mid-cretaceous rudists (bivalvia: …ibayasuhiro.com/publications_files/rao_et_al_2015_2.pdf ·...

MID-CRETACEOUS RUDISTS (BIVALVIA:HIPPURITIDA) FROM THE LANGSHAN FORMATION,LHASA BLOCK, TIBET

by XIN RAO1,2, PETER W. SKELTON3, JINGENG SHA1, HUAWEI CAI1 andYASUHIRO IBA4

1State Key Laboratory of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 East Beijing Road,Nanjing, 210008, China; e-mails: [email protected], [email protected], [email protected] of Chinese Academy of Sciences, Beijing, 100039, China3Department of Earth, Environment and Ecosystems, The Open University, Milton Keynes, MK7 6AA, UK; e-mail: [email protected] of Natural History Sciences, Hokkaido University, N10W8, Kitaku, Sapporo, Hokkaido 060-0810, Japan; e-mail: [email protected]

Typescript received 9 March 2015; accepted in revised form 11 May 2015

Abstract: The Cretaceous Langshan Formation, which

crops out along the northern portion of the Lhasa block, is

composed mainly of grey or grey-black fine-grained lime-

stones, argillaceous limestones and a few thin-bedded

siltstones and silty mudstones. From specimens collected

and/or photographed in the field, combined with a revision

of taxa previously recorded from the Langshan Formation,

four rudist taxa are recognized: (1) Auroradiolites gen. nov.;

(2) Eoradiolites cf. davidsoni; (3) Magallanesia rutogensis sp.

nov.; and (4) Sellaea sp. We propose Auroradiolites gen.

nov. for the grouping of SW Asian to Pacific radiolitid spe-

cies formerly attributed to Eoradiolites and characterized by

a compact (non-celluloprismatic) outer shell layer. The new

genus is represented in the Langshan Formation by

A. biconvexus (Yang et al., 1982), including several other

synonymized taxa. Taxonomic confusion in the literature

has resulted from the varied appearance of the strongly

developed ligamentary infolding in such forms: in the right

valve, it has a forked inner termination, but common trans-

verse sections through both valves show the latter juxta-

posed against its broad, rounded counterpart in the left

valve, giving the combined terminations a bulb-shaped

appearance. Associated Eoradiolites cf. davidsoni is of smal-

ler size and, by contrast, shows the celluloprismatic struc-

ture of the outer shell layer. The recently described

canaliculate polyconitid genus, Magallanesia Sano et al.,

2014, is represented by M. rutogensis sp. nov., which differs

from the type species in possessing more numerous and

ventrally extended canals in the left valve. Possibly belong-

ing with the latter species are isolated right valves, similar

to Horiopleura, but containing tabulae. A probable Sellaea

sp. is represented by poorly preserved right valves neverthe-

less showing typical features, including a very thin outer

shell layer and an erect posterior myophoral wall, separating

off a broad posterior ectomyophoral cavity. Based on the

presence of Magallanesia and the orbitolinid record, this

rudist fauna is probably no older than late Albian, while a

younger age cannot be excluded. Both Magallanesia and

Auroradiolites are characteristic of a SW Asian/central

Pacific faunal province, while the other two genera are

more widely distributed.

Key words: Radiolitidae, Auroradiolites gen. nov., Polyconi-

tidae, orbitolinids, Albian, SW Asian – Pacific province.

THE elaborate morphologies of rudist bivalves (Cestariand Sartorio 1995; Skelton 2013a) enabled them tooccupy shallow marine habitats in great abundance onthe vast carbonate platforms that developed within theequatorially encircling Tethyan–Atlantic–tropical Pacificoceanic realm of the Late Jurassic and Cretaceous. There,they congregated in loose to dense clusters, often formingbroad shelly ‘meadows’ (Gili et al. 1995) or low moundssometimes preserved as laterally extensive biostromes tobioherms within carbonate platform successions.

One such area of rudist colonization was the Lhasablock, which was located on the northern margin of theeastern Tethys Ocean in mid-Cretaceous times (Changet al. 1986; Zhang 2000), and is today bounded on itsnorthern side by the Bangong–Nu Suture and on itssouthern side by the Indus–Yarlung Suture (XZBGM1993; Fig. 1). Massive mid-Cretaceous carbonate depositsare exposed in the northern portion of the Lhasa block,referred to as the Langshan Formation (Leier et al. 2007).This formation is distributed in a belt encompassing

© The Palaeontological Association doi: 10.1002/spp2.1019 1

[Papers in Palaeontology, 2015, pp. 1–24]

Rutog, Gar, Gegyai, Zhongba, Gerze, Coqen, Nyima,Xainza and Baingoin counties (XZBGM 1993; Fig. 1), andit contains abundant orbitolinids, gastropods, corals andrudists (Zhang 1982, 1986, 1991; Yu and Xia 1985; L€oserand Liao 2001; Scott et al. 2010).The rudists from the Langshan Formation have been

described in Rutog, Gegyai, Gerze and Baingoin counties.Nineteen species have been described in the previous lit-erature, assigned to 11 genera (Yang et al. 1982; Yang1984; Gou 1994; Gou and Shi 1998; Scott et al. 2010).However, in most cases, the diagnostic characters of thegenera or even families have not been well established inthese publications, so their identification remains prob-lematic and in need of revision according to recentadvances in rudist classification. Such revision is of par-ticular interest given the intermediate palaeogeographicalposition of this fauna between the classically documentedrudist faunas of the Mediterranean/Middle Eastern Teth-yan region and those of Pacific provenance, largelydescribed in the last few years (Masse and Shiba 2010;Sano and Masse 2013; Skelton et al. 2013; Sano et al.2014).The most abundant rudists from the Langshan Forma-

tion are some radiolitids that were originally assigned toPraeradiolites (Yang et al. 1982; Gou and Shi 1998), but

later to Eoradiolites and Sphaerulites (Scott et al. 2010),though possessing an entirely compact outer shell layer.Similar forms were also reported from Iran, centralAfghanistan, northern Pakistan, northern India and Japan(Montenat et al. 1982; Pudsey et al. 1985; Mathur andVogel 1988; Masse and Gallo Maresca 1997; Sano andMasse 2013; Upadhyay 2014 and see also Upadhyay 2001,fig. 4b). Yang et al. (1982) and Gou and Shi (1998) estab-lished many new species for these radiolitids based mainlyon supposed differences in the form of their ligamentaryinfolding in transverse section. But both Masse and GalloMaresca (1997) and Scott et al. (2010) questioned thereliability of such variation in ligament form as a diag-nostic criterion. Moreover, these south-west Asian andwestern Pacific radiolitids differ from typical Eoradiolites(Pons et al. 2010) not only in the structure of the outershell layer, but also in the shape of the shell and the myo-cardinal apparatus.Accordingly, these Lhasa block radiolitid species are

revised herein based on the investigation of materialnewly collected and/or photographed in the field by thesenior author, and revision of the previous publications.In addition, three further rudist species are also describedfor the first time from China and their stratigraphical andpalaeogeographical significance clarified.

F IG . 1 . Map showing the outcrop of the Langshan Formation in the Lhasa block (modified from XZBGM, 1993). BNS, Bangong–NuSuture; IYS, Indus–Yarlung Suture.

2 PAPERS IN PALAEONTOLOGY

GEOLOGICAL SETTING

The rudist bivalves described herein were collected fromthe limestone beds of the following three sections inTibet: (1) the north side of Lameila Mountain at thejunction of Rutog County and Gar County (N32°42051″, E 79°51025″; Fig. 2A); (2) the south bank ofNyer Lake in Gegyai County (N 32°14031.5″, E82°07058.9″; Fig. 2B); and (3) the south-west bank ofZabuye Lake in Zhongba County (N 31°18016″, E83°55034″; Fig. 2C).

In the tectonic division of Tibet, these areas belong tothe northern part of the Lhasa block, which is sandwichedbetween the Bangong–Nu Suture and the Indus–YarlungSuture (Fig. 1; Yin and Harrison 2000; Zhang 2000; Leieret al. 2007). Here, thick shallow marine carbonates ofmid-Cretaceous age, called the Langshan Formation, arewidely distributed. This formation is interpreted to have

been deposited in the Gangdese retroarc foreland basin(Zhang et al. 2011).

The Langshan Formation crops out widely along abelt encompassing Rutog, Gar, Gegyai, Zhongba, Gerze,Coqen, Nyima, Xainza and Baingoin counties, extendingfrom Rutog County in the west to Namtso Lake in theeast (Fig. 1; XZBGM 1993). It was deposited in a rela-tively stable carbonate sedimentary environment and ismainly composed of grey or grey-black fine-grainedlimestones, argillaceous limestones and a few thin-bed-ded siltstones and silty mudstones (Fig. 2D; Marcouxet al. 1987; Zhang et al. 2004; Leier et al. 2007). Thethickness of the Langshan Formation is poorly con-strained; we regard it as about 1000 m based on Zhanget al. (1986) and Yin et al. (1988). Based on its litholog-ical character, the Langshan Formation was separatedinto lower and upper parts, the former containinglimestones, siltstones and marlstones, and the latter

A C D

B

F IG . 2 . The fossil localities inRutog (A), Gegyai (B) and Zhongba(C) counties, and the lithologicalfeatures of the Langshan Formationand the beds yielding rudists (basedon Zhang et al. 1986) (D).

RAO ET AL . : M ID-CRETACEOUS RUDISTS FROM TIBET 3

dominated by limestones (Zhang et al. 1986). The rudistspecimens described herein were all collected from theupper part of the Langshan Formation (Fig. 2D). Thisformation conformably overlies the Lower Cretaceous si-liciclastic Duoba Formation, and is in turn unconform-ably overlain by the Upper Cretaceous coarse-grainedclastic Jingzhushan Formation (BGMRXAR 1997; Zhaoet al. 2001).Abundant marine fossils are present in the Langshan

Formation and the main elements are orbitolinids, ru-dists, gastropods and corals (Zhang 1982, 1986, 1991;Liao and Xia 1985; Yu and Xia 1985; BGMRXAR 1997;L€oser and Liao 2001; Scott et al. 2010; Deng and Wang2013).During the course of this study, RAO and IBA re-stud-

ied the orbitolinid records of Zhang (1986, 1991) in thelight of the recent evolutionary and biostratigraphicalscheme of the Eopalorbitolina–Palorbitolinoides andPraeorbitolina–Mesorbitolina lineages (Schroeder 1975;Schroeder et al. 2010) and confirmed the age assignmentof the Langshan Formation to the early Barremian toearly Cenomanian according to the presence of Eopalor-bitolina charollaisi, Palorbitolina lenticularis, Palorbitolinaultima, Praeorbitolina cormyi, Mesorbitolina parva, M. tex-ana, M. subconcava and M. aperta (see StratigraphicalCorrelation, below).

SYSTEMATIC PALAEONTOLOGY

This published work and the nomenclatural acts it contains havebeen registered in Zoobank: http://www.zoobank.org/References/A00A7D9A-58D2-403B-8836-06B790A2DE5B

All numbered specimens are conserved in the collections ofthe Nanjing Institute of Geology and Palaeontology, ChineseAcademy of Sciences, 39 East Beijing Road, Nanjing 210008,China (NIGP).

The suprageneric classification and morphological terminol-ogy used herein follow Skelton (2013a, b).

Abbreviations. am, anterior myophore; ab, anterior radial band;at, anterior tooth; ats, anterior tooth socket; bc, body cavity; ct,central tooth; cts, central tooth socket; ecto, ectomyophoral cav-ity; ib, interband; Llv, ligamentary infolding in left valve; Lrv, lig-amentary infolding in right valve; LV, left valve; ol, outer(calcitic) shell layer; pm, posterior myophore; pb, posterior radialband; pt, posterior tooth; pts, posterior tooth socket; RV, rightvalve.

Order HIPPURITIDA Newell, 1965 (nom. correct. Scarlato and

Starobogatov, 1971 pro Order Hippuritoida Newell, 1965;

emend. Bouchet et al. 2010)

Suborder HIPPURITIDINA Newell, 1965 (Skelton, 2013b)

Superfamily RADIOLITOIDEA d’Orbigny, 1847

Family RADIOLITIDAE d’Orbigny, 1847

Genus AURORADIOLITES gen. nov.

LSID. urn:lsid:zoobank.org:act:AC0FD0E6-2E5C-4249-A935-77B57B7A0AA3

Type species. Praeradiolites gilgitensis Douvill"e, 1926a; from theupper Aptian of Kohistan, northern Pakistan.

