiii/~, 3(1' • Ceratomyopsiv Opis I Trigonopis ) i 7i i i i i i i i i i i i i i i \ \ ............. \ • ::i+i!!iiii: ..... 30: Borea[ Bixalve Province Mediterranean Bivalve Province Ethiopian Bivalve Province S o u t h e r n t r a n s i t i o n a l Prox incc • .~. . . . Fig. 2. Balhonian bixal\c provinces of the Proto-Atlnlltic and southern margin of the Tethys. and distribution of some characteristic taxa. Base map al'ter Barl-Oll el al. (19gl). modilicd after Zieglcr ( 198gL Dashed line=realm boundar> based on Zltllnlonilcs. yields more detailed information than that carried out at the generic 'subgeneric level, if the taxonomic basis is sound. Nevertheless an argunlent l\~r restricting the analysis to genera subgenera is that such a sound taxonomic data base is at present not feasible (see also Hallam, 1977). Although this is generally true, distract progress has been made in Recent decades on the taxonomy of Jurassic bivalves from previously poorly documented areas (e.g. Fthsich, 1982: Damborenea, 1987a,b: Yin and F/]rsich. 1991; Aberhan, 1994: Jaitly el al., 1995: Holzapfel, 1997) or of particular groups such as the pectinids (Johnson. 1984) ol- the bakevelliids (Muster, 1995). Thus. while accepting Hallam's (Hallam, 1977) argumel'lt Oll the whole. Heinze (1996), in comparing bivalve species flom the Ethiopian faunal province with those fl-om Europe, was able to demonstrate that the distinctness o f the Ethiopian faunal province was consider- ably enhanced compared to an analysis at the generic level. This is particularly evident when the nunlber of endemic species is considered ( Fig. 9). However, the necessary information is not yet available to carry out such an analysis for larger areas and we have to be content with the cruder picture offered by genus-based data sets. 5.2. Biva/vc versus altmtonite pazterns Ammonites have received far more attention as palaeobiogeographic indicators in the Jurassic than bivalves. This is understandable because, being the main biostratigraphic tools ill the Jurassic, they were studied m much greater detail than bivalves. Nevertheless, as Hallam (1977) remarked, bivalves are the most diverse and abundant macroinvertebrate group in the Jurassic, occurring in a wide range of environments. Moreover, being nearly ('. Liu et al, Pa/aeo~,,eo:,,rap/s ' Palaeoclimatolo%y, Pa/aeoecolo<v 137 : 1998) 127 151 133 60 ° 30 ° ® • • • • X arcomytilus Arca(Eonavicula) arctica Lopatinia Modiolus (Strimodiolus) Praebuchia ~] Boreal Bivalve Province -[7~ "))i i i i i i : . i i i 2 i i i i i i i i i ................ " i i i i Ethiopian Bivalve Province [III] Northern transitional Province ~] Southern transitional Province i i i i i i i i i t ........ ........ ...iiii. Mediterranean Bivalve Province 1~ 0° i i i i i .... " ............. i i i i i 7 i i i i i i i i i i i i i i( , i. . . . . . . . . ' 30 ° " '"-*)iiiiii' Fig. 3. C a l l o v i a n bivalve p r o v i n c e s o f the P r o t o - A t l a n t i c a n d s o u t h e r n m a r g i n o f the Tethys, a n d d i s t r i b u t i o n o f s o m e c h a r a c t e r i s t i c t a x a . Base m a p a f t e r B a r t o n el al. ( 1981 ). m o d i f i e d a f t e r Z i e g l e r ( 1 9 8 8 ) . D a s h e d l i n e = r e a l m b o u n d a r y b a s e d o n amn3oniles. exclusively benthic, bivalve faunas can be expected to show regional differences in addition to their distinct facies dependence. However, this facies dependence might overprint and obscure the influence of larger-scale biogeographic factors such as climate, physical barriers and tectonic processes. Along the margins of the Proto-Atlantic, biogeographic boundaries based on ammonites and bivalves coincide remarkably well during the Bathonian, Oxfordian, Kimmeridgian and Tithonian (Figs. 2 and 4 6). In contrast, the boundary between the Boreal and Tethyan Realm differs drastically in the Pliensbachian ( Fig. 1 ) with the boundary based on ammonites crossing through southern Spain. whereas that based on bivalves was situated in northern France. During the Callovian (Fig. 3), the discrepancy was less pronounced; the realm boundary based on a m m o nites was situated in northern Spain, while that based on bivalves was in northern France. Apparently, the southward spread of Boreal ammonites in the Callovian, after the physical barrier of the Mid North Sea High had become ineffective, was not mirrored by a similar spread of Boreal bivalves, For the Pliensbachiam and to a lesser degree also for the Callovian, a feasible explanation for the discrepancy in the ammonite and bivalve distribution pattern could be that, on the whole, bivalves spread passively during their planktic larval stage whereas ammonites, as nektic and nektobenthic organisms, also spread actively. In other words, the distribution pattern of bivalves should be much more influenced by current patterns than that of ammonites. This would imply that northward-directed currents dominated in the area of western Europe, at least in the Pliensbachian and Callovian, and caused the discrepancy in the boundaries of faunal realms. 134 C Liu e t a / . ' PalaeoL, eo¢~rap/ly, Palaeoclimcttolo~+v, Pa/aeoecohs~,,y 137 ( 1 9 9 8 ) 127 151 iiiiiii iiiiiiii I 6°0 ~iiiiiiiiii iiiiiiiiiiiilJ ....... iiiiiiiiiii iiiiiiii.. !iiiiiiiii!iiiiiiii!.:!:i iiiiiiiiiiiiiiiiii~ 30° iiiiiiiiiiiiiiiiiiiiiiiii~ ® A rcomytilus • A rctica [] Discoloripes Lopatinia • Buchia (Buchia) • Praebuchia \ " " " ~ Boreal BivalveProvince ~ Mediterranean BivalveProvince -[[~ Northern Transitional Province .......................... Fig. 4. Oxfordian bivalve provinces and occurrence of some characteristic taxa in the Proto-Atlantic. Base map after Barron et al. ( 1981 ), moditied after Ziegler (1988). Dashed line= realm boundary based on ammonites. 5.3. The distinctness o]the Ethiopian Bivalve Province As the cluster analysis for the Bathonian Callovian time interval shows, the Ethiopian Bivalve Province is less distinct than other bivalvebased provinces. The reason for this lies in the transitional nature of the Arabian region which, in the Middle Jurassic, apparently was a spreading centre not only for ammonites (e.g. Enay and Mangold, 1982; Enay et al., 1987) but also for some bivalves which originated in that area and migrated both westward along the southern margin of the Tethys towards North Africa and southern Europe, and southward into the I n d o M a l a g a s y Gulf (see below). With respect to diagnostic genera/subgenera, Heinze (1996) has shown that the province does not include true endemic genera/subgenera. Thus, Eligmus occurs abun- dantly, but is by no means confined to the Ethiopian Province. A characteristic and widespread genus is the arcoid hutogmmmatodon, which managed, however, to cross the Tethyan Ocean by the Bathonian, as is evidenced by its occurrence in northern Tibet. The well known Bathonian "6~)rbula" lyrata, a new genus that awaits description, is abundant in nearshore Bathonian sediments of Kachchh, but also occurs from Madagascar to Morocco and contributes to the blurred boundary of the Ethiopian Bivalve Province. Less common is the oyster AJ)'icogo,phaea, which is known from the Middle East and Somalia and also rarely occurs in Tunisia. The recently erected subgenus hutoweyla is known only from the Callovian of Kachchh (Jaitly et al., 1995), but is very rare: it may represent a true endemic. In contrast, Heinze (1996) showed that the C. Liu el al. Pa&eogeography,Palaeoclimatology, Palaeoecology137 (1998) 127 151 135 ii 160o i!: " • Eomiodon~ " i ~i i i i i i i i i i i i i i i i i i i i i i i ~ • Laternula • Arctica [] Discoloripe • Buchia(Buchia) X~::~X~ • Praebuchia X~K:x.