Bachitherium sp.

Métais, Grégoire, Coster, Pauline, Licht, Alexis, Ocakoğlu, Faruk & Beard, K. Christopher, 2023, Additions to the late Eocene Süngülü mammal fauna in Easternmost Anatolia and the Eocene-Oligocene transition at the periphery of Balkanatolia, Comptes Rendus Palevol 22 (35), pp. 711-727 : 717

cf. Bachitherium sp.

( Fig. 2 R-S)

REFERRED MATERIAL. — Sü-2017 , lingual half of a right molar (M1 or M2) ; Sü-2019 , talonid of a left p4.

DIAGNOSIS (from Métais & Vislobokova 2007). — Elongated muzzle with small orbits placed relatively posteriorly (the anterior border of the orbit is situated above M2), postorbital bar partly formed by the jugal apophysis, sagittal crest developed and extending posteriorly into temporal crests, small auditory bullae, lacrimal fossa absent, strong paraoccipital apophysis, limited postero-lateral exposure of the mastoid, ethmoidal fissure moderately developed, post-glenoid apophysis well-developed; the angular region of the mandible is extended posteriorly and upward, and the coronoid apophysis is much higher than the articular joint suggesting strong abductor muscles. Dental Formula: 0/3 1/1 3/4 3/3, upper canine tusk-like, slightly curved posteriorly, and occludes against the anterior side of the caniniform p1, P1 lost, very long diastema between C and P2, lower incisors as the incisiform canine are small, small diastema between c and p1, and long diastema between p1 and p2. Radius and ulna separate, tibia and fibula partially fused, distal extremity reduced to a malleolar bone, Mc III and IV unfused, Mt III and IV fused proximally, lateral metatarsals absent, astragalus with unparallel trochlea, crural index close to that of the extant genus Moschus .

LOCALITY. — Outcrop of unnamed rock unit consisting of c. 40 cm thick tuffite bed containing gastropod operculae superposed by c. 30 cm thick white silty limestone with silicified nodules exposed in a streambed roughly two km northwest of Süngülü, Ardahan Province, Turkey (de Bruijn et al. 2003: fig. 1). Latest Eocene according to de Bruijn et al. (2018, 2019).

DESCRIPTION

The lingual half of the molar preserves highly worn protocone and metaconule; there is no lingual cingulum or entostyle, and there is both a mesial and a distal cingulum; the enamel is relatively thick and smooth. The talonid of p4 which is tentatively referred to the same taxon as the fragmentary upper molar (mostly on the basis of large size) displays in occlusal view a pattern of crests that sets it apart from tragulids and lophiomerycids. The top of the protoconid is preserved and the transversely wide talonid is delimited by two crests extending posteriorly; the labial postprotocristid reaching the posterior margin, where it forms the posterior border of the tooth; the lingual postprotocristid seems to be interrupted at the level of a marked transverse crest which is connected to the labial postprotocristid but without totally reaching the lingual border of the tooth. The disto-lingual border of the tooth is badly damaged but it does not appear that a lingual opening existed. Posterior to the protoconid, there are enamel foldings.

TAXONOMIC ATTRIBUTION

The lack of a lingual cingulum on the upper molar Sü-2017 is a character seen in Bachitherium ( Geraads et al. 1987) . If the premolar Sü-2019 belongs to the same taxon as we hypothesize here, its morphology is also consistent with a p4 of Bachitherium . This “gutter-like” p4 showing transverse cristids subdividing the talonid into two basins is also present in the type species Bachitherium curtum Filhol, 1877 (MNHN-Qu3917- holotype), and in Bachitherium insigne Filhol, 1882 (MNHN-Qu3902; MNHN-Qu3918- holotype). However, the p4 of B. curtum and B. insigne display not one but two neocristids extending transversely from the lingual and labial postprotocristids respectively, but they do not fuse and leave a notch, medially situated, between the extremities of these two crests. This morphology is not what is observed on Sü-2019, where a single transverse neocristid extends transversely from the labial postprotocristid to reach the lingual wall of the posterior valley, which is devoid of a neocristid. We consider that Sü-2017 and Sü-2019 are closer to Bachitherium than to any other ruminant taxon, but the morphology of Sü-2019 does not match any species of Bachitherium known so far. Given the paucity of the current fossil material, we refer Sü-2019 to cf. Bachitherium sp. The morphology of Sü-2019 is also unlike the p4 of Bachitherium thracencis Mennecart, Geraads, Spassov & Zagorchev, 2018a from the late Eocene of Bulgaria, in which the gutter-like talonid of p4 lacks a transverse cristid ( Mennecart et al. 2018a). Thus, if our assumption is correct, it worth noting that there are two forms of bachitheriids in the late Eocene of Balkanatolia, suggesting a diversification of the family in this biogeographic province before its dispersal to Western Europe at MP 23 (“ Bachitherium dispersal event” of Mennecart et al. 2018a, b).