Derivation of name. From the Roman goddess of dawn, ‘Aur-ora’, contracted with Radiolites, in reference both to the orientalprovenance of the new genus, and to its early evolutionaryappearance within the radiation of the family Radiolitidae.

Included species. Auroradiolites gilgitensis (Douvill"e, 1926a);A. griesbachi (Douvill"e, 1926b); A. biconvexus (Yang et al., 1982)(and synonyms Praeradiolites ngariensis Yang et al., 1982;?P. gegyainensis Gou and Shi, 1998; ?P. daxungensis Gou andShi, 1998; ?P. bangoinensis Gou and Shi, 1998; ?P. perbellus Gouand Shi, 1998; and ?P. coquenensis Gou and Shi, 1998).

Diagnosis. Moderate- to large-sized radiolitid (adult com-missural diameters from 2 to over 10 cm). LV externallyalmost flat to strongly domed; RV subcylindrical to obtu-sely conical with thick (0.5 to several cm), entirely com-pact (non-celluloprismatic) ol, forming projecting laminargrowth rugae and ornamented externally by coarserounded to subangular costae. Radial bands weakly tomoderately protruding from posteroventral flank of RV,separated by relatively narrower, indented ib. Ligamentaryinfolding strongly developed, producing elongate inwardextension of ol with widely forked inner termination inRV, although transverse sections across the two articu-lated valves show the Lrv inner termination juxtaposedagainst its broad, rounded counterpart in the LV givingthe combined inner terminations a bulb-shaped appear-ance. LV teeth prominent, straddling elongate ridge-likect in RV, which is separated from Lrv by small accessorycavity. LV myophores robust, facing outwards onto innerwall of RV: LV pm a protruding plate attached via emb-ayed neck to pt, am somewhat less protruding.

Remarks. This new genus is proposed to embrace thedistinctive grouping of south-west Asian to Japaneseradiolitid species that are characterized by a compact(non-celluloprismatic) outer shell layer and which wereformerly attributed to Eoradiolites by Masse and GalloMaresca (1997) and Sano and Masse (2013). Despiteshowing a similar configuration of radial bands to that inEoradiolites, Auroradiolites differs from the former genusin its possession of the following characters: (1) anentirely compact fibrous prismatic, as opposed to cellu-


loprismatic ol in the RV (contrary to the erroneousdescription of polygonal cells in a specimen of A. gilgiten-sis by Skelton, in Pudsey et al. (1985), as confirmed hereby Fig. 3A–C); (2) an almost flat to strongly convex LV(e.g. Fig. 4A), contrasting with the externally concaveform of that in Eoradiolites (‘with nearly flat centre andinwards inclined margin’ in the diagnosis of the lattergenus given by Pons et al. (2010)); and (3) a more robustmyocardinal apparatus, with a much more strongly devel-oped ligamentary infolding (e.g. Figs 3, 4B–C). The com-pact outer shell layer, in particular, suggests thepossibility that this endemic north-eastern Tethyan towestern Pacific genus arose independently from the ances-tral radiolitid genus Agriopleura in the late Aptian, whilethe first celluloprismatic forms, Archaeoradiolites and

Eoradiolites, were evolving in the Mediterranean Tethyanregion at almost same time (Fenerci-Masse et al. 2006).Published photographic figures of Radiolites abraensis

Coogan, 1973, from the upper Albian of Mexico, alsoappear to show an entirely compact ol (Coogan 1973, pl. 7,fig. 1b; Alenc"aster and Garc"ıa-Barrera 2008, figs 3.4 and8.3), despite the somewhat ambiguous text descriptions ofthe ol by those authors (‘. . .composed of finely reticulatevertical and radial funnel plates’ (Coogan 1973, p. 63), and‘. . .polygonal cells of less than 0.5 mm’ (Alenc"aster andGarc"ıa-Barrera 2008, p. 579)). Moreover, the ol microstruc-ture in the closely similar R. costata Scott, 1990, isdescribed as ‘. . .prismatic rather than coarsely cellular as inmost other radiolitids’ (Scott 1990, p. 78). Nevertheless, weexclude both these Mexican species from Auroradiolites, as

A

C D

B

F IG . 3 . Auroradiolites gen. nov.; transverse sections of RV in adumbonal view plus myocardinal elements of LV. A–C, A. gilgitensisfrom Kohistan (see Pudsey et al. 1985; specimen re-photographed by PWS); A, transverse section passing through ligamentary infold-ings of both valves; B, detail from white square in (A); C, transverse section cut at a lower level in a neighbouring RV. D, A. biconvex-us (= paratype of Praeradiolites ngariensis Yang et al., 1982, copied from Scott et al. 2010, fig. 9.1). All scale bars represent 10 mm.Abbreviations: at, anterior tooth; ct, central tooth; Llv, ligamentary infolding in left valve; Lrv, ligamentary infolding in right valve; LV,left valve; pm, posterior myophore; pt, posterior tooth; RV, right valve.


they differ from the latter in the following respects: (1) theradial bands of Auroradiolites are weak to moderate protru-sions (Figs 4D, 5A), as opposed to the smooth indentedfurrows in R. abraensis and R. costata; (2) the ligamentaryinfolding of Auroradiolites is much more strongly devel-oped than that in the Mexican species.

Another New World Albian genus possessing arelatively thick, compact ol that invites comparison isPseudopetalodontia Masse et al., 2007. Yet this genus isagain distinguished from Auroradiolites by its markedlyless pronounced ligamentary infolding and externallyindented radial bands (Masse et al. 2007, fig. 6), as wellby its robust ct in the RV, which projects inwards pastthe ligamentary infolding with no intervening dorsalaccessory cavity, in our view justifying its attribution tothe Monopleuridae by Masse et al. (2007). Thus, we judgeits similarities with Auroradiolites to be convergent.At the species level, a number of similar radiolitid spe-

cies from the Langshan Formation erected by Gou andShi (1998) and attributed by them to Praeradiolites (P. ge-gyainensis, P. daxungensis, P. bangoinensis, P. perbellusand P. coquenensis) were distinguished on the basis ofapparent differences in the form of their ligamentaryinfoldings in transverse section. As noted by Scott et al.(2010), however, lack of information concerning the

developmental variability of this shell feature rendersthese distinctions open to question. Indeed, Masse andGallo Maresca (1997, fig. 2) had already pointed out intheir figured transverse sections of ‘Praeradiolites? sp.’from Afghanistan, and ‘Eoradiolites gilgitensis (Douvill"e)’,from Iran, that ‘the ligamental crest appears bulbous: afeature linked with the position of the section, close tothe upper valve with the ligamental crest inside in theanterior/dorsal part of the posterior tooth. Sections cutfurther down, away from the commissure, exhibit thetypical triangular pediculate shape of the ligamental crestof the lower valve’. This critical observation is confirmedhere in newly photographed transverse sections of ‘P. gil-gitensis Douvill"e’ from Kohistan, N. Pakistan, describedby Pudsey et al. (1985) that show both the bulbousappearance of the opposed ligamentary infoldings of bothvalves in a transverse section close to the commissure(Fig. 3A–B) and the broadly Y-shaped inner terminationof the infolding seen in a RV sectioned at a lower level(Fig. 3C). Hence, the ‘bulb-shaped’ transverse section ofthe ligamentary infolding supposed to be characteristic of‘P. ngariensis’ by Yang et al. (1982), but likewise shownonly in a section containing the opposed ligamentary in-foldings of both valves (Fig. 3D), can no longer beregarded as a diagnostic character. Synonymy of the latter

A

C D

B

F IG . 4 . Auroradiolites biconvexus specimens from Langshan Formation in Zhongba County (A) and Rutog County (B–D), Tibet. A,polished radial section of articulated shell (NIGP no. 160500), showing the myophore of LV; it is unclear whether the myophoreshown in this section is the anterior or posterior one. B, polished transverse section of two conjoined RVs (NIGP no. 160502). C, bro-ken transverse section of RV (NIGP no. 160507). D, detail from white rectangle in (C). All scale bars represent 10 mm. Abbreviations:ab, anterior radial band; bc, body cavity; ib, interband; Lrv, ligamentary infolding in right valve; LV, left valve; pb, posterior radialband; RV, right valve.


species with ‘P. biconvexus’ Yang et al. (1982), stated byits authors to have a narrow ligament ridge, is accord-ingly proposed below.

Auroradiolites biconvexus (Yang et al., 1982)

Figures 3D, 4A–D, 5, 6A (pars), B–D, 7A

1982 Praeradiolites biconvexus Yang et al., p. 299, pl. 1,

fig. 4a–c.1982 Praeradiolites ngariensis Yang et al., p. 299, pl. 1,

figs 2, 3.

1997 Praeradiolites? sp. Masse and Gallo Maresca, p. 105,

fig. 2a.

?1998 Praeradiolites gegyainensis Gou and Shi, p. 261,

pl. 1, figs 1, 2.

?1998 Praeradiolites daxungensis Gou and Shi, p. 261,

pl. 1, figs 3, 4.

?1998 Praeradiolites bangoinensis Gou and Shi, p. 263,

pl. 2, fig. 1.

?1998 Praeradiolites perbellus Gou and Shi, p. 263, pl. 2,

fig. 2.

?1998 Praeradiolites coquenensis Gou and Shi, p. 263, pl.

2, figs 3–5.2010 Eoradiolites gilgitensis (Douvill"e, 1926b); Scott et al.,

pp. 455–457, fig. 9.1–9.3 (= types of P. ngariensis

Yang et al., 1982, re-photographed; non Auroradio-

lites gilgitensis (Douvill"e, 1926a)).

2010 Sphaerulites biconvexus (Yang et al., 1982); Scott

et al., p. 457, fig. 9.4–9.7 (= holotype of P. biconv-

exus Yang et al., 1982, re-photographed, together

with copy of the latter authors’ fig. 6).

Revised diagnosis. Relatively large species of Auroradiolites(commissural diameter may exceed 100 mm) with mark-edly convex LV and radially oriented undulations on theinner margin of the ol.

Material. All specimens are preserved in section in indurated darkgrey limestone: one polished radial section of both valves (NIGPno. 160500, Fig. 4A); 11 RVs in transverse section, either polished(NIGP no. 160501, NIGP no. 160502 (2 specimens, Fig. 4B),NIGP no. 160503, NIGP no. 160504 (Fig. 6A–B)) or natural sec-tions (NIGP no. 160505 (Fig. 6C–D), NIGP no. 160506 (twomutually attached specimens, of which one is shown in Fig. 5D),NIGP no. 160507 (Fig. 4C–D), NIGP no. 160508 (Fig. 5A–C),NIGP no. 160509); plus some other RV fragments (includingNIGP no. 160510 (Fig. 5E)); together with a number of specimensphotographed in the field by the first author.

Localities. The north side of Lameila Mountain at the junctionof Rutog County and Gar County (N 32°42051″, E 79°51025″),the south bank of Nyer Lake in Gegyai County (N 32°14031.5″,E 82°07058.9″) and the south-west bank of Zabuye Lake inZhongba County (N 31°18016″, E 83°55034″).

Description. The external shell form of specimens that are unam-biguously attributable to this species shows a markedly domedLV and a RV of broadly conical form with widely flared, laminargrowth rugae (Figs 4–6, 7A). Commissural diameters may exceed100 mm (Fig. 5A). Some radiolitid specimens with flared growthrugae, but with an almost flat LV and slender conical RV, werealso photographed in the field (Fig. 7B–C), but their relativelysmall size and uncertainty as to whether their ol structure is com-pact or cellular makes their attribution to this species question-able; alternatively, they may represent another species, or – iftheir ol is celluloprismatic – they might belong to Eoradiolites,which is also described herein from the Langshan Formation.

The ol of A. biconvexus is moderately thick (up to 5 mm)over the LV (Figs 4A, 7A) and, in the RV, may extend from thejunction with the inner shell for over 40 mm to the perimeter ofthe valve in larger specimens (Fig. 5A). Externally, it is orna-mented by coarse subangular costae (Fig. 5C) that are matchedon the inner margin of the ol by radial undulations (Fig. 5D),which frequently give rise to ‘zigzag’ growth lines in transversesections of the RV (Fig. 4B). The radial bands are situated ven-trally, ab being a little wider than pb (Fig. 5A), and are separatedby a narrow, indented ib (Figs 4D, 5A). However, the relativeprotrusion of the radial bands is quite variable, such that theycannot always be clearly identified in transverse sections (e.g.Fig. 6B). The compact fibrous prismatic microstructure of the olis visible on close inspection of clean broken surfaces even tothe naked eye (Figs 4D, 5E) and lacks any signs of the cellulopr-ismatic mesostructure that characterizes most other radiolitidtaxa: that this lack of cells is not merely an effect of diageneticalteration is confirmed by the clear preservation of cellulopris-matic mesostructure in nearby specimens of Eoradiolites(Figs 6A, 8A–B).