~ Modiolus (Strimodtolus) [~ Boreal BivalveProvince ~] Mediterranean BivalveProvince r~lq Northern Transitional Province .i 30° iiiiiiiiiiiiiiiiiiiii 0° Fig. 5. Kimmeridgian bivalve provinces and occurrence of some characteristic taxa in the Proto-Atlantic. Base map after Barron et al. (1981), modified after Ziegler (1988). Dashed line=realm boundary based on ammonites. Ethiopian Bivalve Province can be well defined at the species level. For example, in the Bathonian and Callovian of Kachchh about 25% of the bivalve species are endemic. Together with an additional 10% of the species, which are restricted to the Ethiopian Bivalve Province, this province has a high degree of distinctness. An example is the pectinid Chlamys curvivarians, which is confined in the Bathonian and Callovian to the Ethiopian Province and only rarely occurs outside of it, in Tunisia, during the Late Callovian and Oxfordian (see also Hallam, 1977). The limid Plagiostoma harronis, according to Hallam (1977) another characteristic endemic species known from Somalia and Abyssinia (Daqu& 1905; Weir, 1929), is, in fact, Late Jurassic in age and may well be a synonym of another, more widespread species of Plagiostoma. A comparison of the Ethiopian Bivalve Province with the Ethiopian Ammonite Province (often also called the Indo-Malagasy Province) is less straightforward because the boundaries of the latter are less clearly defined. Moreover, based on ammonites, some authors differentiate between an Erythrean Province (Arabia and the Middle East) and an Indo East African Province (Gill and Tintant, 1975; Krishna and Cariou, 1990). According to Cariou et al. (1985), the Ethiopian Ammonite Province in the Bathonian extended as far north as Egypt, Israel and Lebanon ("faunes arabo-malgaches"), but several elements characteristic of the region migrated westward across northern Africa (e.g. Enay and Mangold, 1982), thus blurring the boundary. Based on bivalves, the Arabian region occupied an intermediate position between the Mediterranean and the Ethiopian Bivalve Provinces. According to the cluster analysis, the province boundary runs between Arabia 136 I'<tl~co<gco<Wap/. ', Palacoclimalolo~' Palacuccotok,.' 137 ~ 199<~7 127 151 C Liu cl al. \ . ® • [] • • ~] Eomiodon A rcom.vtilus Arctoti~ Discoloripe; Buchia (Buchi . . . . . . . . . . . . . . . . . . . . . . . . 7 30° ......... Mo,liolu., ( S t r t ~ / u s ~ i Camptonectes (mclearnia) "q~ i " Pty{ homya (Ptrchom~a) N N i;;iil;iiiiiiiiiiiiiil.J"~"~,_ Boreal BivalveProvince ~ ' ~ Medilerranean Bivah'cProvince -~m Northern Transitional Province Fig. 6. T i t h o n i a n bix.ah.e pro,Anccs and occurrence o1 some characteristic taxa m the Proto-Atlantic. Base m a p after Barron ct al. { 19811, modilied after Ziegler (19881. Dashed line=realm boundary based on ammoniles. and East Africa (Fig. 2). In the Callovian, the boundaries of the ammonite- and bivalve-based provinces coincide and run north of Egypt and the Middle East (Fig. 3). Thus, on the whole, ammonite- and bivalve-based Ethiopian provinces show the same dilemma: a poorly defined northern boundary due to the marginal position of the spreading centre ( Middle East). 5.4. Diversity patterns Ever since it was recognised that the diversity of organisms exhibits distinct latitudinal gradients both in terrestrial and marine realms (e.g. Fischer. 1960: Stehli et al., 1967 in the case of bivalves), diversity has been regarded a biogeographic tool that should be easily applicable to the fossil record. Hallam (1972) was the first to investigate latitudinal diversity trends of Jurassic bivalves m a semi-quantitative study of bivalve genera of the margins of the Proto-Atlantic. from Greenland to Morocco. He demonstrated that, contrary to expectations, there was an increase in the overall diversity of bivalves northwards. Ftirsich and Sykes (1977) investigated diversity trends of Oxfordian bivalves in the same area and lbund a northward decrease of the overall diversity which they interpreted as resulting from a decrease m environmental heterogeneity in the same direction. Recognising the important influence of litcies oll diversity (species richness), F~rsich and Sykes (1977) then continued to look at diversity trends within particular facies types and could not discriminate a consistent pattern. They explained the lack of distinct diversity trends within specific facies types as due to the influence of regional environmental instability and differing environmental heterogeneity with respect to other environ- C Liu el al. Pa/aeogco~rap/y, [ 1.0 A r- 137 Palaeoc/imato/og. ', Palaeoeco/o:,tv 137 (1998) 127 151 i 0.5 t 0 Arab N-France Portugal Morrocco "Tethys" Bivalve Province SW-France E-Africa Tunisia Mexico t_ Kachchh N-England S.Traasifional Province I-S-England Boreal Bivalve Province Greenland B Boreal Bivalve Province I 1.0 ~ 05 I 0 [- Greenland (S) S-England (C) S-England (S) S.Transitional Provine. S-England (T) N-England (C) N-England (T) N-England (S) Tunisia (T) Tunisia (C) Morocco (T) Mediterranean Bivalve Province Morocco (C) SW-France (C) Portugal (C) N-Fraaee (C) Arab (T) ] _] t Arab (C) E-Africa (T) E-A fric~ (C) Ethiopian Bivalve Province Kacl~hh (T) Kachchh (S) Kachckh (C) Fig. 7. Dendrograms of a Q-mode cluster analysis with the WPGMA, based on the Simpson Similarity Coefficient, fbr Bathonian bivalve genera/subgenera of the Proto-Atlantic and South Tethys margin. (A) Taking 12 areas into consideration (see Table I: Spain has been omitted due to insufficient data) three provinces can be recognized. Note that the Ethiopian Bivalve Province, which clearly shows up in (B) cannot be separated by means of the cluster analysis. (B) Using areas and facies types four provinces are distinguished. C - carbonates; S - sandstones" T - clay and siltstones. 138 C. Liu et al. ' Palaeogcography, Palaeoclintatology, Palaeoccolo~L'137 (19982 127 151 A 1.0 05 I Kachchh E-Africa Arab Ethiopian Bivalve Province 1t Portugal V Morocco Mediterranean Bivalve Province Mexico Tunisia ~ N-France S.Transitional Province SW-Franee ~ N-England N.Transitional Province S-England -Greenland Boreal Bivalve Province B I 1.0 I Kachchh(T) Kachchh(S) E-Africa(T) - S-England (C) Ethiopian Bivalve Province Boreal Bivalve Province ~. Greenland(S) i Ethiopian Bivalve Province 015 S-England(S) S-England(T) N-England(S) N-England(C) N-England(T) N.Transitional Province VlediterraneanBivalve Province 0 Morocco(T) Morocco(C) Tunisia(T) Tunisia(C) Portugal(C) I Arab(T) Arab (C) SW-France(C) E-Africa(C) Kachchh(C) Fig. 8. Dendrograms of a Q-mode cluster analysis with the WPGMA, based on the Simpson Similarity Coefficient, for Callovian bivalve genera :subgenera of the Proto-Atlantic and South Tethys margin. (A) Talking 12 areas into consideration (see Table I; Spain has been omitted due to insufficient data) five provinces can be recognized. (B) Using areas and titcies types five provinces are distinguished. Due to insufficient data Spain and northern France have not been included in the analysis. ("--carbonates: S-sandstones; T-clay and siltstones. Note that the Ethiopian Bivalve Province is split in two clusters, depending on Facies. 