Boine snake Bavarioboa from the Oligocene/Miocene of eastern Turkey with comments on connections between European and Asiatic snake faunas ZBIGNIEW SZYNDLAR and IZZET HOŞGÖR ably two different forms), and two fragmentary dentaries of Myotis−like bats. Remains of Oligocene/Miocene terrestrial ver− tebrates are practically unknown from eastern Turkey. The only exception is the find of a few mammals and a crocodile of simi− lar age reported recently from the Ka—izman−Tuzluca Basin (Sen et al. 2011), located near the Armenian frontier, north of the Mendikdere Formation. Fossil remains of the extinct boine snake Bavarioboa, thus far known exclusively from several localities of western and central Europe, are reported for the first time outside Eu− rope. The new fossil record is from the Mendikdere Forma− tion in easternmost Turkey, dated Late Oligocene to Early Miocene. The finding provides strong evidence of links con− necting ophidian faunas of Europe and southwestern Asia in the past, and confirms the supposition that Anatolia may have had close terrestrial connections with Europe around the Oligocene/Miocene boundary. Institutional abbreviation.—AUNHL, Ankara University Natu− ral History Laboratory. Introduction Geological setting Living members of the subfamily Boinae (family Boidae) are distributed mainly in the tropical Americas, while in the Old World their range is restricted to Madagascar and some western Pacific islands. In the past, however, they dominated in Euro− pean and perhaps also Asiatic snake faunas. The genus Bavario− boa, an extinct member of the subfamily Boinae, was one of the most common and most recognizable snakes of the European Oligocene and Early Miocene. This extinct genus was first described, as Bavarioboa hermi, from the late Early Miocene (MN 4) of Petersbuch 2 in southern Germany (Szyndlar and Schleich 1993). Subsequent studies (Szyndlar and Rage 2003) identified numerous remains of Ba− varioboa in 14 fossil sites located in several European countries (mainly France and Germany) and ranging in age from the Mid− dle Oligocene (MP 25) to early Middle Miocene (MN 5); alto− gether, the genus is represented by seven species. Although it was speculated that the Asiatic continent was the most probable place of origin of Bavarioboa, this was not sup− ported by any fossil record. Here we report Bavarioboa in the Mendikdere Formation in easternmost Turkey, near the Iranian frontier. At the same time, this is the first record of the genus in the Asiatic continent. The finding provides strong evidence of similarities (hence, exchanges) between snake assemblages in− habiting western Europe and southwestern Asia around the Oligocene/Miocene boundary. The fossil remains were discovered during field work under− taken by IH in April 2010. The entire fossil material of verte− brates in the Mendikdere Formation is limited, consisting of two trunk vertebrae of Bavarioboa sp., described in detail below, two hardly identifiable trunk vertebrae of anguine lizards (prob− The study area is situated in the Hinis−Muş−Van Basin. This is an intermontane basin, 80 km long and 20 km wide, located in the eastern part of the North Anatolian Fault zone. The snake fossils were found near the village Kurucan, in the Saray district of Van province, near the Turkish−Iranian frontier (Fig. 1). One stratigraphic section located east of Kurucan was mea− sured and sampled during field work in 2010. This measured stratigraphic section is 115 m thick approximately and mostly composed of clastic lithologies. The Mendikdere Formation is located at the bottom of the section and is characterized by green, well−sorted, thick−bedded sandstone and conglomerate at the bottom and interfingering with clayey limestone, siltstone, greyish to green sandstone−shale intercalations at the top (Fig. 2). The fossil remains were deposited in a green−brown, roughly laminated siltstone, consisting of quartz, pyrite, and some glau− conite (54m–S10AT1). No reliable age estimates of the formation are available from the literature. The very shallow marine to brackish facies is gen− erally difficult to date because of the absence of planktonic foraminifera. Age correlation ranges broadly from the Late Oligocene to Early Miocene (Şenel et al. 1984). The most char− acteristic and most abundant macrofossils occurring at the base of the Mendikdere Formation from the Kurucan section are oys− ters, in particular Crassostrea cyathula (Lamarck, 1806) (IH, unpublished material), widely recognised as an important strati− graphic marker for the Oligocene (Hoşgör and Okan 2009). Their presence generally points to a Rupelian–early Chattian age, enabling at least the lower stratigraphic boundary to be de− fined. Hence, lacking any other biostratigraphic tools, the oyster faunas are used to estimate the age of the overlying deposits. Acta Palaeontol. Pol. 57 (3): 667–671, 2012 http://dx.doi.org/10.4202/app.2011.0075 668 ACTA PALAEONTOLOGICA POLONICA 57 (3), 2012 Systematic paleontology Reptilia Laurenti, 1768 Squamata Oppel, 1811 Serpentes Linnaeus, 1758 Boidae Gray, 1825 Boinae Gray, 1825 Genus Bavarioboa Szyndlar and Schleich, 1993 Type species: Bavarioboa hermi Szyndlar and Schleich, 1993 from Petersbuch 2, Germany; Early Miocene (MN 4). Bavarioboa sp. Fig. 3. Material.