3

Click here to load reader

A Late Cretaceous ceratopsian dinosaur from Europe with Asian affinities

  • Upload
    david-b

  • View
    214

  • Download
    0

Embed Size (px)

Citation preview

Page 1: A Late Cretaceous ceratopsian dinosaur from Europe with Asian affinities

LETTERS

A Late Cretaceous ceratopsian dinosaur from Europewith Asian affinitiesAttila Osi1, Richard J. Butler2 & David B. Weishampel3

Ceratopsians (horned dinosaurs) represent a highly diverse andabundant radiation of non-avian dinosaurs1–5 known primarilyfrom the Cretaceous period (65–145 million years ago). This radi-ation has been considered to be geographically limited to Asia andwestern North America1–3, with only controversial remains reportedfrom other continents. Here we describe new ceratopsian cranialmaterial from the Late Cretaceous of Iharkut, Hungary6, from acoronosaurian ceratopsian, Ajkaceratops kozmai. Ajkaceratops ismost similar to ‘bagaceratopsids’ such as Bagaceratops andMagnirostris, previously known only from Late Cretaceous eastAsia3,5,7,8. The new material unambiguously demonstrates thatceratopsians occupied Late Cretaceous Europe and, when con-sidered with the recent discovery of possible leptoceratopsid teethfrom Sweden9, indicates that the clade may have reached Europe onat least two independent occasions. European Late Cretaceous dino-saur faunas have been characterized as consisting of a mix ofendemic ‘relictual’ taxa and ‘Gondwanan’ taxa, with typical Asianand North American groups largely absent10,11. Ajkaceratopsdemonstrates that this prevailing biogeographical hypothesis isoverly simplified and requires reassessment. Iharkut was part ofthe western Tethyan archipelago, a tectonically complex series ofisland chains between Africa and Europe12, and the occurrence of acoronosaurian ceratopsian in this locality may represent an earlyLate Cretaceous ‘island-hopping’ dispersal across the Tethys Ocean.

Ceratopsia Marsh, 1890Neoceratopsia Sereno, 1986Coronosauria Sereno, 1986

Ajkaceratops kozmai gen. et sp. nov.

Etymology. Ajka: the town of Ajka, which is close to the type locality;ceratops (Greek): horned face. (Intended pronunciation: oi-ka-sera-tops.) The species is named in honour of Karoly Kozma.Holotype. MTM V2009.192.1, fused premaxillae and rostral boneswith fragments of the maxillae (Fig. 1a, b); housed in the collectionsof the Hungarian Natural History Museum, Budapest, Hungary.Referred material. MTM V2009.193.1 (Fig. 1c–e), V2009.194.1,V2009.195.1, V2009.196.1: four predentary bones.Horizon and locality. Csehbanya Formation (Upper Cretaceous,Santonian13), Iharkut, Veszprem County, Bakony Mountains, Trans-danubian Range, western Hungary. Iharkut has previously yielded atypical European Late Cretaceous dinosaur assemblage of endemicrhabdodontid ornithopods and basal nodosaurid ankylosaurs6,14.Diagnosis. Differs from all other coronosaurian ceratopsians in thefollowing character combination (* indicates autapomorphies): (1)large oval accessory fenestra are present between the premaxilla andmaxilla, with the nasal excluded from its margin; (2*) the part of thepremaxilla ventral to the external naris and the accessory fenestra isdorsoventrally shallow relative to its rostrocaudal length; (3*) the

caudolateral process of the premaxilla is curved along its lengthbecoming nearly horizontal caudally; (4*) the buccal margins ofthe predentary are sharp and not bevelled.Description. The short rostral bone strongly curves ventrally to asharp point, similar to the condition in Archaeoceratops oshimai15,Protoceratops andrewsi1 and Magnirostris dodsoni8. Well-developedlateral processes of the rostral extend along more than half of theventral margin of the premaxilla. The surface of the rostral is heavilypitted, indicating high vascularization and a covering keratinousbeak.