139 C Liu el aL /' Palaeogeoeraphy, Palaeoclh72atolog~y, Palaeoecology 137 (1998) 127-151 BIVALVE DIVERSITY n taxa Z 120 8o uJ tarnilies ,o GENERA SPECIES GENERA SPECIES 4O Fig. 9. Comparison of numbers of endemic genera/subgenera and species in the Bathonian and Callovian of the Ethiopian Bivalve Province. mental parameters than substrate. Incidentally, Oxfordian ammonites showed a clear northward decrease in diversity across the same area both at the generic and the species level. A generic decrease in diversity from the Tethyan Realm to the ' B o r e a l ~ t l a n t i c Province" is also shown by Middle Jurassic to Late Cretaceous belemnites (Doyle, 1987). However, belemnite diversity increases again from his Boreal-Atlantic to his Arctic Province. Fig. 10 shows patterns in species richness, expressed by the number of taxa, at the generic/subgeneric and the familial level for the Bathonian, Oxfordian and Kimmeridgian. Whereas overall bivalve diversity decreases northward from northern France to East Greenland, in the Oxfordian and Kimmeridgian, corroborating the findings of Farsich and Sykes (1977), the Bathonian diversity pattern differs in showing low values for northern France and thus no regular decrease northward. This shows again the limitations of the use of diversity patterns. Most likely the relatively low Bathonian diversity in northern France reflects the low environmental heterogeneity, nearly exclusively shallow carbonate platform environments, and not so much a latitudinal pattern. The fact that for the Bathonian, Oxfordian, and Kimmeridgian of northern France and southern England generic diversity values are higher than in East Greenland probably reflects i 40 S ~ A A A A N Fig. 10. Changes in bivalve diversity (expressed by the number of genera and families, respectively) from Northern France to East Greenland for the Bathoniam Oxfordian and Kimmeridgian. both the expected poleward decrease of diversity and low environmental heterogeneity in East Greenland, because carbonate regimes are not present. The reason that the diagram in Fig. 10 does not include data from southern Europe, Morocco and Texas/Mexico is the lack of sufficient data for most of these areas, which automatically leads to low (and misleading) diversity values. Similarly, lack of sufficient data also forced us to restrict the diversity analysis of the southern Tethys margin to the Bathonian Callovian time interval. When looking at bivalve diversities along the southern margin of the Tethys, from Kachchh in the south to Tunisia in the north during the Bathonian and Callovian (Fig. 11), there appears to be a weak trend towards lower values in the north. However, the low values of Tunisia are largely responsible for this 'trend'. In the Bathonian, sediments of southern Tunisia pre- 140 ('. I.iu et a/. Pak~eoyeo,,,raldU'. Pa/ucoc/in~alo/oL,3'. Pa/aeoecolovy 1.77 ~ 199X; 127 151 apply also to some other groups such as belcmnites (Doyle, 1987). BIVALVE DIVERSITY n taxa I [,° I 5.5. t 40 "El S t - - - - m-,- .,,-.4a i --.,~ • ,~ 1 %%~ %% 4%%,%% NW Fig. 11. Changes in bival,,e diversity, expressed by the number o1"genera and IEmilies respecti\,eb, along the southern margin of tile Tcthys. The slight decrease in dix,ersity in a northx~estward direction is an artefacl caused b3 limited reformation from some areas, especially Tunisia. dominantly represent marginal marine, high stress environments (Holzapfel, 1997). In the Callovian. shallow nearshore environments prevail, still to a large extent influenced by fluctuations in salinity. It is therefore not surprising that regional diversity values do not match those from other areas such as Kachchh where fully marine conditions prevailed and environmental heterogeneity was high (pers. observations). In addition, limited exposure and far less extensive sampling of Jurassic rocks of East and Northeast Africa probably contributed considerably to the observed pattern. In Kachchh, in contrast, bivalve faunas have been collected and described systematically over the last decade (e.g. Jaitly et al., 1995: Pandey et al., 1996) and provide a much more comprehensive data base than that of other regions. In conclusion, diversity gradients are a poor tool for defining provinces when applied to Jurassic bivalves, partly due to the lack of sufficient data, partly due to the role of environmental heterogeneity. Similar conclusions were reached by Hallam (1977), when he compared the latitudinal diversity curve of Recent bivalves with Jurassic data from different palaeolatitudes. These findings Di,vwrsal patterns Biogeographic units are rarely static. Temporal changes ill lhe position of province boundaries arc caused to a large extent by expansions or contractions of the occupied ecospace, m other words by dispersal. For a detailed analysis of dispersal patterns high stratigraphic resolutiol~ is a prerequisite. In the case of bivalves, detailed stratigraphic information is often missing and migration analyses have to be carried out at the stage level. Even at this crude level, the northward dispersal of several bivalve taxa of Tethyan affinity, starting in the Aalenian and extending throughout the Bathonian, is a conspicuous feature of the Proto-Atlantic. Fig. 12 shows some representative taxa. such as Arcomrlilu.s, Pachyrisma, Actinostreon, E/i~mus, Triclfites and Dacrvom)'a. In the case of Eli t,mus, dispersal from the southern margin of the Tethys to northern France via Morocco look place within the Bathonian. thus reaching the limit of the stratigraphic resolution. This "Tethyan Spread" of biwflves can be correlated with the regional amelioration of climate at that time (e.g. Frakes et al., 1992), which is corroborated by the existence of tropical carbonate platl\~rms across large areas of France. In addition, the Mid North Sea High served as an efficient barrier (e.g. Ziegler, 1982, 1988) preventing cool currents from the north fi-om entering western and southern European shell" seas. Larger-scale migration patterns are very difficult to establish because of collection failure and insufficient temporal resolution. The few examples shown in Fig. 13 are among the best and demonstrate that several taxa migrated westward along the southern margin of the Tethys, some of them as far as the Proto-Atlantic. The genus El(~,,mus, for example, apparently originated in the Late Bajocian of the Middle East (Israel) and moved southwards into East Aft'lea by the Middle Bathonian to reach Kachchh by the Late Bathonian. In the other direction, the genus reached northern Aft'lea (Tunisia) and France by the Late Bathonian. (The record of Eli~,,mus from C. Liu et al. /Palaeogeography, Palaeoclimatology, Palaeoecology 137 (1998) 127 151 NORTHWARD DISPERSAL OF TETHYAN BIVALVES Bath Baj Aal Hie • ...,-..-......,~• ° Dacryomya Bath Bai Aal Plie Bath Ba] Aal Pile Bath Baj Aal Pile Trichites m • Eligmus •~ u ..._--~• Actinostreon .