The inner shell of the specimens inspected by us – originallyaragonitic, but now replaced by white sparry calcite – is thicklydeveloped, especially in the myocardinal area, though naturallyappearing thinner in transverse sections close to the margin ofthe RV (e.g. Fig. 6C). The ol infoldings associated with theinvaginated ligament are strongly developed, in transverse sec-tion showing either the bulbous form of the opposed infoldingsof both valves in sections close to the commissural plane(Fig. 5B and partially shown in Fig. 6D) or an asymmetricallyY-shaped form in sections at lower levels in the RV (Fig. 6B), asdiscussed above.

The two projecting LV teeth (at and pt) straddle the narrowct in the RV, which appears in transverse section as an elongateridge with a swollen inner termination and is separated from theligamentary infolding by a small accessory cavity (Figs 3D, 6C–D). The myophores are not well displayed in the materialinspected by us, but, as in the original specimens of Yang et al.(1982; e.g. Fig. 3D herein), the robust pm appears to projectmore strongly than the am, judging from its preferential appear-ance in some transverse sections (Fig. 5A–B).

Remarks. As its name suggests, the most striking charac-ter of A. biconvexus is its strongly domed LV (Yang et al.1982, pl. 1, fig. 4a–c; see also Figs 4A, 7A herein) – if thesmaller specimens with an almost flat LV shown inFig. 7B–C can be excluded from the species as suggested


above. In this respect, A. biconvexus also differs from thetype species of the genus, A. gilgitensis (Douvill"e, 1926a),in which the LV is considered to be ‘not strongly convex’(Sano and Masse 2013, p. 320). The latter species, more-over, lacks any clear indication of the radial undulationsof the ol inner margin seen in A. biconvexus (Sano andMasse 2013; see also Fig. 3A herein). In contrast, not onlyare these marginal undulations clearly shown in the holo-type of A. biconvexus (refigured by Scott et al. 2010,fig. 9, 4–7), but they are also visible in the original holo-type and paratype of ‘P. ngariensis’ of Yang et al. (1982;refigured by Scott et al. 2010, fig. 9, 1–3), which is hereregarded as a synonym of A. biconvexus. As ‘P. biconvex-us’ was established before ‘P. ngariensis’ on p. 299 ofYang et al. (1982), we designate it as the senior synonym.Auroradiolites griesbachi, which was described from

the region of Herat, in Afghanistan, is very similar toA. gilgitensis (see Sano and Masse 2013), although Scottet al. (2010) mentioned some slight differences betweenthese two species in the shapes of the central tooth and

myophores. The ligamentary infolding of specimensassigned to A. griesbachi is narrower and longer than thatof A. gilgitensis (see Montenat et al. 1982, pl. 1, fig. 4).However, similar variation is seen in A. biconvexus: forexample, a specimen with a distinctly narrow, elongate lig-amentary infolding is illustrated by Gou and Shi (1998,pl. 2, fig. 2b). As noted previously, we do not regard suchvariation in the shape of the ligamentary infolding as a reli-able diagnostic criterion for distinguishing species in Auro-radiolites. In conclusion, the specific status of A. griesbachi,as well as its age, remains uncertain. So, pending furtherstudy, the species is provisionally retained herein.The descriptions and figures of the rudist species

erected by Gou and Shi (1998) are not sufficiently infor-mative for a clear understanding of their taxonomicstatus. Nevertheless, the five new species from theLangshan Formation proposed by them and attributed toPraeradiolites are here regarded as likely junior synonymsof A. biconvexus, as tentatively indicated in our synonymylist above. Notably, the descriptions all refer to well-

A C

D E

B

F IG . 5 . Auroradiolites biconvexus specimens from Langshan Formation in Zhongba County (A–C, E) and Rutog County (D), Tibet.A, commissural surface of large RV (NIGP no. 160508). B, detail from white rectangle in (A). C, ventral flank of RV in (A), showingexternal ribbing (arrowed). D, internal margin of RV, showing radial undulations of surface (NIGP no. 160506). E, natural radial sec-tion of part of RV outer shell layer, showing flaring growth rugae and compact fibrous prismatic microstructure (NIGP no. 160510).All scale bars represent 10 mm. Abbreviations: ab, anterior radial band; at, anterior tooth; ib, interband; Llv, ligamentary infolding inleft valve; Lrv, ligamentary infolding in right valve; pm, posterior myophore; pb, posterior radial band; pt, posterior tooth.


A

C D

B

F IG . 6 . Auroradiolites biconvexus specimens from Langshan Formation in Zhongba County (A–B) and Rutog County (C–D), Tibet.A, polished cut slab showing transverse sections of RVs, in abumbonal view, of A. biconvexus (white rectangle) together with smallerEoradiolites cf. davidsoni (NIGP no. 160504). B, detail of A. biconvexus RV section from white rectangle in (A). C, natural brokentransverse section of RV in adumbonal view (NIGP no. 160505). D, detail of ligamentary infolding and dentition from white rectanglein (C). All scale bars represent 10 mm. Abbreviations: ats, anterior tooth socket; ct, central tooth; Lrv, ligamentary infolding in rightvalve; pts, posterior tooth socket.

A CB

F IG . 7 . Radiolitids in the Langshan Formation, photographed in the field by the first author. A, Auroradiolites biconvexus (ZhongbaCounty). B–C, undetermined small radiolitids with flat LV (Rutog County). All scale bars represent 10 mm. Abbreviations: at, anteriortooth; Llv, ligamentary infolding in left valve; LV, left valve; RV, right valve.


developed ligamentary infoldings, albeit with differencesin appearance in transverse section that are here regardedas diagnostically invalid, as discussed above. In addition,the descriptions of Gou and Shi (1998) variously sharementions of growth rugae and radial undulations of theol inner margin (in those cases where that shell layer ispreserved); thus, ‘Funnel plates overlapping unevenly andwavy’ (‘P. gegyainensis’, p. 261); ‘Outer ostracum consist-ing of irregular lamellar structure; wavy layers appear intransverse section’ (‘P. daxungensis’, p. 261); ‘Outer ostra-cum consisting of thin, wavy funnel plates’ (‘P. bangoin-ensis’, p. 263); ‘Funnel plates. . . appearing as slightly wavycircular structures’ (‘P. coquenensis’, p. 265). Althoughnone of the original plate figures is sufficiently clear toshow whether or not any celluloprismatic mesostructureis present in the ol, the oblique section of that in‘P. bangoinensis’ (Gou and Shi, 1998, pl. 2, fig. 1c)appears to be compact, contrary to those authors’ asser-tion of a ‘fine quadrangular network’ being present (p.263). Moreover, the myocardinal apparatus of ‘P. gegyain-ensis’ (Gou and Shi, 1998, pl. 1, figs 1b, 2b), which is theonly form in which it is adequately illustrated, is essen-tially identical to that in A. biconvexus. Meanwhile, theshape of the LV is illustrated only for ‘P. bangoinensis’(Gou and Shi, 1998, pl. 2, fig. 2a), albeit with the ol miss-ing, but is distinctly convex. Finally, mention should alsobe made of the confused account of another new species,‘Lapeirousia lomata’, given by Gou and Shi (1998,p. 265), which is stated in their text description to comefrom the ‘Upper Zongshan Formation (Campanian–Maastrichtian), Zongshan, Gamba County, Tibet’, but inthe plate figure captions, it is attributed to the LangshanFormation of the Zhilongashan area, Gegyai, Tibet. Whilethe specimens illustrated by Gou and Shi (1998, pl. 1, figs

5, 6 and pl. 2, fig. 6) clearly show marginal undulationsof the ol, it is not possible from these figures to confirmthe absence of a ligament, as asserted in the text. Conse-quently, the taxonomic status and age of this last formremains unclear. Hence, careful revisionary study of thetaxa erected by Gou and Shi (1998) is evidently necessary.In addition, Sano and Masse (2013) mentioned a possi-

ble new record of A. biconvexus from Kandahar region,Afghanistan, which was previously assigned to Praeradio-lites? sp. by Masse and Gallo Maresca (1997, fig. 2a). Therelevant specimen has a strongly convex LV as in the Tibe-tan specimens.Current stratigraphical data suggest that A. biconvexus

is most probably of late Albian age (see StratigraphicalCorrelation, below), hence somewhat younger than A. gil-gitensis, which is of late Aptian – Albian age, based onthe accompanying orbitolinids (Pudsey et al. 1985; Sanoand Masse 2013). So, based on present data, it seemsmost likely that A. biconvexus is an evolutionary descen-dant of A. gilgitensis.

Genus EORADIOLITES Douvill"e, 1909

Type species. Radiolites davidsoni Hill, 1893.

Eoradiolites cf. davidsoni (Hill, 1893)

Figures 6A (pars), 8

Material. Three cut and polished slabs of indurated grey lime-stone: one containing several RV specimens in transverse section(NIGP no. 160504, Figs 6A, 8A); one with a small RV in trans-verse section, contained within a RV of A. biconvexus (NIGP no.

A B

F IG . 8 . Eoradiolites cf. davidsoni specimens from Langshan Formation in Zhongba County (A) and Rutog County (B), Tibet. A,detail from Figure 6A (upside down) showing group of RVs in transverse section, in abumbonal view (NIGP no. 160504). B, small RVin transverse section, showing celluloprismatic mesostructure in its outer shell layer, preserved within the body cavity of a larger Auro-radiolites, with distinctly contrasting compact ol (NIGP no. 160501). Both scale bars represent 10 mm. Abbreviations: ab, anterior radialband; ib, interband; Lrv, ligamentary infolding in right valve; ol, outer (calcitic) shell layer; pb, posterior radial band.


160501, Fig. 8B); and a slab containing several RV specimensand fragments in random orientations (NIGP no. 160511).

Localities. The north side of Lameila Mountain at the junctionof Rutog County and Gar County (N 32°42051″, E 79°51025″)and the south-west bank of Zabuye Lake in Zhongba County(N 31°18016″, E 83°55034″).

Description. In transverse section, the RV is relatively small(commissural diameters generally reach around 20 mm) with arounded to subquadrate outline. A slight dorsal indentationmarks the position of the short ligamentary infolding, which isof broad triangular cross-sectional form. Weakly projecting, flat-faced radial bands are present on the posteroventral flank, abbeing slightly broader than pb, separated by a narrow, indentedib (Fig. 8A). Otherwise, much of the outer valve surface isshown in section to be ornamented by rounded ribs, 2–3 mmwide (Fig. 8A), although these may be lacking in smaller speci-mens (Fig. 8B).

The ol is mostly celluloprismatic, although the outer rim of thelayer may be compact locally, particularly around the radial bands(Fig. 8A). The cellular mesostructure is well preserved, even insmall specimens, which thus contrast distinctly with the whollycompact structure of associated specimens of A. biconcavus(Fig. 8B). The side walls of the cells have a strongly radial pattern,traversed by frequent commarginal intersections with the floorsof the cell layers, reflecting a relatively steep inward marginalinclination of the ol in the RV. The inner shell is largely missing

from the specimens we have studied, although relics of it,replaced by white to colourless calcite spar, are locally preserved.

Remarks. The size and external form of the RV, partic-ularly of the radial bands, and the inner marginal incli-nation and radially dominated cell wall pattern of theol in the Tibetan specimens are consistent with Eoradio-lites, while the ribbing on the remaining parts of theouter valve surface suggests E. davidsoni, according tothe revision of Pons et al. (2010). Eoradiolites davidsoniis an externally highly variable species, as illustrated bySteuber (1999) and Pons et al. (2010). Although rela-tively small for the species, our specimens neverthelessoverlap with the size range for specimens of E. david-soni illustrated from the Albian of Boeotia, Greece, bySteuber (1999). We therefore tentatively assign ourspecimens to the species, an alternative possibility beingthat they represent a distinct small species with wavyradial walls.

Family POLYCONITIDAE Mac Gillavry, 1937

Genus MAGALLANESIA Sano et al., 2014

Type species. Magallanesia canaliculata Sano et al., 2014.

A B

C D

F IG . 9 . Magallanesia rutogensis sp. nov., LV specimens collected from Langshan Formation (all in adumbonal view) in Rutog Coun-ty, Tibet. A, holotype LV (NIGP no. 160512). B–C, paratype LV in natural transverse section (B) and in polished section (C) a fewmm below (NIGP no. 160513). D, interpretative sketch of section shown in (B); ‘biocl.’ is a bioclastic fragment unconnected with therudist specimen. All scale bars represent 10 mm. Abbreviations: am, anterior myophore; at, anterior tooth; bc, body cavity; cts, centraltooth socket; Llv, ligamentary infolding in left valve; ol, outer (calcitic) shell layer; pm, posterior myophore; pt, posterior tooth.