The body of the premaxilla is shallow, and the narial fossa extendscloser to the oral margin than in other non-ceratopsid ceratopsians1.The premaxillae and rostral bones form a strongly concave, vaulted, ovalsecondary palate that is defined laterally by sharp, edentulous cuttingmargins. The caudal edge of the premaxilla and the caudodorsal processform the rostral and dorsal margins of a large, fully perforate and ovalaccessory fenestra between the premaxilla and the maxilla, as in theAsian basal coronosaurian ceratopsians Bagaceratops rozhdestvenskyi6

and Magnirostris dodsoni15. Although the nasal is not preserved, abevelled surface on the medial surface of the caudodorsal process re-presents the articular surface for this bone. The nearly horizontal ori-entation of the caudodorsal process suggests that, as in Bagaceratops andMagnirostris16, the nasal was excluded from the margin of the accessoryfenestra. An accessory fenestra between the premaxilla and the maxilla isalso present in the North American non-ceratopsid coronosaurianZuniceratops and in some basal chasmosaurine and centrosaurineceratopsids16–18. However, in these taxa the nasal forms the dorsal mar-gin of this accessory fenestra, which is proportionally much smaller andis relatively much more dorsally positioned (its ventral margin is farabove that of the external naris). This fenestra occurred in severallineages of Coronosauria and may have been secondarily lost in derivedchasmosaurines and centrosaurines16.

The four predentaries represent different ontogenetic stages ofAjkaceratops (Supplementary Fig. 2). The rostral ends of the preden-taries are acutely pointed. The largest specimen is more stronglycurved in lateral view than smaller specimens, with a stronglyupturned rostral tip. Unusually, the edges of the predentary aresharp, and not bevelled as occurs in many ceratopsians. Caudally,the predentary has a flattened and very broad, parallel-sided ventralprocess. The surfaces of the predentaries are heavily ornamented bypits and grooves, similar to the surface of the rostral.

Three unambiguous synapomorphies demonstrate the ceratopsianaffinities of Ajkaceratops: the presence of a neomorphic rostral bone(unique to ceratopsians); the presence of a strongly vaulted premaxillarypalate; and the presence of a ventral process of the predentary that ismore than half of the transverse width of the predentary1,3,5. InCeratopsia, the presence of a strongly pointed rostral bone with well-developed lateral processes supports referral to Neoceratopsia3. The

1Hungarian Academy of Sciences, Hungarian Natural History Museum, Research Group for Paleontology, Ludovika ter 2, Budapest 1083, Hungary. 2Bayerische Staatssammlung furPalaontologie und Geologie, Richard-Wagner-Straße 10, Munich 80333, Germany. 3Center for Functional Anatomy and Evolution, Johns Hopkins University, Baltimore, Maryland21205, USA.

Vol 465 | 27 May 2010 | doi:10.1038/nature09019

466Macmillan Publishers Limited. All rights reserved©2010

Page 2: A Late Cretaceous ceratopsian dinosaur from Europe with Asian affinities

fully perforate accessory opening between the premaxilla and maxilla ispresent in a number of Late Cretaceous coronosaurians, such as theAsian Bagaceratops (including several bagaceratopsids that are probablysynonymous with this genus3,5,7) and Magnirostris8 (possibly synony-mous with Bagaceratops7), as well as the North American Zuniceratops18

and basal ceratopsids16,17. This opening is absent in earlier and morebasal ceratopsians (including non-coronosaurians and leptocera-topsids)1,3,4. The proportionally large size of the accessory openingand the exclusion of the nasal from its margin are most similar to thebagaceratopsids Bagaceratops and Magnirostris (some recent workquestions whether bagaceratopsids form a distinct clade as generallyhypothesized16). These similarities suggest that Ajkaceratops is acoronosaurian, close to bagaceratopsids and basal to theZuniceratops 1 Ceratopsidae clade.