—AUNHL IZ100401a, anterior trunk vertebra (Fig. 3A); AUNHL IZ100401b, middle trunk vertebra (Fig. 3B). Alluvium Saray Formation (Pliocene) Mendikdere Formation (Oligocene–Miocene) Yücelendere Formation (Upper Paleocene–Eocene) Reef limestone member (Upper Paleocene) ªehittepe Formation (Lower–Middle Paleocene Mehmetalan peridotites (Upper Cretaceous) Tepedam metamorphics (Lower Paleozoic) fossil locality Fig. 1. Map of the Eastern Anatolia subbasins, and the location of snake fos− sil site (modified from Şenel et al. 1984; Bozkurt 2001; Sancay et al. 2006). Description.—The vertebra from the middle trunk portion of the column (AUNHL IZ100401b) is almost completely preserved except for slightly eroded tips of the prezygapophyseal pro− cesses and posterior surface of the condyle. The centrum length (measured as the distance between the cotyle lip and the end of the condyle) is 5.0 mm approximately; the centrum width (the width of the interzygapophyseal constriction) is 7.1 mm. In lateral view, the vertebra is slightly higher than long. The neural spine is approximately as high as long, with anterior and posterior margins vertical, occupying one half the length of the neural arch, and beginning above the zygosphenal articular fac− ets. The anterodorsal portion of the neural spine is slightly rounded. The lateral foramina are small but distinct. The para− diapophyses are subsquare in shape, somewhat higher than long anteroposteriorly, with indistinct subdivision into para− and diapophyseal portions. The subcentral ridges are prominent. The haemal keel is distinct, with its ventral margin straight. In dorsal view, the vertebra is distinctly wider than long. The notch in the posterior border of the neural arch is moderately deep. The interzygapophyseal constriction is well expressed. The neural spine is moderately thick. The zygosphene is provided with three indistinct and wide lobes. The long axis of prezygapophyseal fac− ets is weakly oblique. The prezygapophyseal articular facets are subtriangular in shape. The prezygapophyseal processes are not visible. In ventral view, the centrum is distinctly wider than long. The haemal keel looks like a biconcave lens owing to the presence of a distinct constriction, at the level of the subcentral foramina, and prominent broadenings at the anterior and posterior ends; it is tri− angular in cross section. The subcentral grooves are relatively deep. The subcentral foramina are small but distinct. The post− zygapophyseal articular facets are subtriangular in shape. The prezygapophyseal processes are weakly developed (their tips are lost). In anterior view, the zygosphene is slightly concave dor− sally. It is as wide as the cotyle, the latter being moderately flat− tened dorsoventrally. The prezygapophyses are located clearly above the floor of the neural canal and are weakly inclined. The BRIEF REPORT 669 paradiapophyses project downwards slightly beyond the cotyle lip. The paracotylar foramina are absent. In posterior view, the neural arch is weakly vaulted. The neural spine is moderately thick. The condyle is slightly flat− tened dorsoventrally. The other vertebra (AUNHL IZ100401a), as characteristic of the cervical region of the column, is distinctly higher than long, the neural arch is more vaulted in comparison with the middle trunk vertebra, the neural spine is shorter and higher, and the haemal keel is replaced by a hypapophysis. The distal por− tion of the latter structure is broken off, but the preserved base indicates that the hypapophysis was prominent. Remarks.