RAO ET AL . : M ID -CRETACEOUS RUDISTS FROM TIBET 11

Magallanesia rutogensis sp. nov.

Figures 9, 10, 11?

LSID. urn:lsid:zoobank.org:act:CC8BE863-5D2F-41E8-BDEB-7525D721F4D5

Derivation of name. From Rutog County, in northern Tibet, inwhich the type locality for the new species is situated.

Material. All specimens are preserved in section in indurateddark grey limestone.

Holotype. LV in oblique transverse section in block (NIGP no.160512, Fig. 9A).

Paratypes. One incomplete LV in block, also shown in polishedsection (NIGP no. 160513, Fig. 9B–D) and a small rounded LVfragment displaying canals (NIGP no. 160514); plus two possibleRVs in transverse section (NIGP no. 160515 and NIGP no.160516, Fig. 11A–C). Also, several specimens photographed inthe field by the first author.

Localities. The north side of Lameila Mountain at the junctionof Rutog County and Gar County (N 32°42051″, E 79°51025″)and the south-west bank of Zabuye Lake in Zhongba County (N31°18016″, E 83°55034″).

Diagnosis. Magallanesia with a single row of numerous(usually more than 12) posteroventral pallial canals in theLV that continue for some distance around the ventralmargin beyond the base of the pm.

Description. The holotype (NIGP no.160512, Fig. 9A) is a par-tially preserved LV, of about 40 mm anteroposterior diameter,exposed in an oblique adumbonal natural transverse section.The section shows the valve to be of low capuloid shape andto possess a relatively thick (~2 mm), dark calcitic ol. The pos-terior part of the valve is well displayed, with a single row ofat least 15 narrow pallial canals that decrease in size towardsthe ventral margin, where they stop a small distance ventrallyfrom the base of the pm (Fig. 9A). The apparent extreme radialelongation of the canals in this specimen is due to the obli-quely inward orientation of the plane of section. The partiallyexposed teeth (at and pt) are likewise shown in extendedoblique section.

The slightly larger paratype LV (Fig. 9B–D) is again shown inadumbonal transverse section, as demonstrated by the decreasein diameter of the body cavity in the polished section (Fig. 9C)that was made a short distance beneath the natural sectionexposed on the surface of the block (Fig. 9B, D). The two sec-tions are relatively dorsally situated in the valve, so cuttingacross the two teeth (Fig. 9D). The anterior part of the valve aswell as most of the ol has been lost as a result of boring and/orerosion, but the single row of posteroventral pallial canals isagain well displayed. At least 21 canals of more or less ellipticalcross-sectional outline are here visible, passing in this examplearound the larger part of the ventral margin.

A number of sections of articulated shells photographed inthe field in the south-west bank of Zabuye Lake in ZhongbaCounty show the characteristic single row of canals runningradially from the umbo across the posteroventral flanks of theLV and the thickened dark ol in both valves (Fig. 10A–D) aswell as subequal teeth in the LV (Fig. 10A). In addition, thesespecimens show the pm of the RV as a robust, low ridge, separatedfrom the posterior wall of the valve by a broad ectomyophoral

A B

C D

F IG . 10 . Magallanesia rutogensissp. nov., specimens in LangshanFormation photographed in thefield by XR (Zhongba County). Allscale bars represent 10 mm. Abbre-viations: am, anterior myophore; at,anterior tooth; bc, body cavity; ct,central tooth; ecto, ectomyophoralcavity; ol, outer (calcitic) shell layer;pm, posterior myophore; pt, poster-ior tooth.


cavity (Fig. 10A), while the am forms a broad, flat shelf,opposed by its counterpart in the LV (Fig. 10B).

Lacking canals, isolated specimens of the RV are naturallyharder to identify as such, but a number of strikingly thick-shelled specimens of polyconitid affinity (Fig. 11A–C) – withsubequal sockets for the LV teeth (ats and pts; Fig. 11A–B), andagain, a thickened ol – are here considered to be likely candi-dates. Notable features of these specimens include the presenceof tabulae within at least the ats (Fig. 11B) and the bc (Fig. 11C)and paired indentations of the posteroventral margin corre-sponding to radial bands (Fig. 11A). One articulated specimenphotographed in the field (Fig. 11D), although lacking most ofthe canaliculate part of the LV in the visible section, similarlyshows subequal LV teeth and greatly thickened shell walls withbroad, platform-like am surfaces.

Remarks. With their diagnostic combination of simplepallial canals around the posteroventral flank, subequalteeth and relatively thickened ol, the LV specimensdescribed here can be attributed without doubt to therecently described canaliculate polyconitid genus Magalla-nesia Sano et al., 2014. As with the type species of thatgenus, M. canaliculata Sano et al., 2014, the subequal LVteeth and thickened ol exclude the attribution of the pres-ent material to any caprinoid or caprinulid taxa (sensuSkelton 2013a). By contrast, Sano et al. (2014) explainedhow the simple subdivision by radial laminae of the LV

ectomyophoral cavity in a polyconitid such as Praecapro-tina Yabe and Nagao, 1926, probably gave rise to Maga-llanesia. Nevertheless, the present specimens – despiteconstituting only a small sample – can be distinguishedfrom the type species by their more numerous canals,which extend relatively further along the ventral valvemargin: whereas M. canaliculata has up to 12 canals,only, situated behind the posterior myophore (Sano et al.2014), the holotype of the species described here has atleast 15 (Fig. 9A) canals and the paratype, at least 21(Fig. 9B), while specimens observed in the field (Fig. 10)appear to attain similar numbers. Moreover, the canals inthe Langshan specimens continue for some distancebeyond the base of the pm around part of the ventralmargin – so apparently representing a further advancetowards the condition seen in the plagioptychids, as pos-tulated by Sano et al. (2014). As M. canaliculata is cur-rently considered to be of late Albian age, based onaccompanying orbitolinids at the type locality in CebuIsland, in the Philippines, the age of M. rutogensis, withits slightly more derived development of canals, is thusconsidered here to be probably not older than lateAlbian.With their thickened ol, subequal sockets for the LV

teeth and prominent pm, the RV specimens illustratedin Figure 11A–C at first suggest affinity with Horiopleura

A B

C D

F IG . 11 . ?Magallanesia rutogensis sp. nov., probable specimens from Langshan Formation in Zhongba County. A–C, RV specimens;A, adumbonal view (NIGP no. 160515); B, detail of ligamentary infolding and dentition from (A); C, in presumed abumbonal view,showing remains of tabulae in body cavity (NIGP no. 160516). D, articulated specimen photographed in the field by XR; note that thissection largely misses the canals, of which just two can be seen in oblique section of the LV at top right in the photograph. All scalebars represent 10 mm. Abbreviations: am, anterior myophore; at, anterior tooth; ats, anterior tooth socket; bc, body cavity; ct, centraltooth; Lrv, ligamentary infolding in right valve; LV, left valve; ol, outer (calcitic) shell layer; pm, posterior myophore; pt, posteriortooth; pts, posterior tooth socket; RV, right valve.


Douvill"e, 1889, but tabulae are unknown in that well-studied genus. Given their co-occurrence with theundoubted specimens of Magallanesia described above,and their similar size, attribution to the new species ofthe latter genus described herein thus seems most prob-able at present. As information on the RV of the typespecies of Magallanesia was somewhat limited by thestate of preservation of the material studied by Sanoet al. (2014), the additional features noted here – thetabulae and the two indented radial bands – are of spe-cial interest. In particular, the indented radial bands,together with a ligamentary infolding (as well as tabu-lae), are shared by another, as yet unnamed form,‘Polyconitid new taxon 3’, recently described from theJapanese seamounts by Skelton et al. (2013), althoughthe latter taxon lacks canals in the LV. Nevertheless, thesimilarities noted here between these Albian polyconitidsof the NE Tethyan to central Pacific region furtherstrengthen the case for close relationships between them

and thus for an endemic radiation of polyconitids inthe region at that time, as postulated by Skelton et al.(2013).Finally, the identification of this recently recognized

canaliculate polyconitid genus in northern Tibet castsdoubt on the un-illustrated reports of some other cana-liculate rudist taxa from the Langshan and/or likelyequivalent formations by Gou and Shi (1998), whichwere already dubious on stratigraphical grounds.Such suspect identifications include those of Caprinad’Orbigny, 1822, Praecaprina Paquier, 1905 and Amphi-triscoelus Harris and Hodson, 1922 (the last two generabeing of early Aptian age), in the Chenba Formation ofthe Zedang region, and of Plagioptychus Matheron,1842 (a Late Cretaceous genus) in Langshan limestonesof the Coqen-Gerze region. Careful taxonomic revisionof those specimens is therefore needed, as with the ra-diolitids described by Gou and Shi (1998) discussedabove.

A C

D E F

B

F IG . 12 . Sellaea sp., RVs from Langshan Formation in Rutog County, Tibet. A, transverse section in abumbonal view (NIGP no.160517). B, transverse section in adumbonal view, with position of broken-away posterior valve wall indicated by dashed white lines(NIGP no. 160518). C, external mould with thin, faintly ribbed outer shell layer locally attached (NIGP no. 160519). D, posteroventralflank of valve, showing ventral sulcus (NIGP no. 160520). E, partial radial section showing apparent tabulae in unidentified accessorycavity (NIGP no. 160521). F, specimens closely associated with RV fragment of Auroradiolites (NIGP no. 160519). All scale bars repre-sent 10 mm. Abbreviations: am, anterior myophore; ats, anterior tooth socket; bc, body cavity; ct, central tooth; ecto, ectomyophoralcavity; Lrv, ligamentary infolding in right valve; ol, outer (calcitic) shell layer; pm, posterior myophore; pts, posterior tooth socket.


Family CAPRINULIDAE Yanin, 1990

Genus SELLAEA Di Stefano, 1889

Type species. Caprotina (Sellaea) zitteli Di Stefano, 1889.

Sellaea sp.

Figures 12–13

Material. All specimens from the Langshan Formation are pre-served in section in indurated pink to pale grey limestone, fromthe north side of Lameila Mountain at the junction of RutogCounty and Gar County (N 32°42051″, E 79°51025″), andinclude two RV specimens in polished transverse section, NIGPno. 160517 (Fig. 12A) and NIGP no. 160518 (Fig. 12B), a cutblock containing several fragments, NIGP no. 160519 (Fig. 12C,F), plus other loose RV fragments NIGP no. 160520, NIGP no.160521 (Fig. 12D–E). In addition, a partial articulated specimenwas photographed in the field by the first author (Fig. 13A), inthe Sangzugang Formation, which is regarded as time equivalentto the Langshan Formation (see below).

Description. The slender cylindrical RV specimens have arounded subquadrate cross-sectional form, of between 30 and40 mm anteroposterior diameter, with a dorsal indentationmarking the site of ligamentary infolding and a broad but shal-low ventral sulcus (Fig. 12A–B, D). Preservation of the shellmaterial is poor, with the thick inner shell (originally aragonitic)replaced by white to transparent cavity-filling calcite spar, andthe ol almost entirely missing. However, a relic of the ol, adher-ing to part of an external mould (Fig. 12C), shows it to havebeen very thin (<1 mm) with an external ornamentation of finemm-scale riblets.

Internally, the robust ct is flanked by apparently subequalsockets for the LV teeth (Fig. 12A). The pm forms a thick verti-cal wall connecting the ct to the posteroventral margin of thevalve and is separated from the posterior valve wall by a rela-tively broad ectomyophoral cavity (Fig. 12A–B). The am is lessdistinct, but is presumed to be represented by a slight thickeningof the anterior valve wall (Fig. 12B). No clear evidence of pallialcanals is visible in the anterior valve wall, although the poorpreservation of the material makes it impossible to decidewhether they were completely absent. A longitudinal section of aRV fragment (Fig. 12E), however, shows indeterminate tabulae-like structures in some unidentified small internal cavities.

In addition to the specimens collected from the LangshanFormation, described above, the first author also photographedin the field a natural section of a partial articulated specimen inthe Sangzugang Formation (Fig. 13A). The Aptian–Albian Sang-zugang Formation crops out in the area of Xigaze region in theLhasa block. It was deposited in the northern margin of theXigaze Forearc Basin of the Indus Arc and is regarded as timeequivalent to the Langshan Formation (Wan et al. 2003; Scottet al. 2010). This specimen provides further valuable informa-tion on the myocardinal organization of the LV and particularlyits accessory cavities, including a prominent posterior ectomy-ophoral cavity.