--~• ~ArcO ~ ~ • - - ' ~ MA O Bath Baj Aal Plie Pachyrisma • mytilus 141 R J' sP A sw-F ,t N-F •• A S-E Bath Aal Plie * N-E N Fig. 12. The northward dispersal of Tethyan bivalves (the Tethyan Spread) during the Early and Middle Jurassic is interpreted as being a result of climatic amelioration. Plie=Pliensbachian; Aal=Aalenian; Bq/=Bajocian; Bath=Bathonian; Mor= Morocco; Sp = Spain: S W-F= southwestern France; N-F= northern France; S-E= southern England; N-E = northern England. the Bajocian of Morocco by Dresnay (1963) has to be regarded as doubtful, due to the lack of biostratigraphic control.) Examples of migrating species include Falcimytilus jurensis and Chlamys curvivarians. The first occurrence of the former is in the Bathonian of Kachchh. By the Callovian, the species had migrated to Tanzania, Arabia, the Middle East and Tunisia and reached the margins of the ProtoAtlantic ('Europe') by the Oxfordian. The oldest record of C. curvivarians is from the Bajocian of Somalia (Stefanini, 1939). In Kachchh the species is known from Bathonian and Callovian strata. By the Callovian it had spread to Arabia and Israel; by the Oxfordian it had reached Tunisia. Another example of migration is 'Corbula' lymta, a relatively large, equivalved corbulid occurring characteristically, but not exclusively, in waters of reduced salinity (Mongin, 1967; own observations). The earliest record is from presumably Lower Bathonian strata of Kachchh (Fiirsich et al., 1994). From there it migrated to North Africa, where it is known from the Upper Bathonian of Morocco (Mongin, 1967), and the Callovian of Tunisia (Holzapfel, 1997). The species is also well known from the Bathonian of Madagascar (e.g. Newton, 1895; Besairie, 1936; Nicolai, 1950). Due to limited stratigraphic resolution it is, however, not clear whether it qriginated in Madagascar or in Kachchh. Some species, such as Propeamussium pumilum, Spondylopecten palinurus, Eopecten velatus and Radulopecten scarburgensis, occur in Europe earlier than in areas of the Ethiopian Bivalve Province. Although these species appear to have migrated southward (Johnson, 1984), the fact that they are not known during this critical time interval from the southern margin of the Tethys (Algeria, Tunisia 142 C Liu el al. Pahteo~eo%ral)hy, Pahteoclimatolo,~v, Palaeoecolo%y 137 (1998) 127 151 60°, ....... :i:ii:~iCr~i:::" ~'i'ii:' 60° : i 30 ° i °° 6 0° Eligmus 30° . . .... '~! \\ -~ 30: Fig. 13. Dispersal patterns of the bivalves "CorhuN" Ivrala and Falc#nytilus/ureusis (A) and (71hmLv.s curvivarians and El&,mus (B) from western India to Europe along the southern margin of the Tethys. The black arrows indicate the general dispersal direction. Bj-Bajocian: B - B a t h o n i a n : E B - E a r l y Bathonian: M B = Middle Bathonian: L B - L a t e Bathonian; C - C a l l o v i a n : O - O x f o r d i a n . and Libya) favours migration across the Tethys at a more easterly position rather than migration along the southwestern shelf areas of the ocean. This would agree with the clockwise current pattern envisaged to have existed in the western part of the Tethyan Ocean in the Middle Jurassic (see below; Heinze, 1996). Moreover. some of these bivalves, such as Eopecten velatus and Propeamussium pumilum, are known from Europe and the Andean Basin of South America from strata of the same age (Aberhan, 1994). Their centre of origin is therefore still uncertain. The preferred northwestward migration of bivalve taxa along the southern margin of the Tethys, from Arabia to North Africa, into Europe in the Middle Jurassic corresponds to the generally assumed current pattern within the Tethys (e.g. Parrish and Barron, 1986). In these models an E W directed equatorial current splits up into a westward-directed northern equatorial current and a southward-directed southern equatorial current at the northeast margin of Africa. The latter, by reaching south into the Malagasy Gulf, could have facilitated the migration of taxa such as Eligmus from the Arabian/Middle East region to East Africa and Kachchh (Heinze, 1996). On the other hand, this southward-directed current must have been accompanied by some undercurrents moving in the opposite direction. This can be postulated on the basis of the northward-directed migration of some taxa from Madagascar and Kachchh. Thus, while the large-scale current pattern does explain westward and southward-directed migration along the southern margin of the Tethys, the C. Liu et al. / Palaeogeogrophy, Palaeoclimatology, Palaeoecology 137 (1998) 127-151 current pattern within the Malagasy Gulf was probably complex and the existence of undercurrents or smaller, wind-induced gyres enabled the larvae of some taxa to migrate in the opposite direction. 5.6. Causes of provincialism Whereas the existence of a pronounced faunal provincialism in the Jurassic is beyond doubt, its origin is far from clear. In the past, explanations for the distribution pattern of organisms, usually ammonites, included changes in salinity, temperature differences, physical barriers, facies, and sea level changes, to name but the most important (see Hallam, 1975 for an extensive review). In the case of bivalves, Hallam (1977) favoured eustatic sea level changes as the main cause of their provincialism. All these explanations have one thing in common: they are monocausal. It is, however, highly unlikely, that for a period of at least 70 million years the distribution pattern of a group of organisms has been governed by a single factor. Potential factors governing the distribution of marine organisms are either biological, such as larval type, life habits, reproductive strategies, and competition, or else abiotic. The latter include: salinity, facies, temperature, marine currents, marine connections, physical or physiological barriers, sea level changes, and global or regional tectonic processes. It may be assumed with some confidence that the influence of biological factors, although certainly not negligible, remained relatively constant throughout the Jurassic; they are therefore not further discussed. In contrast, the various abiotic factors appeared to have differed in importance through time. For example, the formation of the Mid North Sea High in the Late Bajocian and its continued existence in the Bathonian (e.g. Ziegler, 1982, 1988) surely contributed significantly to faunal provinciality by severing connections between northernmost parts of the Proto-Atlantic and areas further south. As a result, this physical barrier formed the boundary between the Boreal and the Tethyan Realm and between the Boreal Bivalve Province and the Southern Transitional Province. Moreover, the barrier enabled warm water species of Tethyan origin, such 143 as the bivalve Eligmus, to spread comparatively far north. The opening of marine connections, usually in the form of straits or seaways, in turn facilitated the faunal exchange with the Proto-Atlantic and the Tethys. For example, the existence of the so-called Hispanic Corridor since the Pliensbachian (e.g. Hallam, 1983; Smith, 1983; Smith and Tipper, 1986; Riccardi, 1991 ) increased the faunal similarity between the Proto-Atlantic and parts of the Tethys/Palaeo-Pacific distinctly, whereby migration of Palaeo-Pacific bivalves through the corridor eastwards dominated over migration of Tethyan taxa in the opposite direction. A classic example is the sudden appearance of the Palaeo-Pacific genus Weyla in the ProtoAtlantic as far north as southwestern France (Fig. 14; Damborenea and Mancefiido, 1979). By the Kimmeridgian, the influence of Palaeo-Pacific bivalves on the