—To some extent, the Mendikdere fossils resemble trunk vertebrae of the Erycinae (another subfamily of the family Boidae), in particular those of the living Eryx–Gongylophis complex. However, vertebral centra of the latter snakes are (usu− ally) relatively shorter, neural arches much more depressed, prezygapophyseal processes longer, and haemal keels either un− derdeveloped or (if distinct) uniform in width (ZS, unpublished observations). The fossil remains display clearly diagnostic features of the extinct genus Bavarioboa—see Szyndlar and Rage (2003) for detailed diagnoses of members of Bavarioboa and comparisons with other, extant and extinct, genera of the Boinae. More spe− cifically, the vertebrae from the Mendikdere Formation most re− semble several Late Oligocene and Miocene members of the ge− nus, in particular B. crocheti from the French Late Oligocene (MP 28). Especially, the Turkish Bavarioboa and B. crocheti share the peculiar biconcave−lens−like haemal keel, the feature not observed in other species of the genus; besides, both snakes are characterized by weakly vaulted neural arches. However, the material from the Mendikdere Formation, consisting of two vertebrae only, is neither sufficient for the identification to spe− cific level nor for the description of a new species. What is more, precise comparisons of the Turkish fossils with B. crocheti make great difficulties, considering that the latter is character− ized by a very broad spectrum of intraspecific variation in its vertebral morphology (Szyndlar and Rage 2003). Discussion Our knowledge of the extinct snakes of Asia Minor is limited. There are only four published reports of ophidian fossils from the Asiatic part of Turkey. The snake fauna from the youngest (mid−Pleistocene) locality of Emirkaya−2 (Kessler and Venczel 1993; Venczel and Sen 1994) consists of several extant (or pre− sumably extant) species. The Late Pliocene fauna of Çalta con− tains erycines (Eryx–Gongylophis complex), Naja (originally reported as Palaeonaja) as well as unidentifiable members of Scolecophidia and Colubridae (Rage and Sen 1976). The oldest locality of Bes−Konak seems most interesting, considering its age estimated as Early or Middle Miocene; unfortunately, its snake remains are badly preserved and offer little information (identified as “Colubroïde”; Paicheler et al. 1978). Members of the Boinae have never been reported from Turkey. Fig. 2. The Kurucan section and sample location. Most fossil remains of “Booidea” from Asia were referred to either erycines or pythons (Szyndlar and Rage 2003, and refer− ences therein; Head 2005; Böhme 2007). There is only one credi− ble report of a member of the Boinae from Asia, a single (but well preserved) vertebra from the late Middle or early Late Miocene of Togay (Balkay Lake; Rage and Danilov 2008). The vertebra re− sembles those of Bavarioboa but differs from them by having a short (anteroposteriorly) neural spine. Another presumed member of the Boinae was reported from the Early Oligocene of the Zaisan Basin in Kazakhstan (Chkhikvadze 1985); unfortunately, this fos− sil has never been described or illustrated. Most snakes inhabiting Europe before the Oligocene became extinct at the Eocene/Oligocene boundary. The existing fossil re− cord indicates that Early Oligocene ophidian faunas were repre− sented exclusively by several forms of very small dimensions; larger snakes, including Bavarioboa, appeared in Europe around the Middle Oligocene (Szyndlar and Rage 2003; Szyndlar et al. 2008). Where did the snakes in Europe arrive from? The most ob− vious answer was Asia, in particular its southwestern part, but thus far there has not been any fossil record in favour of this hypothesis. Interestingly, Bavarioboa disappeared from the European conti− http://dx.doi.org/10.4202/app.2011.0075 670 ACTA PALAEONTOLOGICA POLONICA 57 (3), 2012 5 mm Fig. 3. Two vertebrae of boine snake Bavarioboa sp. from the Mendikdere Formation, Kurucan, Turkey, Oligocene/Miocene. A. AUNHL IZ100401a, ante− rior trunk vertebra in right lateral view. B. AUNHL IZ100401b, middle trunk vertebra, in right lateral (B1), ventral (B2), anterior (B3), posterior (B4), and dorsal (B5) views. nent before the end of the Oligocene (MP 28), perhaps following the growing aridization, and reappeared in the late Early Miocene (MN 3–4), arriving in Europe along with a wave of modern snakes, mainly colubroids. Apart from Bavarioboa, we know a number of other ophidians that, since the beginning of the Oligo− cene, invaded the European continent, withdrew from it, and then reappeared (Szyndlar et al. 2008). Again, also in such cases, southwestern Asia seems the most suitable place for survival and/or migratory route for the taxa that became extinct in Europe. Recently published studies of Oligocene and Miocene rhino− cerotoids and associated faunas from several Turkish localities (Antoine at al. 2008; Sen at al. 2011) indicate that Anatolia had close terrestrial connections with Asia and Europe during the Late Oligocene and Early Miocene. These observations contra− dict most palaeogeographic maps issued in the last decade (e.g., Popov et al. 2004), depicting marine barriers across the hypo− thetical migratory routes linking Anatolia with Asia and Europe. The present report of the occurrence of Bavarioboa in eastern Turkey brings additional evidence of links between terrestrial faunas of Europe and southwestern Asia. Acknowledgements.