Evidence for Gondwanan ceratopsians19,20 is highly controversial andmost reviews have assumed that ceratopsians were restricted to centraland east Asia and western North America1,3. Two recent papers reportputative ceratopsians on the basis of scarce isolated teeth from LateCretaceous northern Europe9,21. The identification of ceratopsians basedon isolated teeth must be treated with caution because ornithischiandinosaur dentition is well-documented as highly homoplastic, withparticularly striking evolutionary convergence observed between cera-topsian and ornithopod ornithischians2,21–23. Only a few dental charac-ters support the probable ceratopsian affinities of several teeth from theCampanian of Sweden (see Supplementary Information). These char-acters indicate referral of this material to Leptoceratopsidae9, althoughnon-dental remains are required to establish this identity with certainty.The historic tooth taxon Craspedodon from the Late Cretaceous ofBelgium was recently reinterpreted as possibly ceratopsian, but mayinstead represent a hadrosauroid ornithopod21. The comparably muchbetter sampled Late Cretaceous terrestrial sites of southern Europe havepreviously failed to yield ceratopsian material11. The presence ofAjkaceratops kozmai in the island chains of southern Europe and theprobable presence of a phylogenetically distinct lineage of leptoceratop-sids on the northern Europe craton indicates at least two independentoccupations of the continent by horned dinosaurs.

The largest specimens of Ajkaceratops are considerably smaller(body length estimated at approximately 1 m) than other closely

related and near contemporaneous coronosaurians such asProtoceratops and the largest Asian bagaceratopsids. Although severalfeatures are potentially consistent with some degree of osteologicalmaturity, ontogenetic stage cannot be assessed with certainty at pre-sent. Further discoveries are thus required to determine whetherAjkaceratops is a dwarfed taxon, as has been proposed for other LateCretaceous dinosaurs from the western Tethyan archipelago24–27.

Ajkaceratops seems to be part of a Late Cretaceous radiation ofcoronosaurian ceratopsians otherwise known only from Asia andNorth America. The historical biogeography of coronosaurian cer-atopsians is complex, with several inferred dispersal events between

r

en

a b

c d e

rlp

acf

1 cm

rem

vpp

mp

smpvp

vp

asn

Figure 1 | Anatomy of Ajkaceratops kozmai gen. et sp. nov. a, b, HolotypeMTM V2009.192.1, fused rostral and premaxillae in lateral (a) and ventral(b) views. c–e, Referred material MTM V2009.193.1, predentary in lateral(c), ventral (d) and dorsal (e) views. acf, inferred position of accessoryfenestra between the premaxilla and maxilla; asn, articular surface for nasal;

em, edentulous margin of premaxilla; en, external naris; mp, fragments ofrostral processes of maxillae; r, rostral bone; rlp, lateral process of rostral;smp, sharp margin of predentary; vp, ventral process of predentary; vpp,vaulted premaxillary palate.

a

c

5 cm

b

Figure 2 | Late Cretaceous palaeogeography and bagaceratopsid skulloutlines. a, b, Skull outlines, to scale, for Ajkaceratops kozmai (a) (preservedbones in black, remainder of skull based upon Magnirostris) and the Asianbagaceratopsid Magnirostris (b). Although the exact phylogenetic positionof Ajkaceratops is unclear it is probably closely related to the Asianbagaceratopsids Magnirostris and Bagaceratops. c, Late Cretaceouspalaeogeographical map showing locations of Ajkaceratops and Asianbagaceratopsids (map courtesy of R. Blakey).

NATURE | Vol 465 | 27 May 2010 LETTERS

467Macmillan Publishers Limited. All rights reserved©2010

Page 3: A Late Cretaceous ceratopsian dinosaur from Europe with Asian affinities

western North America and Asia3. However, the similarities betweenAjkaceratops and the Asian bagaceratopsids Bagaceratops andMagnirostris indicate that the presence of a ceratopsian in the westernTethyan archipelago probably results from an early Late Cretaceousdispersal event from Asia, possibly by island-hopping across theTethys Ocean (Fig. 2). This dispersal would have been facilitatedby subduction of the Tethys Ocean along the southern Asiatic marginwith resultant insular island arches in this region12. Previous biogeo-graphical hypotheses have suggested isolation of European vertebratefaunas from Asia and North America during the Early Cretaceous,followed by the development during the Late Cretaceous of anendemic and supposedly relictual dinosaur fauna (including rhabdo-dontid ornithopods and basal nodosaurid ankylosaurs) influencedby immigration of Gondwanan forms such as bothremydid turtles,titanosaurian sauropods and abelisauroid theropods10,11,28. By contrast,the discovery of Ajkaceratops kozmai indicates that the European islandfaunas of the western Tethys comprised not only endemic andGondwanan forms, but also taxa typical of the Late Cretaceous ofAsia and North America. An emerging and unexpected pattern ofEuropean/Asian faunal links in the Late Cretaceous may also be sup-ported by the possible leptoceratopsids from Sweden9 and by recentdescriptions of European hadrosauroids and hadrosaurids26,29,30.Ajkaceratops kozmai thus adds new complexity to our understandingof Late Cretaceous dinosaur faunas and demonstrates the need for re-evaluation of current biogeographical hypotheses.

