Amebelodon Barbour, 1927

Geraads, Denis, Kaya, Tanju & Mayda, Serdar, 2005, Late Miocene large mammals from Yulafli, Thrace region, Turkey, and their biogeographic implications, Acta Palaeontologica Polonica 50 (3), pp. 523-544 : 530-531

Genus Amebelodon Barbour, 1927

Type species: Amebelodon fricki Barbour, 1927 ; Cambridge quarry, Nebraska, middle Pliocene.

Amebelodon grandincisivus ( Schlesinger, 1917) Mastodon (Bunolophodon) grandincisivum Schlesinger, 1917

Type locality and age: Maragheh , Iran, Turolian, Miocene .

Material from Yulafl I.—TTMEU−CY−143, left i2 ( Fig. 4F); TTMEU−CY−203, isolated left M2 ( Fig. 4E).

Length (mm)

Fig. 8. Length versus width plot of Tetralophodon and “ Mastodon” grandincisivus m3s.

Description.—The incisor is slightly curved outwards ( Fig. 4F 2 ). It has a rather flattened transverse section (145 × 65 mm). There is no trace of enamel. On the dorsal side there are two grooves; the labial one being deeper than the lingual one. On the ventral side, there are five shallow grooves. The tip of the incisor has a thin cement layer, underneath follows a concentric laminated dentine layer (10 mm), then a uniform core dentine (max. 41 mm). The basal part has a regular dentine layer (10 mm), but the core of the tooth (max 44 mm) consists of rod−cone structures, or tubular dentine ( Tassy 1999).

TTMEU−CY−203 is a nearly complete, heavily worn left M2 with four lophs, a weak anterior cingulum and a well−developed double talon. The tooth is rectangular. The pretrite half−lophs are composed of a main cone, a smaller mesoconelet and anterior and posterior accessory conules that exhibit trefoil wear pattern in the first two half−lophs. On the posttrite side, the posterior accessory conules are present on the first and third half−lophs; and an incipient anterior accessory conule on the second half−loph. The last two half−lophs show an incipient anancoidy with the pretrite side positioned anterior to the posttrite side. There is a trace of cement at the base of the interlophs.

Comparisons.—The systematic position of “ Mastodon ” grandincisivus among late Neogene elephantoids has long been debated. It was included in Stegotetrabelodon by Tobien (1978), and more recently in the amebelodonts ( Tassy 1999). We follow this latter opinion, since the dorso−ventral compression and tubular dentine structure of the holotype ( Schlesinger 1922) clearly demonstrate affinities with this group.

Similar incisors are known from some (but not many) localities in Eurasia, Africa, and North America, of middle to late Miocene age, but the differences in outline of the cross−section point to the occurrence of two or more different species (and/or sexual dimorphism) in the Old World. The structure of the rod cones from Yulafli is similar to that of an incisor from Arapli, near Tekirdaǧ, referred to Amebelodon (Platybelodon) by Gaziry (1976: pl. 3: 2), but the latter was certainly more dorso−ventrally compressed, and more regularly compressed dorsally. At Kerch in Crimea ( Pavlov 1904) and Sahabi in Libya ( Gaziry 1987), the section is almost rectangular, whereas it is more irregular, and thus more similar to that of Yulafli, at Maragha and Pestszentlörincz in Hungary ( Schlesinger 1917; 1922; comparisons in Tassy 1999: fig. 18.8), and perhaps also in Gansu, China ( Tobien et al. 1986: fig. 24). However, the incisor from Yulafli is smaller and more convex ventrally than the teeth from Maragha and Pestszentlörincz ( Table 2 and Fig. 9 ). Perhaps the specimen most similar to the tooth from Yulafli is from Orjachovo in Bulgaria ( Bakalov and Nikolov 1962: pl. 66).

The M2 TTMEU−CY−203 differs from that of T. longirostris from Yulafli, which has a simpler lophid structure and a clearly smaller size. It resembles the Mannersdorf sample of “ Mastodon” grandincisivus ( Schlesinger 1917: pl. 15: 1). They share the pretrite trefoil pattern in the first two half−lophs, posterior accessory conules on the posttrite side, and secondary conules blocking the interlophs, but the teeth from Mannersdorf are larger. The mastodont from Arapli is trilophodont, and this is a further difference, if in both sites the molars are correctly referred to the same taxon as the incisors.

Thus, there is little doubt that these two teeth should be referred to what is often called “ Mastodon” grandincisivus , but what should more correctly, even if provisionally, be called Amebelodon . Its smaller size than the similar form from Kerch and Maragha suggests that it might be earlier, but given the great size variability in proboscideans, this conclusion is, admittedly, quite weak.

Figures

Fig. 2. Indarctos arctoides, TTMEU−CY−46, Yulafli, Turkey, Vallesian, late Miocene. Left mandibular ramus, lateral (A) and occlusal (B) views of p2–m2 (stereo).

Fig. 9. Measurements of the cross−section of the lower i2s in various Proboscideans. From Tassy (1986: fig. 14) and y = Yulafll.