faunas of Mexico had become so great that in the various statistical analyses the area formed a distinct cluster and no longer represents part of the Mediterranean Bivalve Province (Fig. 5; see also Liu, 1995). The importance of marine currents in governing faunal distribution has been outlined above. As Heinze (1996) showed, marine currents were not only responsible for migration of bivalves along the southern Tethys margin westward and southward, but probably also for the spread of some taxa (e.g. Mytiloperna patchamensis, Eligmus rollandi and Indogrammatodon) from the southern margin to the northeastern margin of the Tethys and vice versa (e.g. Buchia and Linotrigonia). Differences in the position of ammonite- and bivalve-based province boundaries in the ProtoAtlantic, such as, for example, during the Pliensbachian and Callovian, most likely also reflect the influence of currents; passive dispersal of planktotrophic larvae of bivalves surely producing a different pattern than actively swimming ammonites. Hallam (1977) assumed changes in sea level to be the dominant control of diversity, extinction and endemism of Jurassic bivalves, arguing that certain other important biogeographic factors were more or less constant. In Fig. 15 the degree of endemism and the diversity (expressed by the 144 ('. Liu el al. ,; Palaeo~Zeo~raldly, Palaeoclimatolo~zy, Palaeoecology 137 (1998/ 127 151 Distribution of Weyla ) SW-France I~ Spain I~ ->>>>>>>>; i:i:!:i:i:i:i:i:i: Portugall~ ...................................... ,.....:........................... Morocco • i!iiiiii!i!!iiiiiiii!!!!iiiiiiii!!!iiiii!!!!iiiiiiill i•!iiiii•i••!!iiii••!•i!iiii•••!!!iiiii•!!iii•i•••iiiii•i••••iiii•i•••••iiiiii• Mexico • Andean Basin ii••iiii••!!iii•!!iii•i••i!i•i••i!iiii!iiii••••!iiii•iiiii••iiiii•!iiii••!iiiii•iiii••iiii••iiii•••iiiii time Hettangian Sinemurian Pliensbachian Toarcian t Hispanic Corridor open Fig. 14. The sudden appearance of H}_'3'/a ill tile Proto-Atlantic and neighbouring shelf seas has been related to the opening of the so-called Hispanic Corridor in Central America. Data from Damborenea and Mancefiido (1979) and R. Schmidt-El:fing (pers. commun., 1996). number of genera/subgenera) of bivalves within the Proto-Atlantic is plotted against the sea level curves of Hallam (1988), Haq et al. (1987) and Vail and Todd ( 1981 ). The curves expressing diversity and degree of endemism are more or less inversely correlated. On the whole, a high degree of endemism goes hand in hand with low diversity values. Theoretically, a high global sea level should facilitate faunal exchange between areas, reduce the degree of endemism and, due to larger shelf areas, should lead to an increase in overall species diversity (e.g. Hallam, 1977; Doyle, 1987). Conversely, a low sea level leads to isolation of basins and produces endemic taxa. However, this relationship only partly holds true when sea level curves and the degree of endemism are compared (Fig. 15). Thus, the high sea level of the Toarcian is mirrored by a low degree of endemism among bivalves, and the same relationship is seen in the Oxfordian (when Hallam's curve is taken as a model). During the general rise in sea level from the Hettangian to the Toarcian, the degree of endemism gradually declines. The low degree of endemism in the Bathonian, however, is not accompanied by a high sea level. The diversity curve with peaks in the Pliensbachian, Bathonian and Kimmeridgian corresponds neither to high levels of endemism nor C. Liu et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 137 (1998) 127 151 Stage Diversity Degree of endemism Sea level rise 145 (Number of genera/subgenera) Tithonian Kimmeridgian Oxfordian ) Callovian Bathonian Bajocian Aalenian Toarcian Pliensbachian Sinemurian f Hettangian A B C I I 50 40 I 30 SE = 5% I 20 % I I I I 150 100 50 0 Fig. 15. Comparison of sea level curves, degree of endemism and diversity of bivalves in the Jurassic. The three sea level curves are after Hallam (1988) (A), Haq et al. (1987) (B) and Vail and Todd ( 1981 ) (C). Diversity is expressed by the number of genera/subgenera within the Proto-Atlantic; degree of endemism is expressed by the percentage of endemic taxa. Taxa were regarded as endemic when occurring only in one area (as defined in Table I) of the Proto-Atlantic. SE=standard error. to sea level highstands. Part o f the explanation is that the data for some o f the stages, such as the Hettangian, Aalenian, and Bajocian, are too meagre to be discussed in this context. The diversity trends shown in Fig. 15 are therefore difficult to interpret. It seems that only during some intervals o f the Jurassic sea level exerted a m a j o r influence on the b i o g e o g r a p h y o f bivalves. M a n y Jurassic bivalves are strongly faciesdependent (e.g. Hallam, 1969, 1971, 1977; Ftirsich, 1976). Correspondingly, the distribution pattern o f Jurassic bivalves is related to facies, a fact that had been noticed as early as the end o f the last century (Nikitin, 1886; O r t m a n n , 1896) and is clearly shown by the clusters in Figs. 7B and 8B, where regions with the same facies type show the greatest preference for clustering together. As a result, bivalve diversity will be lower where, due to a particular climatic or bathymetric situation, the heterogeneity o f facies will be lower. This is the case in deeper shelf settings where fine-grained siliciclastics (clay, silt and marl) prevail. This a r g u m e n t has been used by Fiarsich and Sykes (1977) to explain the lower overall bivalve diversity in Boreal environments in the Jurassic. Thus, finally, the distribution pattern o f bivalves is 146 C. Liu et aL/ Palaeolzeography, Palaeoclimatolo~v. Pahwoecolo~,:v 137 (1998) 127 151 strongly influenced by the palaeogeographic setting, which changed considerably in the course of the Jurassic. The importance of palaeosalinity, promoted by Hallam (1969) in the late sixties as the decisive factor for regional differences between faunas in northwestern Europe, has been overestimated. This was recognised by Hallam as early as 1971. Changes in salinity may have played some role where, due to marked regression or regional uplift and relatively humid conditions, marginal marine environments occurred on a regional scale. An example is the Bathonian Great Estuarine Group of Scotland and the 'Upper Estuarine Series' of Eastern England, which formed in connection with the uplift of the Mid North Sea High (e.g. ZiegleL 1982) and in which brackish and hypersaline environments were widespread. Another example is the Late Tithonian of Western Europe, in which widespread brackish and hypersaline conditions were the result of the late Tithonian regression. Temperature gradients have been regarded by several authors (e.g. Neumayr, 1883: Uhlig, 1911; Enay, 1973; Riccardi, 1991 ) as the main cause of Jurassic faunal provincialism. As has been pointed out repeatedly (e.g. Hallam, 1975, 1985: Frakes, 1979) latitudinal temperature gradients in the Jurassic were less steep than at present. For this reason, Hallam (e.g. Hallam, 1975; see also Doyle, 1987) discounted temperature as the decisive factor causing faunal provincialism. On the other hand, the reduced solar radiation at higher latitudes must have resulted in some temperature gradient, and recent palaeoclimatological studies (e.g. Kemper, 