Received 17 November 2009; accepted 17 March 2010.

1. You, H. & Dodson, P. in The Dinosauria (eds Weishampel, D. B., Dodson, P. &Osmolska, H.) 478–493 (Univ. California Press, 2004).

2. Dodson, P., Forster, C. A. & Sampson, S. D. in The Dinosauria (eds Weishampel, D.B., Dodson, P. & Osmolska, H.) 494–513 (Univ. California Press, 2004).

3. Sereno, P. C. in The Age of Dinosaurs in Russia and Mongolia (eds Benton, M. J.,Shishkin, M. A., Unwin, D. M. & Kurochkin, E. N.) 480–516 (Cambridge Univ.Press, 2000).

4. Xu, X., Makovicky, P. J., Wang, X.-L., Norell, M. A. & You, H.-L. A ceratopsiandinosaur from China and the early evolution of Ceratopsia. Nature 416, 314–317(2002).

5. Makovicky, P. J. & Norell, M. A. Yamaceratops dorngobiensis, a new primitiveceratopsian (Dinosauria: Ornithischia) from the Cretaceous of Mongolia. Am.Mus. Nov. 3530, 1–42 (2006).

6. Osi, A. The first dinosaur remains from the Upper Cretaceous of Hungary(Csehbanya Formation, Bakony Mts). Geobios 37, 749–753 (2004).

7. Maryanska, T. & Osmolska, H. Protoceratopsidae (Dinosauria) of Asia.Palaeontologia Polonica 33, 133–182 (1975).

8. You, H. & Dong, Z. A new protoceratopsid (Dinosauria: Neoceratopsia) from theLate Cretaceous of Inner Mongolia, China. Acta Geol. Sin. 77, 299–303 (2003).

9. Lindgren, J. et al. The first neoceratopsian dinosaur remains from Europe.Palaeontology 50, 929–937 (2007).

10. Holtz, T. R., Chapman, R. E. & Lamanna, M. C. in The Dinosauria (eds Weishampel,D. B., Dodson, P. & Osmolska, H.) 627–642 (Univ. California Press, 2004).

11. Pereda-Suberbiola, X. Biogeographical affinities of Late Cretaceous continentaltetrapods of Europe: a review. Bull. Soc. Geol. Fr. 180, 57–71 (2009).

12. Csontos, L. & Voros, A. Mesozoic plate tectonic reconstruction of the Carpathianregion. Palaeogeogr. Palaeoclimatol. Palaeoecol. 210, 1–56 (2004).

13. Knauer, J. & Siegl-Farkas, A. Palynostratigraphic Position of the Senonian BedsOverlying the Upper Cretaceous Bauxite Formations of the Bakony Mts in the AnnualReport of the Hungarian Geological Institute of 1990, 463–471 (1992).

14. Osi, A. Hungarosaurus tormai, a new ankylosaur (Dinosauria) from the UpperCretaceous of Hungary. J. Vertebr. Paleontol. 25, 370–383 (2005).

15. You, H.-L. & Dodson, P. Redescription of neoceratopsian dinosaurArchaeoceratops and early evolution of Neoceratopsia. Acta Palaeontol. Pol. 48,261–272 (2003).

16. Kirkland, J. I. & DeBlieux, D. D. in New Perspectives on Horned Dinosaurs (eds Ryan,M. Chinnery-Allgeier, B. & Eberth, D. A.) (Indiana Univ. Press, in the press).

17. Kirkland, J. I. & DeBlieux, D. D. In Ceratopsian Symposium: Short Papers, Abstractsand Programs (ed. Braman, D. R.) 90–95 (Royal Tyrrell Museum Palaeontol.,2007).