Remarks. The elongate cylindrical shape of the RV, withits ventral sulcus, and the thick pm wall, apparent lack ofpallial canals, at least around most of the valve walls, andthin ol are all consistent with the characters of both theprimitive caprinid Pachytraga and the caprinulid Sellaea(taken here to include the forms assigned to ‘Caprotina’by Di Stefano (1889); see Skelton (2013a)). However,both the relatively wide ectomyophoral cavity (Fig. 12A)and the apparently subequal LV teeth argue in favour ofSellaea and against Pachytraga, which is characterized by

A B

F IG . 13 . Sellaea sp. A, partial articulated shell (LV below, RV above) in natural oblique transverse section, photographed in the fieldby XR in the Sangzugang Formation at the village of Qiabulin, about 40 km west of the town of Xigaze, Tibet (note that pt is not visi-ble in this oblique section). B, comparative oblique transverse section of LV of ‘Caprotina (Sellaea) ciofaloi Di Stefano, 1889’ (adumbo-nal view), from Termini-Imerese, Sicily (Di-Stefano Collection, Museo di Paleontologia e Geologia ‘G. G. Gemmellaro’, Palermo,Sicily, specimen no. 209; figured by Di Stefano, 1889, as pl. 11, fig. 1d). Note the basic similarity of myocardinal organization to thatin (A), including accessory cavities, despite the slight difference in the section angle. Both scale bars represent 10 mm. Abbreviations:am, anterior myophore; at, anterior tooth; bc, body cavity; ct, central tooth; ecto, ectomyophoral cavity; LV, left valve; pm, posteriormyophore; pt, posterior tooth; RV, right valve.


a narrow ectomyophoral cavity in the RV and a pt that isdistinctly smaller than the at in the LV (Skelton andMasse 1998). Moreover, Pachytraga has not been recordedabove the lower Aptian (Skelton and Masse 1998),whereas the presence of a fragment of Auroradiolites inclose association with the specimens described here(Fig. 12F) indicates a higher stratigraphical level, consis-tent with the Aptian–Albian stratigraphical range of Sel-laea (Skelton and Gili 2012; PWS unpub. data). Furtherevidence in favour of the identification of Sellaea in Tibetcomes from the partial articulated specimen observed inthe Sangzugang Formation (Fig. 13A). Comparison withan oblique transverse section of a LV of S. ciafaloi (DiStefano, 1889) in the type collection from Termini Imer-ese, Sicily (Fig. 13B), reveals the same basic myocardinalorganization and arrangement of accessory cavities,including a broad posterior ectomyophoral cavity typicalof Sellaea, in contrast to Pachytraga (Skelton and Masse1998).Di Stefano (1889) erected several nominal species in

two subgenera that he assigned to the genus Caprotina(namely Caprotina ‘sensu stricto’, and Sellaea) althoughnone of his species in fact compare closely with the typespecies of the genus, Caprotina striata d’Orbigny, 1840.Di Stefano (1889) distinguished his two ‘subgenera’ onlyon the basis of possession, or not, of pallial canals out-side the RV am – a feature that could nevertheless besubject to intraspecific variation, by analogy with thatseen in Pachytraga paradoxa (Skelton and Masse 1998).Pending much needed modern revision of Di Stefano’s(1889) Sicilian material, the Tibetan specimens describedhere are thus placed without specific assignment in thegenus Sellaea, taken to include those forms attributed toCaprotina by Di Stefano (1889). In any case, the poorstate of preservation of our specimens precludes certaintyas to whether or not canals were present outside theRV am.Coralliochama anomalusa Gou and Shi, 1998, was

described from the Langshan Formation in GegyaiCounty, but only the external form of one articulatedshell and a RV were figured. Not only is the attributionto that relatively derived Campanian–Maastrichtian plagi-optychid genus (Mac Gillavry 1937) extremely doubtfulon stratigraphical grounds, but no mention was made offine pallial canals in the inner shell, which are diagnosticof Coralliochama. The external shell form of the RV issimilar to that of the Tibetan Sellaea, both having a slen-der cylindrical RV and a shallow ligamentary groove.Another important similarity is that their external sur-faces both show fine vertical riblets and circular growthlines (Fig. 12C). Although Gou and Shi (1998, p. 265)state ‘Outer ostracum 3–4 mm in thickness; inner ostra-cum 2 mm in thickness’, it is unclear how their so-calledouter and inner ostraca relate to the calcitic outer (ol),

and originally aragonitic inner shell layers of rudists rec-ognized herein. Indeed, the uppermost part of the RV ofthe articulated specimen figured by Gou and Shi (1998,pl. 2, fig. 7) appears to show a thin ol that has partlyspalled off and is quite similar in appearance to that seenin the Tibetan Sellaea sp. (Fig. 12C). As Gou and Shi(1998) did not figure a transverse section of C. anoma-lusa, the form of its myocardinal apparatus, which is cru-cial in the classification of rudists, remains obscure. Tosummarize, although C. anomalusa is externally similar toSellaea. sp., resolution of its taxonomic status must awaitfurther revision of the type material.

STRATIGRAPHICAL CORRELATION

The Langshan Formation has been dated mainly usingforaminifers (Zhang 1982, 1986, 1991), but its ageremains controversial. The most abundant foraminifers ofthis formation are orbitolinids. Zhang (1986, 1991)described and clearly illustrated the orbitolinids of theLangshan Formation from Tibet and established manynew genera and species, which we nevertheless regard asvery similar to some previously established species as wediscuss below. So revision and correct identification ofthese forms is of paramount importance for determiningthe age of the Langshan Formation.The structure and size of the embryonic apparatus is

the most important factor for generic- and species-levelclassification of orbitolinids (Schroeder 1975; Schroederet al. 2010), and several phylogenetic lineages wereestablished based on the detailed study on the develop-ment pattern of the embryonic apparatus (Schroeder1975; Schroeder et al. 2002, 2010). The first authortherefore measured the maximum diameters of embry-onic apparatuses and protoconchs on the published pho-tographs of median sections from Zhang (1986, 1991).Our revision is mainly based on these data together withthe structures of the embryonic apparatuses. Nine spe-cies have been recognized for age determination(Fig. 14), eight of them belong to the Eopalorbitolinapertenuis – Palorbitolinoides hedini lineage (Schroederet al. 2002, 2010) and Praeorbitolina cormyi – Mesorbito-lina aperta lineage (Schroeder 1975; Schroeder et al.2010), and the last one is Mesorbitolina birmanica.M. birmanica, which was characterized by a biconvexembryonic apparatus and plane convex to triangularprotoconch, was revised recently by Schlagintweit andWilmsen (2014). It has been reported from Myanmar,Iran and Tibet and probably Ladakh, India and Tunisia(Zhang 1982, 1986, 1991; Cherchi et al. 1984; Schlagintweitand Wilmsen 2014), and a possible upper Aptian toupper Albian stratigraphic range could be assumed basedon these occurrences.


Zhang (1986) described the orbitolinids from Xainzaand Baingoin counties. In this area, the Langshan Formationwas divided into lower and upper parts (Zhang et al.1986). According to our revision of the orbitolinids, thelower part contains two species: Eopalorbitolina charollaisiand Palorbitolina lenticularis which was formerlydescribed as Eorbitolina lentiformis by Zhang (1986;Table 1). On the basis of the two species recognizedherein, the lower part of the Langshan Formation shouldthus be lower Barremian to lower Aptian (Schroeder et al.2010).Zhang (1986) described 15 species from upper part of

the Langshan Formation in the same area including 10new species. We recognized five species that are useful forage-dating: Mesorbitolina parva, M. texana, M. subconcav-a, M. birmanica and M. aperta (Table 2). It should bementioned that the M. birmanica specimens from Zhang(1986) may represent the advanced form of this species,

as the size of their embryo is bigger and the structure oftheir subembryonic zone is more complicated comparedwith the Myanmar and Iran records (Schlagintweit andWilmsen 2014). In Zhang (1986), Palorbitolina lenticulariswas also reported from the upper Langshan Formation,but only one transverse section was illustrated, in whichthe structure of the embryonic apparatus was not shown,so this record was discounted. According to the speciescited above, we bracket the upper Langshan Formationwithin the upper Aptian to lower Cenomanian interval(Schroeder 1962, 1979; Schroeder and Neumann 1985;Schroeder et al. 2010).

Zhang (1991) described 25 species collected from theLangshan Formation in western Tibet (including Gerze,Gegyai, Rutog and Gar counties). Nine species arerecognized herein for the age determination: Eopalor-bitolina charollaisi, Palorbitolina lenticularis, P. ultima,Praeorbitolina cormyi, Mesorbitolina parva, M. texana,

A

C

E

D

FG

H I

B

F IG . 14 . Orbitolinid foraminifera from the Langshan Formation of Lhasa block, Tibet. A, Eopalorbitolina charollaisi (Schroeder andConrad), from Zhang (1991, pl. 2, fig. 4). B, Palorbitolina lenticularis (Blumenbach), from Zhang (1991, pl. 1, fig. 14). C, Palorbitolinaultima (Cherchi and Schroeder), from Zhang (1991, pl. 1, fig. 5). D, Praeorbitolina cormyi (Schroeder), from Zhang (1991, pl. 2, fig.7). E, Mesorbitolina parva (Douglass), from Zhang (1991, pl. 5, fig. 6). F, Mesorbitolina texana (Roemer), from Zhang (1991, pl. 2, fig.21). G, Mesorbitolina subconcava (Leymerie), from Zhang (1991, pl. 4, fig. 12). H, Mesorbitolina aperta (Erman), from Zhang (1991, pl.3, fig. 3). I, Mesorbitolina birmanica (Sahni), from Zhang (1991, pl. 4, fig. 18). All scale bars represent 0.1 mm.


M. subconcava, M. birmanica and M. aperta (Table 3).According to these species, it can be concluded that theLangshan Formation in this region can be assigned tothe lower Barremian to lower Cenomanian interval(Schroeder 1962, 1979; Schroeder and Neumann 1985;Schroeder et al. 2010).

In conclusion, we consider the Langshan Formation tohave been deposited during the early Barremian to earlyCenomanian interval. The lower part of this formation isof early Barremian to early Aptian age, while the upperpart is of late Aptian to early Cenomanian age.

The age of the rudist fauna

Eoradiolites davidsoni has a cosmopolitan distribution (inboth the Old and New worlds) and ranges stratigraphically

from the Albian to the Cenomanian – lower Turonian (Steu-ber 1999; Alenc"aster and Garc"ıa-Barrera 2008), but is mainlyreported from Albian strata in Texas, Mexico, Egypt, Iran,Hungary, Apulia and Somalia (Pons et al. 2010).Sellaea was first reported from Sicily by Di Stefano

(1889, 1898), who assigned a Cenomanian age to his speci-mens based on other rudists and nerineid gastropods fromthe type locality at Termini Imerese. However, the lattersection exposes slope deposits containing blocks of rudist-bearing limestone of different ages. In contrast, Sellaea, aswell as Himeraelites and ‘Caprotina’ sensu Di Stefano(1889), are all preserved in situ in platform limestones ofearly Aptian age (including Offneria and Glossomyophorus)exposed further east, at Cefal#u (Camoin 1983; PWS pers.obs. May 2011). Nevertheless, Sellaea has also beenreported from the Albian of southern Italy, based on mic-ropalaeontological assemblages (Chiocchini et al. 1994;Masse et al. 1998), as well as Oman (Masse et al. 1997) andSinai (Steuber and Bachmann 2002). Whether or not speci-mens from the Albian of Texas that were referred to thegenus by Coogan (1977) truly belong to the genus remainsan open question. In any case, Sellaea thus appears to havean Aptian to Albian range, on present evidence.Magallanesia was recently established by Sano et al.

(2014), who described a new species, M. canaliculata,from the Aptian–Albian of Takuyo-Daini Seamount inthe Japanese seamounts and the upper Albian of CebuIsland, Philippines. These two localities were both locatedin the central Pacific in Aptian–Albian times. The age ofthe Takuyo specimens is only roughly estimated, butgiven the relatively derived characters of M. canaliculatacompared with other Aptian–Albian polyconitids, theauthors considered that M. canaliculata was most likelyof late Albian age (Sano et al. 2014). As discussed in Sys-tematic Palaeontology above, the Tibetan M. rutogensis isyet more derived than M. canaliculata, according to thequantity and distribution of the canals, and is thus con-sidered to be probably not older than late Albian.In summary, the age of the rudist fauna here described

is most likely late Albian.