1987; Frakes and Francis, 1988, 1990) suggest that the polar regions in the Jurassic were not as warm as has been assumed in the past. Arguments for the existence of an Austral Faunal Realm at high latitudes of the southern hemisphere, matching the Boreal Realm of the northern hemisphere, were put forward by Crame (1987). The presence of bivalves such as Retroceramus, Inocerantus, Anopaea, Arctotis, and Aucellina point to a climatic zonation at high latitudes in the southern hemisphere. Similarly, the bipolar distribution pattern of buchiid and inoceramid bivalves (e.g. Crame, 1987; Sha and Ffirsich, 1994) is indicative of temperature-induced faunal provinciality (see also Crame, 1986). The fact that many large and thickshelled bivalves, such as Diceras, Pachyrisma, Pachymegalodon, Pterocardia and Pachymytilus, are restricted to low latitudes further supports some climatic control of Jurassic bivalve faunas (see also Hallam, 1977, 1994). Thus temperature must have contributed to t:aunal provincialism, but and this is the point it was not the only factor. In conclusion, the distribution pattern of Jurassic bivalves cannot be explained by a single, overruling factor, but most likely is the result of a combination of several factors, the importance of which varied considerably throughout the Jurassic (Fig. 16). This attempt at interpreting Jurassic bivalve provincialism admittedly is not very spectacular. However, perhaps it lies closer to the truth than earlier, monocausal explanations. 6. Conclusions From the above the following conclusions can be drawn: ( 1 ) Among various multivariate analytical methods, the Q-mode cluster analysis, with the weighted pair-group method using arithmetic averages and based either on the Simpson Similarity Coefficient or the Simple Matching Coefficient, produced the most meaningful results when applied to a data matrix of Jurassic bivalve genera/subgenera from 13 areas and three facies types within the ProtoAtlantic and the South Tethys margin. (2) Based on the cluster analysis and the distribution pattern of characteristic taxa, four bivalve provinces can be distinguished within the ProtoAtlantic. From north to south these are: (a) the Boreal Bivalve Province: (b) the Northern Transitional Province; (c) the Southern Transitional Province; and (d) the Mediterranean Bivalve Province. The first two provinces belong to the so-called Boreal Realm, the latter two to the Tethyan Realm. (3) Along the southern margin of the Tethys, from North Africa to Kachchh and Madagascar, C. Liu et al. /Palaeogeography, Palaeoclimatology, Palaeoecology 137 (1998) 127 151 147 Main factors governing the distribution pattern of Jurassic bivalves in the Proto-Atlantic Late Jurassic Middle Jurassic ] Early Jurassic Fig. 16. Main factors controlling the distribution pattern of Jurassic bivalves in the Proto-Atlantic and their relative importance through time. two provinces are recognised in the BathonianCallovian; these are the Mediterranean Bivalve Province and the Ethiopian Bivalve Province. The latter province, and in particular its northern boundary, is difficult to define on the basis of genera/subgenera, but quite distinct when based on species. (4) The boundaries of provinces based on ammonites and bivalves do not always coincide. This can be explained by the differing mode of life of the two groups (active swimming versus benthic and passive drifting during the larval stages of ammonites and bivalves, respectively). (5) Diversity patterns are a poor tool for defin- ing and characterizing Jurassic bivalve provinces, partly due to lack of sufficient data, partly due to the role of environmental heterogeneity. (6) Within the Proto-Atlantic, several Tethyan bivalve taxa migrated northwards between the late Early Jurassic and the Middle Jurassic as a result of climatic amelioration. In agreement with an assumed E - W directed equatorial current, bivalves show a preference for migrating in a northwestward direction along the southern margin of the Tethys, from western India/Madagascar to northwest Africa and Europe. (7) Faunal provincialism within Jurassic bivalves is explained as the result of several factors, 148 C. Liu el al. / Palaeogeo~zraphy, Palaeocl#,atology, Palaeoecology 137 (1998) 127 151 the importance of which varied considerably throughout the Jurassic. Acknowledgements We would like Chris McRoberts, the manuscript. from reviews by A.L.A. Johnson, thanks financial to thank Martin Aberhan and Wt~rzburg, for critically reading The paper greatly benefitted A. Hallam, Birmingham, and Derby. We acknowledge with support by the Deutsche Forschungsgemeinschq/t (grants Ful 31/12-1 and Fu131/15-1). References Aberhan, M., 1994. Early Jurassic Bivalvia of northern Chile. Part 1. Subclasses Palaeotaxodonta, Pteriomorphia, and lsofilibranchia. Beringeria 13, 3 115. Arkell, W.J., 1956. Jurassic Geology of the World. Oliver and Boyd, Edinburgh, 806 pp. Barrab6, L., 1929. Contribution ~ l'6tude stratigraphique et petrographique de la partie mediane du Pays Sakalave (Madagascar). M6m. Soc. G6ol. Ft. new set. 12, 1 270, pls. 1 23. Barron, E.J., Harrison, C.G.A., Sloan II, J.L., Hay, W.W., 1981. Paleogeography, 180 million years ago to the present. Eclogae Geol. Helv. 74. 443 470. Besairie, H., 1936. Fossiles du Bathonien moyen. Mem. Acad. Malgache21, 120 122. Cariou, E., 1973. Ammonites of the Callovian and Oxfordian. In: Hallam, A. (Ed.), Atlas of Palaeobiogeography. Elsevier, Amsterdam, pp. 287 295. Cariou, E., Contini, D., Dommergues, J.L., Enay, R., Geyssant, J., Mangold, C.. Thierry. J., 1985. Biogeographie des ammonites et ~volution structurale de la TOthys au cours du Jurassique. Bull. Soc. Geol. Ft. 8, 679 697, Clausen. C.K., Wignall, P.B., 1990. Early Kimmeridgian bivalves of southern England. Mesozoic Res. 2. 97 149. Cox, L.R., 1925. A Bajocian Bathonian outcrop in the Jordan Valley and its molluscan remains. Annu. Mag. Nat. Hist. 15 {9), 169 181. Cox, L.R., 1935a. Jurassic Gastropoda and Lamellibrunchiata. In: Macfayden, W.A. (Ed.), The Mesozoic Palaeontology of British Somaliland 11. London, pp. 148 197. Cox, L.R., 1935b. The Triassic, Jurassic and Cretaceous Gastropoda and Lamellibranchia of the Attock District (Punjab). Mere. Geol. Surv. India, Palaeontol, lndica, new ser. 20 (5), I 27. Cox, L.R., 1940. The Jurassic lamellibranch fauna of" Kuchh (Cutch). Mere. Geol. Surv. India, Palaeontol. lndica, Ser. 9 3 (3), 1 157. Cox, L.R., 1952. The Jurassic lamellibranch fauna of Cutch (Kachh). no. 3, Families Pectinidae, Amusiidae, Plicatulidae. Limidae, Ostreidae, and Trigoniidae (Suppl.). Mem. Geol. Surv. India, Palaeont. lndica, Ser. 9 3 (4), 1 128. Cox, L.R.. 1965. Jurassic Bivalvia and Gastropoda from Tanganyika and Kenya. Bull. Brit. Mus. Nat. Hist. {Geology), Suppl. 1, I 213. Crame. J.A., 1986. Late Mesozoic bipolar bivalve faunas. Geol. Mag. 123,611 618. Crame, J.A.. 1987. Late Mesozoic bivalve biogeography of Antarctica. In: Proc. 6th Gondwana Symp. Am. Geophys. Union, Washington, D.C., pp. 93 102. Crame, J.A., 1991. Bipolar molluscs and their evolutionary implications. Palaeontoh Newsl. 