18. Wolfe, D. G. et al. In Ceratopsian Symposium: Short Papers, Abstracts and Programs(ed. Braman, D. R.) 159–167 (Royal Tyrrell Museum Palaeontol., 2007).

19. Bonaparte, J. F. Cretaceous tetrapods of Argentina. Munch. Geowiss. Abh. A 30,73–130 (1996).

20. Rich, T. H. & Vickers-Rich, P. Protoceratopsian? ulnae from Australia Records of theQueen Victoria Museum Launceston; 113, 1–12 (2003).

21. Godefroit, P. & Lambert, O. A re-appraisal of Craspedodon lonzeensis Dollo, 1883from the Upper Cretaceous of Belgium: the first record of a neoceratopsiandinosaur in Europe? Bull. Instit. Roy. Sci. Nat. Belg. Sci. Terre 77, 83–93 (2007).

22. Horner, J. R., Weishampel, D. B. & Forster, C. A. in The Dinosauria (edsWeishampel, D. B., Dodson, P. & Osmolska, H.) 438–463 (Univ. California Press,2004).

23. Sampson, S. D. & Forster, C. A. Parallel evolution in hadrosaurid and ceratopsiandinosaurs. J. Vert. Pal. 21 (suppl.), 96A (2001).

24. Weishampel, D. B., Norman, D. B. & Grigorescu, D. Telmatosaurus transsylvanicusfrom the Late Cretaceous of Romania: the most basal hadrosaurid dinosaur.Palaeontology 36, 361–385 (1993).

25. Weishampel, D. B., Jianu, C.-M., Csiki, Z. & Norman, D. B. Osteology andphylogeny of Zalmoxes (N. G.), an unusual euornithopod dinosaur from the latestCretaceous of Romania. J. Syst. Palaeontol. 1, 65–123 (2003).

26. Dalla Vecchia, F. M. Tethyshadros insularis, a new hadrosauroid dinosaur(Ornithischia) from the Upper Cretaceous of Italy. J. Vertebr. Paleontol. 29,1100–1116 (2009).

27. Benton, M. J. et al. Dinosaurs and the island rule: the dwarfed dinosaurs fromHateg Island. Palaeogeogr. Palaeoclimatol. Palaeoecol. (in the press).

28. Le Loeuff, J. The Campano-Maastrichtian vertebrate faunas from southernEurope and their relationships with other faunas in the world:palaeobiogeographical implications. Cretac. Res. 12, 93–114 (1991).

29. Prieto-Marquez, A. & Wagner, J. R. Pararhabdodon isonensis and Tsintaosaurusspinorhinus: a new clade of lambeosaurine hadrosaurids from Eurasia. Cretac. Res.30, 1238–1246 (2009).

30. Pereda-Suberbiola, X. et al. The last hadrosaurid dinosaurs of Europe: a newlambeosaurine from the Uppermost Cretaceous of Aren (Huesca, Spain). C. R.Palevol. 8, 559–572 (2009).

Supplementary Information is linked to the online version of the paper atwww.nature.com/nature.

Acknowledgements We thank L. Kocsis, P. Barrett, D. Norman, P. Dodson,J. Kirkland, P. Makovicky, S. Sampson, P. Godefroit, Z. Csiki, J. Palfy and R. Sissonsfor discussion, R. Blakey for the use of the palaeogeographical reconstruction,P. Gulyas for preparation, J. Kirkland for unpublished data and U. Gohlich forlogistical support. Research was supported by the Hungarian Scientific ResearchFund (OTKA PD 73021) and the Hantken Foundation (A.O.), and a HumboldtPostdoctoral Fellowship (R.J.B.). This is MTA–MTM Paleo contribution 109.

Author Contributions A.O. collected the material. A.O. and R.J.B. carried out theanatomical description. A.O. prepared the figures. All authors discussed the resultsand contributed to the manuscript.

Author Information Reprints and permissions information is available atwww.nature.com/reprints. The authors declare no competing financial interests.Correspondence and requests for materials should be addressed to A.O.([email protected]).

LETTERS NATURE | Vol 465 | 27 May 2010

468Macmillan Publishers Limited. All rights reserved©2010