—The sampling was carried out as a part of field studies by TransAtlantic Petroleum (Turkey) Corp. (Ankara, Turkey). We are grateful to the referees, Jean−Claude Rage and Martón Venczel, for their valuable suggestions. References Antoine, P.O., Karadenizli, L., Saraç, G., and Sen, S. 2008. A giant rhino− cerotoid (Mammalia, Perissodactyla) from the Late Oligocene of north− central Anatolia (Turkey). Zoological Journal of the Linnean Society 152: 581–592. Böhme, M. 2007. Herpetofauna (Anura, Squamata) and palaeoclimatic im− plications: preliminary results. In: G. Daxner−Höck (ed.), Oligocene– Miocene Vertebrates From the Valley of Lakes (Central Mongolia): Morphology, Phylogenetic and Stratigraphic Implications. Annalen des Naturhistorischen Museums in Wien 108A: 43–52. Bozkurt, E. 2001. Neotectonics of Turkey. Geodinamica Acta 14: 3–30. Chkhikvadze, V.M. 1985. Preliminary results of the study of Tertiary amphib− ians and squamate reptiles of the Zaisan Basin [in Russian]. Voprosy Gerpetologii – Shestaâ Vsesoûznaâ Gerpetologičeskaâ Konferentsiâ, Taškent, 18–20 sentâbrâ 1985, Avtoreferaty dokladov, 234–235. Akade− miâ Nauk SSSR, Taškent. Head, J.J. 2005. Snakes of the Siwalik group (Miocene of Pakistan): Sys− tematics and relationship to environmental change. Palaeontologia Electronica 8.1.18A: 33 pp. Hoşgör, I. and Okan, Y. 2009. First record of the Crassostrea cyathula (Lamarck 1806) from the Rupelian–Lower Chattian of Saribu—day− Kovancilar (NE Palu), Eastern Taurides, E Turkey. Turkish Journal of Earth Sciences 18: 15–629. Kessler, E. and Venczel, M. 1993. Quaternary vertebrate remains from Emirkaya−2, Turkey. Theoretical and Applied Karstology 6: 213–214. Paicheler, J.C., de Broin, F., Gaudant, J., Mourer−Chauviré, C., Rage, J.C., and Vergnaud−Grazzini, C. 1978. Le bassin lacustre miocène de Bes− BRIEF REPORT Konak (Anatolie−Turquie): Géologie et introduction à la paléontologie des Vertébrés. Geobios 11: 43–65. Popov, S.V., Rögl, F., Rozanov, A.Y., Steininger, F.F., Shcherba, I.G., and Kovac, M. (eds.) 2004. Lithological−paleogeographic maps of Para− tethys. 10 maps Late Eocene to Pliocene. Courier Forschungsinstitut Senckenberg 250: 1–46. Rage, J.C. and Danilov, I.G. 2008. A new Miocene snake fauna from eastern Siberia, Russia. Was the snake fauna largely homogenous in Eurasia during the Miocene? Comptes Rendus Palevol 7: 383–390. Rage, J.C. and Sen, S. 1976. Les amphibiens et les reptiles du Pliocène supérieur de Çalta (Turquie). Géologie Méditerranéenne 3: 127–134. Sancay, R.H., Bati, Z., Işik, U., Kirici, S., and Akça, N. 2006. Palynomorph, foraminifera, and calcareous nannoplankton biostratigraphy of Oligo− Miocene sediments in the Muş Basin, Eastern Anatolia, Turkey. Turk− ish Journal of Earth Sciences 15: 259–319. Sen, S., Antoine, P.O., Varol, B., Ayyildiz, T., and Sözeri, K. 2011. Giant rhinoceros Paraceratherium and other vertebrates from Oligocene and middle Miocene deposits of the Ka—izman−Tuzluca Basin, Eastern Tur− key. Naturwissenschaften 98: 407–423. 671 Şenel, M., Acarlar, M., Çakmako—lu, A., Da—er, Z., Erkanol, D., Örçen, S., Taşkiran, M.A., Ulu, U., Ünal, M.F., and Yildirim, H. 1984. Özalap (Van)−¤ran siniri arasindaki alanin jeolojisi (Başkale−K52). MTA Rapor 7623. Szyndlar, Z. and Rage, J.C. 2003. Non−Erycine Booidea from the Oligocene and Miocene of Europe. 111 pp. Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, Kraków. Szyndlar, Z. and Schleich, H.H. 1993. Description of Miocene snakes from Petersbuch 2 with comments on the lower and middle Miocene ophidian faunas of southern Germany. Stuttgarter Beiträge zur Naturkunde, Serie B 192: 1–47. Szyndlar, Z., Smith, R., and Rage, J.C. 2008. A new dwarf boa (Serpentes, Booidea, “Tropidophiidae”) from the Early Oligocene of Belgium: a case of the isolation of Western European snake faunas. Zoological Journal of the Linnean Society 152: 393–406. Venczel, M. and Sen, S. 1994. Pleistocene amphibians and reptiles from Emirkaya−2, Turkey. Herpetological Journal 4: 159–165. Zbigniew Szyndlar [szyndlar@isez.pan.krakow.pl], Institute of Systematics and Evolution of Animals of the Polish Academy of Sciences, ul. Sławkowska 17, 31−016 Kraków, Poland; Izzet Hoşgör [Izzet.Hosgor@viking−intl.com], Viking International, Akmerkez, Block B Floor 5−6, Nispetiye Caddesi, 34330 Etiler−Istanbul, Turkey. Received 8 July 2011, accepted 19 August 2011, available online 26 August 2011. http://dx.doi.org/10.4202/app.2011.0075