THE DISTRIBUTION AND ORIGIN OFAURORADIOLITES

Stratigraphical and geographical distribution ofAuroradiolites

Auroradiolites gilgitensis has been recorded in upper Ap-tian – Albian limestones from Iran, central Afghanistan,northern Pakistan, northern India and Japan (Montenatet al. 1982; Pudsey et al. 1985; Mathur and Vogel 1988;Masse and Gallo Maresca 1997; Sano and Masse 2013;Upadhyay 2014 and see also Upadhyay 2001, fig. 4b).

TABLE 1 . Revised orbitoline species from the lower part of theLangshan Formation, Lhasa block, Tibet (Zhang, 1986).

Species Synonyms in Zhang (1986) Age

Eopalorbitolinacharollaisi

Eopalorbitolina charollaisi(pl. 1, fig. 1)

Early Barremian

Palorbitolinalenticularis

Eorbitolina lentiformis(pl. 6, figs 2–3)

Late Barremian –early Aptian

TABLE 2 . Revised orbitoline species from the upper part ofthe Langshan Formation, Lhasa block, Tibet (Zhang, 1986).

SpeciesSynonyms inZhang (1986) Age

Mesorbitolinaparva

Orbitolina(Columnorbitolina)absidata (pl. 3,figs 13, 15–16)

Late Aptian

Mesorbitolinatexana

Orbitolina (Orbitolina)sp. (pl. 7, fig. 11)

Late Aptian –late Albian

Mesorbitolinasubconcava

Orbitolina (Mesorbitolina)texana (pl. 4,figs 7–8, 11–15)


Orbitolina (Mesorbitolina)aspera (pl. 5, figs 7–9)Orbitolina (Mesorbitolina)imperilis (pl. 6, figs 5–8)Orbitolina (Mesorbitolina)langshanensis (pl. 8, figs 1–2)Orbitolina (Mesorbitolina)xainzaensis (pl. 8, figs 6–7)

Mesorbitolinabirmanica

Orbitolina (Mesorbitolina)birmanica (pl. 5, figs 2–5)


Mesorbitolinaaperta

Orbitolina (Mesorbitolina)gigantea (pl. 7, fig. 9)

Late Albian –earlyCenomanian


Auroradiolites griesbachi, previously referred to Sphaeru-lites griesbachi, was reported from Herat, in northernAfghanistan (Montenat et al. 1982; Masse and Gallo Mar-esca 1997). The age of this species is poorly constrained,although Masse and Gallo Maresca (1997) envisaged thatit could be late Aptian.Auroradiolites biconvexus is distributed in the Lhasa

block of Tibet and Kandahar region, Afghanistan (Sanoand Masse 2013). The stratigraphical data suggest thatthis species is most probably of late Albian age (seeStratigraphical Correlation, above).To conclude, our data indicate that Auroradiolites was

restricted to those parts of south-west Asia that pertainedto the northern margin of Tethys and to the westernPacific margin. The geological age of this genus is lateAptian to Albian.

The origin of Auroradiolites

The most primitive genus of the Radiolitidae is Hauteri-vian – lower Aptian Agriopleura (Sano and Masse 2013;Skelton 2013a; Masse and Fenerci-Masse 2015), and thecompact nature of its outer shell layer makes it a naturalcandidate for the ancestry of Auroradiolites (see System-atic Palaeontology, above). Agriopleura is distributed inthe Mediterranean and Middle East regions, but question-ably also in Mexico (Masse and Fenerci-Masse 2008,2015).The Mediterranean to Middle Eastern lower Aptian spe-

cies A. libanica (Astre, 1930), recently reviewed by Masseand Fenerci-Masse (2015), is both stratigraphically andpalaeobiogeographically closest to the south-west Asian

examples of Auroradiolites. It is characterized by a concaveLV, a relatively thick, entirely compact outer shell layer,and the presence in the LV of downwardly projected my-ophoral bulges (distinct from true plates; Skelton andMasse 2000; Fenerci-Masse et al. 2006; Masse and Fenerci-Masse 2015). The differences between Auroradiolites andAgriopleura are as follows: (1) the LV of Auroradiolites isalmost flat to strongly convex, contrasting with the con-cave, lid-like LV of Agriopleura; (2) the LV myophores ofAuroradiolites are more robust, looking like protrudingplates as in other radiolitids, while the LV myophores ofAgriopleura project only as low buttresses; (3) the ligamen-tary infolding of Auroradiolites is more strongly developed,producing an elongate inward extension of the ol with awidely forked inner termination in the RV, whereas theligamentary infolding of Agriopleura is much shorter andis triangular in transverse section. In other words, themyocardinal apparatus of Auroradiolites is notably morerobust than that of Agriopleura, possibly to ensure stron-ger attachment between the two valves (see Skelton 1979).For each of the three character-state differences citedabove, the primitive condition is apparent in Agriopleura,its depressed LV (1) and relatively weak ligamentaryinfolding (3) both being retained in Archaeoradiolites andEoradiolites, while its relatively less prominent myophoresreflect its monopleurid ancestry. Based on present data,Auroradiolites is thus most parsimoniously interpreted ashaving descended directly from Agriopleura, with progres-sive convex projection of the LV and strengthening of itsmyocardinal anchorage in the RV. An alternative phyloge-netic hypothesis would be the derivation of Auroradiolitesfrom Archaeoradiolites or Eoradiolites, but that is less par-simonious as it would involve the secondary loss of the

TABLE 3 . Revised orbitoline species from the Langshan Formation in western Tibet, Lhasa block, Tibet (Zhang 1991).

Species Synonyms in Zhang (1991) Age

Eopalorbitolina charollaisi Eopalorbitolina charollaisi (pl. 2, figs 3–4) Early BarremianPalorbitolina lenticularis Palorbitolina nannembryona (pl. 1, figs 11–14) Late Barremian – early AptianPalorbitolina ultima Palorbitolina lenticularis (pl. 1, figs 4–5, 9) Early AptianPraeorbitolina cormyi Praeorbitolina cormyi (pl. 2, figs 7–9) Early AptianMesorbitolina parva Mesorbitolina lh€unzhubensis (pl. 2, figs 10–12) Late Aptian

Mesorbitolina tibetica (pl. 2, fig. 19)Mesorbitolina minuscula (pl. 3, fig. 15)Mesorbitolina parva (pl. 5, figs 6, 8–10)

Mesorbitolina texana Mesorbitolina pengboensis (pl. 2, fig. 21) Late Aptian – late AlbianMesorbitolina texana (pl. 3, figs 17, 19, 21–22)Mesorbitolina subconcava (pl. 4, figs 8, 11)Mesorbitolina confusa (pl. 5, figs 20–21)

Mesorbitolina subconcava Mesorbitolina aperta (pl. 3, fig. 6) Late Aptian – late AlbianMesorbitolina subconcava (pl. 4, figs 9–10, 12)

Mesorbitolina birmanica Mesorbitolina birmanica (pl. 4, figs 1–7, 13–18) Late Aptian – late AlbianMesorbitolina aperta Mesorbitolina aperta (pl 3, figs 1–5, 8) Late Albian – early Cenomanian

Mesorbitolina gregaria (pl. 5, figs 13–16)


celluloprismatic structure of the outer shell layer seen inthe last two genera. Nevertheless, some additional radio-litid specimens containing mixed compact and cellulopris-matic structure in the outer shell layer were collected fromthe Langshan Formation by the first author in 2014. Fromour initial analysis, however, we regard these specimens asmost probably belonging to Eoradiolites (in which it isknown that some compact structure may be locally pres-ent) and not connected with Auroradiolites. Full descrip-tion of this latest material and discussion of itssignificance will follow in a later study.In conclusion, the fossil record shows that Agriopleura

persisted in the Mediterranean to Middle Eastern regionuntil (at least) the end of the early Aptian, while Auroradio-lites spread around the northern Tethyan margin of south-west Asia to the Pacific rim from the late Aptian to the lateAlbian. The most parsimonious interpretation of the mor-phological evidence suggests that the latter genus evolveddirectly from the former, with the modification of the LVbut retention of the primitively compact outer shell layer.

PALAEOBIOGEOGRAPHICALIMPLICATIONS

Of the four rudist genera described from the LangshanFormation herein, one (Eoradiolites) is cosmopolitan, withwidespread Albian records from both the New World andthe Old World (Pons et al. 2010); two more (Auroradio-lites and Magallanesia), by contrast, are considered ende-mic to the NE Tethyan – Pacific region (Masse and GalloMaresca 1997; Sano and Masse 2013; Sano et al. 2014);finally, a fourth genus (Sellaea) presents a palaeobiogeo-graphical puzzle, having previously been unquestionablyrecorded only from central to southern Tethyan carbonateplatforms (Steuber and Bachmann 2002).As the most easterly Tethyan material referred (tenta-

tively) to the species so far reported, the Tibetanspecimens of Eoradiolites cf. davidsoni narrow the palaeo-biogeographical gap between Tethyan populations of thespecies to the west and those of the New World to theeast. On the basis of current knowledge, the Tibetan pop-ulation could have been derived from either the east (viathe Pacific) or the west according to the Albian dispersalroutes discussed by Skelton et al. (2013).

In this respect, it is interesting to note van Waasbergen’s(1995) record of radiolitid fragments with typical cellu-loprismatic ol mesostructure from the central Pacific.These samples were thought also to be Albian in age(van Waasbergen 1995; Sano and Masse 2013). At the veryleast, this record together with that of the Tibetan Eoradio-lites cf. davidsoni confirms the already cosmopolitan distri-bution of radiolitids with celluloprismatic ol in Albiantimes.

The presence of Auroradiolites in the Lhasa blockclearly associates this terrane with the south-west Asian towestern Pacific palaeobiogeographical province. Auroradi-olites is an endemic genus that was distributed along thenorthern margin of eastern Tethys and around to thewestern Pacific (Japan) during the late Aptian to Albian,according to records from Iran, central Afghanistan,northern Pakistan, northern India (Ladakh) and Japan(Montenat et al. 1982; Pudsey et al. 1985; Mathur andVogel 1988; Masse and Gallo Maresca 1997; Upadhyay2001, 2014; Sano and Masse 2013), besides those fromTibet. Masse and Gallo Maresca (1997) summarized thedistribution of late Aptian radiolitids and proposed theconcept of a south-west Asian palaeobiogeographicalprovince based on the acellular forms here placed in thenew genus Auroradiolites. Sano and Masse (2013)reported further examples attributable to the genus fromHokkaido, Japan, thereby expanding the province fromsouth-west Asia to the western Pacific. The latter authorsalso mentioned the corresponding palaeobiogeographicallinks demonstrated by the late Aptian polyconitid proba-ble sister taxa Praecaprotina yaegashii Yabe and Nagao,1926, from northern Japan, and Horiopleura haydeniDouvill"e, 1926a from Pakistan, both of which are foundwith Auroradiolites gilgitensis. Moreover, in the Lhasablock of Tibet, Auroradiolites biconvexus is accompaniedby Magallanesia rutogensis. As discussed in SystematicPalaeontology above, Magallanesia is probably the descen-dant of Praecaprotina. Thus, both the Auroradiolites andthe Horiopleura haydeni–Praecaprotina–Magallanesia lin-eages can be said to characterize the south-west Asian –western Pacific province.Given the context above, the presence also of the genus

Sellaea in the Langshan Formation, otherwise regarded ascharacteristic of the central to southern Tethyan region, isintriguing. Coogan’s (1977) attribution of some speci-mens from the Albian of Texas to Sellaea was based onhis evident confusion of borings in the shell wall outsidethe anterior myophore of the RV for pallial canals (Coo-gan 1977, pl. 3, fig. 6). However, if the Sicilian specimensthat likewise lack such canals, which were attributed byDi Stefano (1889) to Caprotina, can also be included inSellaea (see Systematic Palaeontology, above), then thepresence or absence of these canals may be taxonomicallyimmaterial, after all. Pending revision of the Texan mate-rial, the case for Sellaea in the New World thus remainsopen, hinting at the further possibility of trans-Pacificderivation of the Langshan examples, as for Eoradiolitescf. davidsoni. However, testing of that hypothesis requiresa lot more revision of relevant material, going beyond thescope of the present study.Sano et al. (2014) discussed the systematic relation-

ship of Sellaea, Praecaprotina and Magallanesia usingcladistic analysis. They concluded that Sellaea and


Magallanesia could be sister taxa, both possibly derivedfrom Praecaprotina or some allied form. If correct, thisphylogenetic hypothesis would strengthen the evidencefor a palaeobiogeographical relationship between Tibetand the Pacific. A corollary of this hypothesis would bethat Sellaea would thus originally have spread from theNW Pacific to the central and southern Tethyan plat-forms, possibly via the Lhasa block, and then situatedon the northern margin of the eastern part of theTethys. Countering this hypothesis, however, is our cur-rent knowledge of the stratigraphical distribution of Sel-laea, the earliest records of which come from the earlyAptian of the central and southern Tethyan regions, aspreviously discussed (see The age of the rudist fauna,above). Hence, further testing of the hypothesis is evi-dently required.In this context, it should also be mentioned that L€oser

and Liao (2001) revised the spatial and temporal distribu-tion of Cretaceous coral faunas from Tibet. They con-cluded that the mid-Cretaceous coral fauna from theLangshan Formation and the equivalent Jaggang Forma-tion in the Lhasa block has a closer relationship withTethyan faunas in Asia and Europe than with Caribbeanfaunas. Although this conclusion likewise still needsfuture confirmation, it again hints at an intimate palaeo-biogeographical link between the Asian/Pacific and Euro-pean regions.