12, 9 Damborenea, S.E., 1987a. Early Jurassic Bivalvia of Argentinia. Part 1: Stratigraphical introduction and superfamilies Nuculanacea. Arcacea, Mytilacea and Pinnacea. Palaeontographi c a ( A ) 199, 23 111. Damborenea, S.E.. 1987b. Early Jurassic Bivalvia of Argentinia, l'art 2: Superfamilies Pteriacea, Buchiacea and part of Pectinacea, Palaeontographica (A) 199, I 13 216. Damborenea, S.E., 1993. Early Jurassic South American pectinaceans and circum-Pacific palaeobiogeography. Palaeogeogr. Palaeoclimatol. Palaeoecol. 100. 109 123. Damborenea, S.E.. Mancefiido, M.O., 1979. On the palaeogeographical distribution of the pectinid genus ~ ' y / a ( Bivalvia, Lower Jurassic). Palaeogeogr. Palaeoclimatol. Palaeoecoh 27, 85 102. Daqu6, E., 19(15. Beitrgtge zur Geologie des Somalilandes. Zweiter Teil: Oberer Jura. Beitr. Palfiontol. Geol. Osterr.Ung. 17, 119 159. Daqud, E., 1910. Dogger und Maim aus Ostafrika. Beitr. Pal~ontol. Geol. Osterr.-Ung. 23, 1 62. Desio, A., Rossi Ronchetti, C., lnvernizzi, G., 1960. 11 giurassico dei dintori di Jefren in Tripolitania. Riv. ltah Paleontol. 46, 65 I18. Dietrich, W.O., 1933. Zur Stratigraphie und Palaeontologie der Tendaguruschichten. Palaeontographica, Suppl. 7 I {2). 1 8. Digby, P.G.N., Kempton, R., 1987. Multivariate Analysis of Ecological Communities. Chapman and Hall, London, 203 pp. Douvill6. H.. 1916. Les terrains secondaires dans le massif du Moghara ~t l'est de l'lsthme de Suez, d'apr~s les exploration de M. Couyat Barthoux. M6m. Acad. Sci. Paris 54, I 184. Douvill& H., 1925. Le Callovien darts le Massif du Moghara. Bull. Soc. G6ol. Fr. 25 (IV), 303 328, pls. 5 8. Doyle, P., 1987. Lower Jurassic Lower Cretaceous belemnite biogeography and the development of the Mesozoic Boreal Realm. Palaeogeogr. Palaeoclimatol. Palaeoecoh 61, 237 254. Dresnay. R., 1963. Donnees stratigraphiques complementaires sur le Jurassique moyen des synclinaux d'EI Mers et de Skoura (Moyen-Atlas, Maroc). Bull. Soc. G~ol. Ft. 5, 883 900. Duff, K.L., 1978. Bivalvia from the English Lower Oxford Clay (Middle Jurassic). Palaeontogr. Soc. Monogr., 137 pp. Enay, R., 1973. Upper Jurassic (Tithonian) ammonites. In: C Liu et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 137 (1998) 127 151 Hallam, A. (Ed.), Atlas of Palaeobiogeography. Elsevier, Amsterdam, pp. 297-307. Enay, R., Pal6obiog6ographie et ammonites jurassiques: "'rythme faunique" et variations du niveau marin; voles d'6changes~ migration et domaines biog6ographiques. Livre Jubilaire de la Soci6t6 g6ologique de France, 1880-1980. 1980. Mem. h.-s. Soc. G6ol. Ft. 10, 261 281. Enay, R., Marigold, C., 1982. Dynamic biog6ographique et 6volution des faunes d'ammonites au Jurassique. Bull. Soc. G6ol. Fr. 24, 1025 1046. Enay, R., Le Nindre, Y.M., Mangold, C., Manivit, J., Vaslet, D., 1987. Le Jurassique d'Arabie saoudite centrale: nouvelles donn6es sur la lithostratigaphie, les pal6oenvironnements, les faunes d'ammonites, les figes et les correlations. Geobios, M6m. Spec. 9, 13-65. Fischer, A.G., 1960. Latitudinal variations in organic diversity. Evolution 14, 64 81. Frakes, L.A., 1979. Climates Throughout Geologic Time. Elsevier, Amsterdam, 310 pp. Frakes, L.A., Francis, J.E., 1988. A guide to Phanerozoic cold polar climates from high-latitude ice-rafting in the Cretaceous. Nature 333, 547-549. Frakes, L.A., Francis, J.E., 1990. Cretaceous palaeoclimates. In: Ginsburg, R.N., Beaudoin, B. (Eds.), Cretaceous Resources, Events and Rhythms. Kluwer, Dordrecht, pp. 273 287. Frakes, L.A., Francis, J.E., Syctus, J,I., 1992. Climate Modes of the Phanerozoic. Cambridge Univ. Press, 274 pp. Freneix, S., 1965. Les bivalves du Jurassique moyen et sup6rieur du Sahara Tunisien (Arcacea, Pteriacea, Pectinacea, Ostreacea, Mytilacea). Ann. Pal6ontol. 51, 1 113. F~rsich, F.T., 1976. Fauna-substrate relationships in the Corallian of England and Normandy. Lethaia 9, 343 356. F~rsich, F.T., 1977. Corallian ( Upper Jurassic) marine benthic associations from England and Normandy. Palaeontotogy 20, 337 385. FiJrsich, F.T., 1982. Upper Jurassic bivalves from Milne Land, East Greenland. Gronl. Geol. Unders. Bull. 144, 126 pp. F~rsich, F.T., Sykes, R.M., 1977. Palaeobiogeography of the European Boreal realm during Oxfordian (Upper Jurassic) times: a quantitative approach. Neues Jahrb. Geol. Pal~tontol. Abh. 155, 137 161. Farsich, F.T., Werner, W., 1988. The Upper Jurassic Bivalvia of Portugal. Part I. Palaeotaxodonta and Pteriomorphia (Arcoida and Mytiloida). Comun. Serv. Geol. Portugal (for 1987) 73, 103-144. Fiirsich, F.T., Werner, W., 1989. The Upper Jurassic Bivalvia of Portugal. Part lI. Pteriomorphia (Pterioida exclusive Ostreina). Comun. Serv. Geol. Portugal (for 1988) 74, 105 164. Fiirsich, F.T., Pandey, D.K., Oschmann, W., Jaitly, A.K., Singh, I.B., 1994. Ecology and adaptive strategies of corals in unfavourable environments: Examples from the Middle Jurassic of Kachchh Basin, western India. Neues Jahrb. Geol. Pal~ontol. Abh. 194, 269 303. Gill, G.A., Tintant, H., 1975. Les ammonites calloviennes du 149 Sud d'Israel, stratigraphie et relations pal6ogeographiques. C. R. Somm. S6ances Soc. Geol. Fr. 1975, 103-106. Gordon, W.A., 1975. Origin of the Mesozoic Boreal realm. Geol. Mag. 112, 199 201. Hallam, A., 1969. Faunal realms and facies in the Jurassic. Palaeontology 12, 1 18. Hallam, A., 1971. Provinciality in Jurassic faunas in relation to facies and palaeogeography. In: Middlemiss, F.A., Rawson, P.F., Newall, G. (Eds.), Faunal Provinces in Space and Time. Geol. J. Spec. Issue 4, 129 152. Hallam, A., 1972. Diversity and density characteristics of PliensbachiawToarcian molluscan and brachiopod faunas in the North Atlantic margins. Lethaia 5, 389 412. Hallam, A., 1975. Jurassic Environments. Cambridge Univ. Press, Cambridge, 269 pp. Hallam, A., 1977. Jurassic bivalve biogeography. Paleobiology 3, 58 73. Hallam, A., 1983. Early and mid-Jurassic molluscan biogeography and the establishment of the Central Atlantic Seaway. Palaeogeogr. Palaeoclimatol. Palaeoecol. 43, 181 193. Hallam, A., 1985. A review of Mesozoic climates. J. Geol. Soc. London 142, 433 445. Hallam, A., 1988. A reevaluation of Jurassic eustasy in the light of new data and the revised Exxon curve. SEPM Spec. Publ. 42~ 261 273. Hallam, A., 1994. An Outline of Phanerozoic Biogeography. Oxford Biogeography Series, 10. Oxford Univ. Press, Oxford, 246 pp. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science 235, 1156 1167. Hayami, l., 1984. Jurassic marine bivalve faunas and biogeography in Southeast Asia. In: Kobayashi, T., Toriyama, R., Hashimoto, W. (Eds.), Geology and Palaeontology of Southeast Asia. Univ. Tokyo Press, Tokyo, pp. 229 237. Hayami, I., 1987. Geohistorical background of Wallace's Line and Jurassic marine biogeography. In: Taira. A., Tashiro, M. (Eds.), Biogeography and Plate Tectonic Evolution of Japan and Eastern Asia. Terra, Tokyo, pp. 111 133. Hayami, I., 1989. Outlook on the post-Paleozoic historical biogeography of pectinids in the western Pacific region. In: Ohba, H., Hayami, I., Mochizuki, K. (Eds.), Current Aspects of Biogeography in West Pacific and East Asien Regions. Univ. Tokyo, Nature and Culture, Tokyo, pp. 3 25. Heinze, M., 1991. Evolution benthonischer Faunengemeinschaften im subborealen Jura des Pariser Beckens und in der ~ithiopischen Faunenprovinz des Beckens von Kachchh (Indien) ein Vergleich. Beringeria 4, 3 126. Heinze, M., 1996. Pal/iobiogeographie jurassischer Muschelfaunen: Beziehung zwischen S/id- und Nordrand der Tethys. Pal~iontol. Z. 70, 97 128. Hirsch, F., 1980. Jurassic bivalves and gastropods from Northern Sinai and Southern Israel. Israel J. Earth-Sci. 28, 128-163. Holdhaus, K., 1913. Fauna of the Spiti Shales (Lamellibranchiata and Gastropoda). Mem. Geol. Surv. India, Palaeontol. lndica Ser. 15 4 (2), 397 456, pls. 94 100. 150 C. Liu el aL ,: Pahwogeo£,raph.r, Palaeoe/imaloloKr. Pahwoeeolo,gy 137 (1998) 127 15l Holzapfel, S., 1997. Pal/Skologie benthischer Faunengemeinschaften und Taxonomic der Bivalven im Jura yon Stidtunesien. Beringeria, 20 (in press). Jaboli, D., 1959. Fossili giurassici dell'Harar (Africa orientale) (Brachiopodi, Lamellibranchiati e Gastropodi). Missione Geologica A.G.I.P. (1936 38). Vol. IV. Documentazione Paleontologica 1. Accad. Naz. dei Lincei, Rome. Jaitly, A.K., Ftirsich, F.T., Heinze, M., 1995. Contributions to the Jurassic of Kachchh, western India. IV. The bivalve fauna. Part I. Subclasses Palaeotaxodonta, Pteriomorphia. and lsofilibranchia. Beringeria 16, 147 257. Johnson, A.L.A., 1984, The palaeobiology of the bivalve families Pectinidae and Propeamussiidae in the Jurassic of Europe. Zitteliana I I, 1 235. Kelly, S.R.A., 1984. Bivalvia of the Spilsby Sandstone and Sandringham Sands (Late Jurassic Early Cretaceous) of eastern England. Part 1. Palaeontogr. Soc. M o n o g r , pp. 1 94. Kelly, S.A., 1992. Bivalvia o1" the Spilsby Sandstone and Sandringham Sands ( Late Jurassic Early Cretaceous) of eastern England. Part I1. Palaeontogr. Soc. Monogr., pp. 95 123. Kemper, E., 1987. Das Klima der Kreide-Zeit. Geol. Jahrb. Ser. A 9 6 , 5 185. Kitchin, F.L., 1903. The Jurassic fauna of Cutch. The Lamellibranchiata. Genus Trig,onia. Mem. Geol. Surv. India, Palaeontol. lndica Ser. 9 3 (2), I 122. Krishna, J., Cariou, E., 1990. A m m o n o i d faunal exchanges during Lower Callovian between the lndo East-African and submediterranean Provinces: Implications for the long distance east west correlations. Newsl. Stratigr. 23, 109 122. Liu, Chunlian, 1995. Jurassic bivalve palaeobiogeography of the Proto-Atlantic and the application of multivariate analysis methods to palaeobiogeography. Beringeria 16.3 123. Manivit, J., Le Nindre, Y.M.. Vaslet, D., 1990. Le Jurassique d'Arabie Centrale. Vol. 4. Histoire g6ologique de la bordure occidentale de la plate-forme arabe. Dec. B R G M 194, 559 pp. Mongm, D., 1967. Les mollusques du Bathonien s'aumatre du Moyen Atlas. Not. Mem. Serv. G~ol. Maroc 200~ 37 94. Muster, H., 1995. Taxonomie und Pal~iobiogeographie der Bakevelliidae (Bivalvia). Beringeria 14, 1 161. Nauss, A.L., Smith, P.L., 1988. LiHfiods (Bivalvia) bioherms in the Lower Jurassic of East-Central Oregon, U.S.A. Palaeogeogr. Palaeoclimatol. Palaeoecol. 65, 253 268. Neumayr, M., 1883. Ober klimatische Zonen wiihrend der Juraund Kreidezeit. Denkschr. K. Akad. Wiss. Wien, Math.Naturwiss. K1.47.277 310. Newton, C.R., 1990. Palaeobiogeography. In: Briggs, D.E.G., Crowther, P.R. (Eds.), Palaeobiology. A Synthesis. Blackwell, Oxford, pp. 452 460. Newton, R.B.. 1895. Notes on fossils IYom Madagascar, with descriptions of two new species of Jurassic pelecypoda from that island. Q. J. Geol. Soc. London 45, 331-339. Nicolai, M., 1950. Paleontologic de Madagascar. XXIX. E,tude de quelques gisements fossiliferes du sud-ouest de Madagascar. Ann. Paldontol. 36. 141 168: 37. 1 46. Nikitin, S., 1886. Ober die Beziehung zwischen der russischen und west-europfiischen Juraformation. Neues Jahrb. Mineral. Geol. Pal~ontol. 2, 205 245. Ortmann, A.E., 1896. An examination of the arguments given by N e u m a y r for the existence of climatic zones in Jurassic times. Am. J. Sci. 4 ( 1 ), 257 270. Oschmann. W , 1988. Upper Kimmeridgian and Portlandian marine macrobenthic associations from southern England and northern France. Facies 18, 49 82. Pandey, D.K., Ffirsich. F.T., Heinze, M., 1996. Contributions to the Jurassic of Kachchh, Western India. V. The bivalve fauna. Part II. Subclass Anomalodesmata. Beringeria 18, 51 87. Parnes, A., 1981. Biostratigraphy of the Mahmal Formation ( Middle and Upper Bajocian) in Makhtesh R a m o n ( Negev, southern Israel ). Bull. Geol. Surv. Israel 74. 1 55. Parrish, J.T., Barrom E.J., 1986. Paleoclimates and economic geology. SEPM Short Course 18, 1 162. Riccardi, A.C., 1991. Jurassic and Cretaceous marine connections between the Southeast Pacitic and Tethys. Palaeogeogr. Palaeoclimatol. Palaeoecol. 87. 155 189. Sha, Jingeng, Ftirsich, E.T., 1994. Bivalve Iitunas of eastern Heilongjiang. northeastern China. 11. The Late Jurassic and Early Cretaceous buchiid fauna. Beringeria 12, 3 93. Shi, G.R., 1993. Multivariate data analysis in palaeoecology and p a l a e o b i o g e o g r a p h y - a review. Palaeogeogr. Palaeoclimatol. Palaeoecol. 105, 199 234. Simpson, G.G., 1960. Notes on the measurement of faunal resemblance. Am. J. Sci. 258a, 300 31 I. Smith, P.L., 1983. The ammonite Dayicera.s ~ko'iceroides and Early Jurassic paleogeography. Can. J. Earth Sci. 20, 86 91. Smith, P.L.. Tipper, H.W., 1986. Plate tectonics and paleobiogeography: Early Jurassic (Pliensbachian) endemism and diversity. Palaios 1,399 412. Sowerby, J. de C., 1840. Description of fossils from the Upper Secondary Formation of Cutch collected by C.W. Grant. Trans. Geol. Soc. London 2 (5), pls. 21 23, Figs. 8 17. Stefanini, (3., 1925. Description of fossils from South Arabia and British Somaliland. In: Little, O.H. (Ed.), The Geography of Makalla (South Arabia). Geol. Surv. Egypt, pp. 143 221. Stefanini, G.. 1939. Molluschi del Giuralias della Somalia. Paleontogr. ltal. 32 (4), 103 270. Stehli, F.S.. McAlester, A.L., ffelsley, C.E.. 1967. Taxonomic diversity of Recent bivalves and some implications for geology. Geol. Soc. Am. Bull. 78, 455 466. Stoliczka, F., 1865. Geological sections across the Himalaya Mountains, from Wangtu-bridge on the river Sutley to Sungdo on the Indus: with an account of the formations in Spiti, accompanied by a revision of all known fossils from that district. Mem. Geol. S u r v l n d i a 5(1), I 154, pls. I 10. Uhlig, V.. 1911. Die marinen Reiche des Jura und der Unterkreide. Mitt. Geol. Ver. Wien 3, 329 448. Vail. P.R., Todd, R.G., 1981. Northern North Sea Jurassic unconformities, chronostratigraphy and sea level changes from seismic stratigraphy. In: lllings, L.V.. Hobson, G.D. (Eds.), Petroleum Geology of the Continental Shelf of North-West Europe. Inst. Pet.. London, pp. 216 235. C Liu et al. / Palaeogeography, Palaeoclimatolog), Palaeoecology 137 (1998) 127 151 Weir. J., 1925. Brachiopoda, Lamellibranchiata, Gastropoda and Belemnites. In: Wyllie, B.N.K., Smellie, W.R. (Eds.), The Collection of Fossils and Rocks from Somaliland. Monogr. GeoL Dept. Hunterian Mus. Glasgow Univ. 1, 79 110. Weir, J., 1929. Jurassic fossils from Jubaland, East Africa, collected by V.G. Glenday, and the Jurassic Geology of Somaliland. Monogr. Geol. Dept. Hunterian Mus. Glasgow Univ. 1,1 63. Westermann, G.E.G., 1981. Ammonite biochronology and bio- 151 geography of the Circum-Pacific Middle Jurassic. In: House, M.R., Senior, J.R. (Eds.) The Ammonoidea. System. Assoc. Spec. Publ. 18, 459 498. Yin, Jiarun, Fi~rsich, F.T., 1991. Middle and Upper Jurassic bivalves from the Tanggula Mountains, W-China. Beringeria 4, 127-192. Ziegler, P., 1982. Geological Atlas of Western and Central Europe. Elsevier, Amsterdam, 130 pp. Ziegler, P., 1988. Evolution of the Arctic North Atlantic and the Western Tethys. AAPG, Tulsa, 197 pp.