CONCLUSIONS

Based on the investigation of newly collected material andspecimens photographed in the field, together with therevision of some previously recorded taxa, four rudistgenera and species are recognized from the upper part ofthe Langshan Formation in the Lhasa block:1. Auroradiolites gen. nov. is a new radiolitid genus pro-

posed for the grouping of south-west Asian to wes-tern Pacific radiolitid species that are characterizedby a entirely compact outer shell layer and whichwere formerly attributed to Eoradiolites. The TibetanAuroradiolites is attributed to A. biconvexus (Yanget al., 1982), which differs from the type speciesA. gilgitensis (Douvill"e, 1926a) by its strongly domedLV and the radially ‘wavy’ structure in the outer shelllayer.Auroradiolites most likely evolved directly from themost primitive radiolitid genus Agriopleura, whichalso has a completely compact outer shell layer andpersisted in the Mediterranean to Middle Easternregion until (at least) the end of the early Aptian,while Auroradiolites then spread around the northernTethyan margin of south-west Asia and the westernPacific from the late Aptian to the late Albian.

2. Eoradiolites cf. davidsoni (Hill, 1893) is of smaller size,with celluloprismatic structure in its outer shell layeras well as the typical ribbing of the species.

3. Magallanesia rutogensis sp. nov., a new species of thecanaliculate polyconitid, Magallanesia Sano et al.,2014, differs from the type species in possessing morenumerous and ventrally extended canals in the leftvalves.

4. Sellaea sp., represented by poorly preserved rightvalves from the Langshan Formation together witha partial articulated specimen observed in the cor-relative Sangzugang Formation, shows typical fea-tures of this genus, including a very thin outer shelllayer and an erect posterior myophoral wall in theRV, separating off a broad posterior ectomyophoralcavity.

The revision of the orbitolinids from the LangshanFormation confirms that it was deposited during theearly Barremian to early Cenomanian interval. Therudist fauna described herein from the upper part ofthe formation indicates most probably a late Albianage, based mainly on the presence of Magallanesia ruto-gensis.Both Auroradiolites biconvexus and Magallanesia ruto-

gensis are characteristic of a south-west Asian – Pacificfaunal province. Eoradiolites davidsoni and Sellaea aremore widely distributed, and their likely presence in Tibetcould imply derivation either from the East (via thePacific) or from the West, along the northern Tethyanmargin.

Acknowledgements. This work was supported by the ‘StrategicPriority Research Program (B)’ of the Chinese Academy of Sci-ences (XDB03010101) and National Science Foundation ofChina (40872013). It is also a contribution to UNESCO-IUGSIGCP Project 632. We thank Jianguo Li, Xiaolin Zhang, YaqiongWang and Bo Peng for their help during the field survey inTibet. We thank Shin-ichi Sano for his kind support and fruitfuldiscussion, and both him and Jean-Pierre Masse for their perspi-cacious and constructive reviews.

Editor. Arnaud Brayard

REFERENCES

ALENC "ASTER, G. and GARC"IA-BARRERA, P. 2008. Al-bian radiolitid rudists (Mollusca: Bivalvia) from East-CentralMexico. Geobios, 41, 571–587.

ASTRE, G. 1930. Sur un Agria du Liban. Compte Rendus del’Association Franc!aise Pour l’Avancement des Sciences, SciencesNaturelles. G"eologie et Min"eralogie, 3, 149–151.

BGMRXAR (BUREAU OF GEOLOGY AND MINERALRESOURCES OF XIZANG AUTONOMOUS REGION). 1997.Stratigraphy (Lithostratic) of Xizang Autonomous Region. ChinaUniversity of Geosciences Press, Wuhan, 302 pp. [in Chinese]


BOUCHET, P., ROCROI , J.-P., BIELER, R., CARTER, J.G. and COAN, E. V. 2010. Nomenclator of bivalve familiesand classification of bivalve families. Malacologia, 52 (2), 1–184.

CAMOIN, G. 1983. Plates-formes carbonates et r"ecifs #a ru-distes du Cr"etac"e de Sicile. Th#ese doctorale, Universit"e deProvence. Travaux du Laboratoire de G"eologie historique etde Pal"eontologie, no. 13, Universit"e de Provence, Marseille,244 pp.

CESTARI , R. and SARTORIO, D. 1995. Rudists and Faciesof the Periadriatic Domain. Agip, Milano, 207 pp.

CHANG, C., CHEN, N., COWARD, M. P. et al. 1986. Pre-liminary conclusions of the Royal Society/Academia Sinica1985 Geotraverse of Tibet. Nature, 323, 501–507.

CHERCHI , A., GUPTA, V. J. and SCHROEDER, R. 1984.Late Aptian calcareous algae and larger foraminifera fromKhalsi, Ladakh, India. Bulletin of the Indian Geological Associa-tion, 17 (2), 145–157.

CHIOCCHINI , M., FARINACCI , A., MANCINELLI , A.,MOLINARI , V. and POTETTI , M. 1994. Biostratigrafia aforaminiferi, dasicladali e calpionelle delle successioni carbona-tiche mesozoiche dell’Appennino centrale (Italia). Biostratigrafiadell’Italia Centrale, Studi Geologici Camerti, vol. spec., 128 pp.

COOGAN, A. H. 1973. New rudists from the Albian andCenomanian of Mexico and south Texas. Revista del InstitutoMexicano del Petr"oleo, 5, 51–83.-1977. Early and middle Cretaceous Hippuritacea (rudists)

of the Gulf Coast. 32–70. In BEBOUT, D. G. and LOU-CKS, R. G. (eds). Cretaceous carbonates of Texas and Mexico,applications to subsurface exploration. Report of investigations,89, Bureau of Economic Geology, University of Texas, Austin.

DENG, Z. Q. and WANG, G. X. 2013. Mesozoic Corals andSponges from Ge’gyai and Rutog, Xizang. Acta PalaeontologicaSinica, 52, 202–212. [in Chinese with English abstract]

DI STEFANO, G. 1889. Studi stratigrafici e paleontologici sulsistema Cretaceo della Sicilia. 1. Gli strati con Caprotina diTermini-Imerese. Atti della Reale Accademia di Scienze, Letteree Belle Arte di Palermo, Nuova Serie, 10, 1–44.-1898. Studi stratigrafici e paleontologici sul sistema cre-

taceo della Sicilia. I calcari con Polyconites di Termini Imerese.Paleontographia Italica, 4, 1–46.

DOUVILL"E, H. 1889. Sur quelques rudistes du terrain cr"etac"einf"erieur des Pyr"en"ees. Bulletin de la Soci"et"e G"eologique deFrance, 17 (3), 627–635.-1909. Sur le genre Eoradiolites nov. Comptes Rendus des

S"eances de la Soci"et"e G"eologique de France (S"erie 4), 9, 77.-1926a. Fossiles recueillis par Hayden dans le Kashmir en

1906 et les Pamirs en 1914; leur description. Records of theGeological Survey of India, 58, 349–357.- 1926b. Description de quelques fossiles Cr"etac"es de

l’Afghanistan. Records of the Geological Survey of India, 58,345–348.

FENERCI-MASSE, M., MASSE , J.-P., ARIAS, C. andVILAS , L. 2006. Archaeoradiolites, A new genus from theupper Aptian of the Mediterranean region and the origin ofthe rudist family Radiolitidae. Palaeontology, 49, 769–794.

GILI , E., MASSE , J.-P. and SKELTON, P. W. 1995. Rudistsas gregarious sediment-dwellers, not reef-builders, on Creta-

ceous carbonate platforms. Palaeogeography, Palaeoclimatology,Palaeoecology, 118, 245–267.

GOU, Z. H. 1994. Rudists Fossils of Cretaceous Period in Ge-gyai Region, Tibet. Tibet Geology, 12, 128–138. [in Chinesewith English abstract]-and SHI , H. 1998. Rudists (Bivalvia) from the Cretaceous

of Tibet, China, with descriptions of new species. 255–266. InJOHNSON, P. A. and HAGGART, J. W. (eds). Bivalves: aneon of evolution – paleobiological studies honoring Norman D.Newell. University of Calgary Press, Calgary, Canada, 461 pp.

HARRIS , G. D. and HODSON, F. 1922. The rudists of Trin-idad. Paleontographica Americana, 1, 119–162.

HILL , R. T. 1893. The paleontology of the Cretaceous forma-tions of Texas: the invertebrate fossils of the Caprina lime-stone beds. Proceedings of the Biological Society of Washington,8, 97–108.

LEIER, A. L., DECELLES , P. G., KAPP, P. and GEH-RELS , G. E. 2007. Lower Cretaceous strata in the Lhasa Ter-rane, Tibet, with implications for understanding the earlytectonic history of the Tibetan plateau. Journal of SedimentaryResearch, 77, 809–825.

LIAO, W. H. and XIA, J. B. 1985. Upper Jurassic and LowerCretaceous Scleractinia from Baingoin district of northernXizang (Tibet). Memoir of Nanjing Institute of Geology andPalaeontology, Academia Sinica, 21, 119–178.

L €OSER, H. and LIAO, W. H. 2001. Cretaceous corals fromTibet (China) – stratigraphic and palaeobiogeographic aspects.Journal of Asian Earth Sciences, 19, 661–667.

MAC GILLAVRY, H. J. 1937. Geology of the province of Ca-mag€uey, Cuba, with revisional studies in rudist paleontology.Geographische en Geologische Mededeelingen, Physiographisch-Geologische Reeks, 14, 168 pp.

MARCOUX, J., GIRARDEAU, J., FOURCADE, E., BAS-SOULLET, J. P., PHILIP , J., JAFFREZO, M.,XUCHANG, X. and CHENGFA, C. 1987. Geology andbiostratigraphy of the Jurassic and Lower Cretaceous series tothe north of the Lhasa Block (Tibet, China). GeodinamicaActa, 1, 313–326.

MASSE , J.-P. and FENERCI-MASSE, M. 2008. Time con-trasting palaeobiogeographies among Hauterivian-Lower Ap-tian rudist bivalves from the Mediterranean Tethys, theirclimatic control and palaeoecological implications. Palaeogeog-raphy, Palaeoclimatology, Palaeoecology, 269, 54–65.--2015. Evolution of the rudist bivalve Agriopleura

K€uhn (Radiolitidae, Hippuritida) from the Mediterraneanregion. Palaeontology, 58, 71–100.-and GALLO MARESCA, M. 1997. Late Aptian Radiolit-

idae (rudist bivalves) from the Mediterranean and SouthwestAsiatic regions: taxonomic, biostratigraphic and palaeobiogeo-graphic aspects. Palaeogeography, Palaeoclimatology, Palaeoecol-ogy, 128, 101–110.-and SHIBA, M. 2010. Praecaprotina kashimae nov. sp.

(Bivalvia, Hippuritacea) from the Daiichi–Kashima Seamount(Japan Trench). Cretaceous Research, 31, 147–153.-BORGOMANO, J. and AL MASKIRY, S. 1997. Strati-

graphic and tectonosedimentary evolution of a late Aptian–Albian carbonate margin: the northeastern Jebel Akhdar(Sultanate of Oman). Sedimentary Geology, 113, 269–280.


-GALLO MARESCA, M. and LUPERTO SINNI , E.1998. Albian Rudist faunas from Southern Italy: taxonomic,biostratigraphic and palaeobiogeographic aspects. G"eobios, 31,47–59.-BELTRAMO, J., MARTINEZ-REYES , J. and AR-

NAUD-VANNEAU, A. 2007. Revision of Albian polyconi-tid and monopleurid rudist bivalves from the New World.221–230. In SCOTT, R. W. (ed.). Cretaceous rudists and car-bonate platforms: environmental feedback. SEPM Special Publi-cation, 87.

MATHERON, P. 1842. Catalogue m"ethodique et descriptifdes corps organis"es fossiles du d"epartement des Bouches-du-Rhone et lieux circonvoisins. R"epertoire des Travaux de laSoci"et"e de Statistique de Marseille, 6, 269 pp.

MATHUR, N. S. and VOGEL, K. 1988. Some rudists fromKhalsi limestone of Indus Formation, Ladakh, Himalaya.G"eobios, 21, 693–707.

MONTENAT, C., MOULLADE, M. and PHILIP , J. 1982.Le Cr"etac"e inf"erieur #a Orbitolines et Rudistes d’Afghanistancentral. G"eologie M"editerran"eenne, 9, 109–122.

NEWELL, N. D. 1965. Classification of the Bivalvia. AmericanMuseum Novitates, 2206, 1–25.

ORBIGNY, C. d’. 1822. Notice sur quelques esp#eces nouvellesde mollusques fossiles, du D"epartement de la Charente-Inf"eri-eur. M"emoires du Mus"eum d’Histoire Naturelle, 8, 98–110.

ORBIGNY, A. d’. 1840. Note sur le genre Caprine. Revue Zoo-logique par la Soci"et"e Cuvierienne, 2 (Ann"ee 1839), 168–170.-1847. Sur les Brachiopodes ou Palliobranches (deuxi#eme

M"emoire). Comptes Rendus Hebdomadaires des S"eances del’Acad"emie des Sciences (Paris), 25, 266–269.

PAQUIER, V. 1905. Les rudistes urgoniens. II Serie inverse.M"emoire du Soci"et"e G"eologique de France, Pal"eontologie, 29,49–102.

PONS, J. M., VICENS, E., CHIKHI-AOUIMEUR, F. andABDALLAH, H. 2010. Albian Eoradiolites (Bivalvia: Radio-litidae) from Jabal Na€ımia, Gafsa region, Tunisia, with revi-sional studies on the Albian forms of the genus. Journal ofPaleontology, 84, 321–331.

PUDSEY, C. J., SCHROEDER, R. and SKELTON, P. W.1985. Cretaceous (Aptian/Albian) age for island-arc volcanics,Kohistan, N. Pakistan. 150–168. In GUPTA, V. J. (ed.). Con-tributions to Himalayan Geology, v. 3, Geology of westernHimalayas. Hindustan Publishing Corporation, Delhi.

SANO, S. and MASSE , J.-P. 2013. First record of a primitiveradiolitid rudist from Japan. Paleontological Research, 17, 317–324.-IBA, Y., SKELTON, P. W., MASSE, J.-P., AGUI-

LAR, Y. M. and KASE, T. 2014. The evolution of canalicu-late rudists in the light of a new canaliculate polyconitidrudist from the Albian of the Central Pacific. Palaeontology,57, 951–962.

SCARLATO, O. A. and STAROBOGATOV, Y. I. 1971.Osnovnye Cherty Evolyutsii i sistema Klassa Bivalvia Mor-fologiya, Sistematika i Filogeniya Mollyuskov. Trudy Zoologi-cheskogo Instituta, Akademiya Nauk SSSR, 80, 5–38. [inRussian]

SCHLAGINTWEIT , F. and WILMSEN, M. 2014. Orbitoli-nid biostratigraphy of the top Taft Formation (Lower Creta-

ceous of the Yazd Block, Central Iran). Cretaceous Research,49, 125–133.

SCHROEDER, R. 1962. Orbitolinen des Cenomans S€udwest-europas. Pal€aontolog#ısche Zeitschrift, 36, 171–202.-1975. General evolutionary trends in Orbitolinas. Revista

Espa~nola de Micropaleontolog"ıa, N"umero Especial, 117–128.-1979. Les Orbitolines de l’Aptien: d"efinition, origine et evo-

lution. G"eobios, M"emoire Sp"ecial, 3, 289–299.–––– NEUMANN, M. (eds.) 1985. Les Grands Foraminif#eresdu Cr"etac"e Moyen de la R"egion M"editerran"enne. G"eobios,M"emoire Sp"ecial, 7, 1–140.- CLAVEL, B., CHERCHI, A., BUSNARDO, R.,

CHAROLLAIS , J. and DECROUEZ, D. 2002. Lign"eesphyl"etiques d’Orbitolinid"es de l’intervalle Hauterivien sup"eri-eur – Aptien inf"erieur; leur importance stratigraphique. Revuede Pal"eobiologie, 21 (2), 853–863.-VAN BUCHEM, F. S. P., CHERCHI, A., BAGH-

BANI , D., VINCENT, B., IMMENHAUSER, A. andGRANIER, B. 2010. Revised orbitolinid biostratigraphiczonation for the Barremian-Aptian of the eastern ArabianPlate and implications for regional stratigraphic correlations.GeoArabia Special Publication, 4, 49–96.

SCOTT, R. W. 1990. Models and stratigraphy of mid-Creta-ceous reef communities, Gulf of Mexico. SEPM (Society forSedimentary Geology) Concepts in Sedimentology and Palaeon-tology, 2, 102 pp.-WAN, X., SHA, J. and WEN, S. 2010. Rudists of Tibet

and Tarim Basin, China: significance to Requieniidae phylog-eny. Journal of Paleontology, 84, 444–465.

SKELTON, P. W. 1979. Preserved ligament in a radiolitid ru-dist bivalve and its implication of mantle marginal feeding inthe group. Paleobiology, 5, 90–106.-2013a. Rudist classification for the revised Bivalvia volumes

of the ‘Treatise on Invertebrate Paleontology’. Caribbean Jour-nal of Earth Science, 45, 9–33.-2013b. Rudist classification: nomenclatural correction of

‘Suborder Radiolitidina Skelton, 2013’ to ‘Suborder Hippuri-tidina Newell, 1965’. Caribbean Journal of Earth Science, 45,34.-and GILI , E. 2012. Rudists and carbonate platforms in

the Aptian: a case study on biotic interactions with oceanchemistry and climate. Sedimentology, 59, 81–117.-and MASSE, J.-P. 1998. Revision of the Lower Cretaceous

Rudist Genera Pachytraga Paquier and Retha Cox (Bivalvia:Hippuritacea), and the origins of the Caprinidae. G"eobios,M"emoire Sp"ecial, 22, 331–370.--2000. Synoptic guide to Lower Cretaceous rudist biv-

alves of Arabia. Middle East models of Jurassic/Cretaceous sys-tems. SEPM (Society for Sedimentary Geology) SpecialPublication, 69, 89–99.-SANO, S. and MASSE , J.-P. 2013. Rudist bivalves and

the Pacific in the Late Jurassic and Early Cretaceous. Journalof the Geological Society, 170, 513–526.

STEUBER, T. 1999. Cretaceous rudists of Boeotia, centralGreece. Special Papers in Palaeontology, 61, 229 pp.-and BACHMANN, M. 2002. Upper Aptian–Albian rudist

bivalves from Northern Sinai, Egypt. Palaeontology, 45, 725–749.


UPADHYAY, R. 2001. Middle Cretaceous carbonate build-upsand volcanic seamount in the Shyok suture, Northern Ladakh,India. Current Science, 81, 695–699.-2014. Palaeogeographic significance of ‘Yasin-type’ rudist

and orbitolinid fauna of the Shyok suture zone, Saltoro Hills,northern Ladakh, India. Current Science, 106, 223–228.

WAASBERGEN, R. J. VAN 1995. Sediment facies and envi-ronments of deposition on Cretaceous Pacific carbonate plat-forms: an overview of dredged rocks from Western Pacificguyots. 471–493. In WINTERER, E. L., SAGER, W. W.,FIRTH, J. V. and SINTON, J. M. (eds). Proceedings of theOcean Drilling Program, scientific results, 143.

WAN, X. Q., WU, Y. H. and LI , G. B. 2003. Distribution ofmid-Cretaceous Orbitolinids in Xizang (Tibet) and its paleo-biogeographic implications. Acta Geologica Sinica, 77, 1–8. [inChinese with English abstract]

XZBGM (XIZANG BUREAU OF GEOLOGY ANDMINERAL RESOURCES) . 1993. Regional geology of Xiz-ang Autonomous Region. Geological Publishing House, Beijing,707 pp. [in Chinese with English summary]

YABE, H. and NAGAO, S. 1926. Praecaprotina, nov. gen.from the Lower Cretaceous of Japan. Science Reports of the To-hoku Imperial University, 2nd Series. Geology, 9, 21–24.

YANG, Z. R. 1984. Marine facies Bivalvia, Hippuritoida. 150–165. In Palaeontological plates of Northwest region, China (III)(Xinjiang). Geology Publishing House, Beijing. [in Chinese]

YANG, Z., NIE , Z., WU, S. and LIANG, D. 1982. Creta-ceous rudists from Ngari, Xizang (Tibet), AutonomousRegion, China and their geologic significance. Acta GeologicaSinica, 56, 293–301. [in Chinese with English abstract]

YANIN, B. T. 1990. Kriterii sistematiki rudistov. 57–69. InMENNER, V. V. (ed.). Sistematika i Filogeniia Bespozvonoc-hikh. Kriterii Vydeleniia Vyschikh Taksonov. Nauka, Moscow.[in Russian]

YIN, A. and HARRISON, T. M. 2000. Geologic evolution ofthe Himalayan–Tibetan Orogen. Annual Review of Earth andPlanetary Sciences, 28, 2211–2280.

YIN, J. X., XU, J. T., L IU, C. J. and LI , H. 1988. The Tibetanplateau: regional stratigraphic context and previous work.Philosophical Transactions of the Royal Society of London, SeriesA, 327, 5–52.

YU, W. and XIA, J. B. 1985. Late Jurassic–Cretaceous Nerine-ids from Northern Xizang. Acta Palaeontologica Sinica, 24,640–650. [in Chinese with English abstract]

ZHANG, B. G. 1982. Orbitolina (Foraminifera) from Xizang.51–80. In The series of the scientific expedition to theQinghai–Xizang (Tibet) Plateau, Paleontology of Xizang(Tibet), 4. Science Press, Beijing. [in Chinese with Englishabstract]-1986. Early Cretaceous Orbitolinids from Xainza and

Baingoin, Xizang. Bulletin of Nanjing Institute of Geology andPalaeontology, Academia Sinica, 10, 101–122. [in Chinese withEnglish abstract]- 1991. Cretaceous larger foraminifera orbitolines from

Ngari, Western Xizang (Tibet). 68–87. In SUN, D. L. andXU, J. T. et al. (eds). Stratigraphy and palaeontology of Perm-ian, Jurassic and Cretaceous from the Rutog region, Xizang(Tibet). Nanjing University Press, Nanjing. [in Chinese withEnglish abstract]

ZHANG, K. J. 2000. Cretaceous palaeogeography of Tibet andadjacent areas (China): tectonic implications. CretaceousResearch, 21, 23–33.

ZHANG, B. G., YANG, S. Q., SUN, D. L. and YE, C. H.1986. Cretaceous System. Bulletin of Nanjing Institute of Geol-ogy and Palaeontology, Academia Sinica, 10, 46–55. [in Chinesewith English abstract]

ZHANG, K. J., XIA , B. D., WANG, G. M., L I , Y. T. andYE, H. F. 2004. Early Cretaceous stratigraphy, depositionalenvironments, sandstone provenance, and tectonic setting ofcentral Tibet, western China. Geological Society of America,Bulletin, 116, 1202–1222.

ZHANG, Q. H., DING, L., CAI , F. L., XU, X. X.,ZHANG, L. Y., XU, Q. and WILLEMS, H. 2011. EarlyCretaceous Gangdese retroarc foreland basin evolution in theSelin Co basin, central Tibet: evidence from sedimentologyand detrital zircon geochronology. 27–44. In GLOAGUEN,R. and RATSCHBACHER, L. (eds). Growth and collapse ofthe Tibetan Plateau. Geological Society, London, SpecialPublications, 353.

ZHAO, Z. Z., LI , Y. T., YE , H. F. and ZHANG, Y. W. 2001.Strata of Qinghai–Tibet Plateau. Science Press, Beijing, 542 pp.[in Chinese with English abstract]


mid-cretaceous rudists (bivalvia: …ibayasuhiro.com/publications_files/rao_et_al_2015_2.pdf ·...

Documents