New Amphicyonids (Mammalia, Carnivora) from Moghra, Early Miocene, Egypt Author Morlo, Michael Author Miller, Ellen R. Author Bastl, Katharina Author Abdelgawad, Mohamed K. Author Hamdan, Mohammed Author El-Barkooky, Ahmed N. Author Nagel, Doris text Geodiversitas 2019 2019-11-07 41 21 731 745 journal article 10.5252/geodiversitas2019v41a21 5687d3cd-124a-450f-9e42-6272f23a94b4 1638-9395 3695843 urn:lsid:zoobank.org:pub:F2985B60-626C-4FF6-8C0F-C62C34531199 Amphicyon giganteus ( Schinz, 1825 ) ( Fig. 4 ) In the synonym list, only first description and citations of the African record is given. Canis giganteus Schinz, 1825: 342 . Canis d’une taille gigantesque – Cuvier 1824 : pl. 193 , fig. 20. Amphicyon giganteus – Kuss 1965: 66 . — Morales et al. 1998 : fig. 7; 2003: 180. — Jiangzuo et al. 2019: 6 . Amphicyonidae gen. et sp. indet. – Hendey 1978 : fig. 4 . Megamphicyon giganteus – Morales et al. 2016: 147 . HOLOTYPE . — M7753, Muséum d’Orléans , France . TYPE LOCALITY. — Avaray, France . AGE AND HORIZON IN AFRICA. — Late Early Miocene, contemporary to European biozone MN4. AFRICAN RECORD. — Left P4 (DPC 14532/1), left M1 (DPC 5426), Left M2 (DPC 8981) from Egypt ( Morlo et al. 2007 ), PQAD 1520, right mandible fragment from Arrisdrift, Namibia ( Morales et al. 1998 , 2003 , 2016 ). NEW SPECIMEN. — CUWM 53, left mandible fragment with crowns of m1-2, and alveoli for a single-rooted p1, double-rooted p2-p4, and a single-rooted m3. DESCRIPTION AND COMPARISONS The left mandible fragment CUWM 53 preserves the crown of m1, part of m2, and alveoli for a single-rooted p1, doublerooted p2-p4, and a single-rooted m3. No diastemata are present. As the inferior part of the mandible is broken, it is not possible to assess its original height. Both m1 and m2 are robust teeth, including their roots. The m1 paraconid-protoconid blade appears to show strong wear, but this feature may be due in part to abrasion of this specimen. The protoconid is the tallest cusp of the trigonid, with the metaconid and paraconid subequal in height. The metaconid is tilted slightly posteriorly. The talonid makes up more than half of the tooth and a large hypoconid, which approaches the metaconid and paraconid in height, is present. The talonid also has a small entoconid, and a cingulid is present. The m2 is broken mesiolingually but a tall protoconid and lower metaconid are preserved. A crest extends from the metaconid distolingually to join the cingulid around the talonid. FIG. 4. — CUWM 53, Moghra, Early Miocene, Amphicyon giganteus ( Schinz, 1825 ) : A , lingual view; B , occlusal view; C , labial view. Scale bar: 100 mm. Specimen CUWM 53 resembles Amphicyon giganteus and is assigned to that species rather than Afrocyon ( Arambourg 1961 ) on the combined basis of having more slender m1 and m2, presence of a massive m1 talonid with the hypoconid as the most prominent cusp, an m2 which is subequal in height to the m1 talonid, lack of a talonid basin on m1, and the presence of an m3 alveolus indicating a single-rooted m3. The Moghra mandible also shares a suite of features with Amphicyon giganteus from Arrisdrift ( Hendey 1978 ; Morales et al. 1998 , 2003 ). These include comparable size, a double-rooted p2, lack of premolar diastemata, m1 with a low metaconid and well-developed hypoconid, the presence of a paraconid on m2, and presence of a single-rooted m3. Morales et al. (2016) referred Amphicyon giganteus from Arrisdrift to the genus “ Megamphicyon ”, a reassignment of the species suggested by the work of Kuss (1965) , but a move that has been rejected by many authors (e.g., Ginsburg & Telles-Antunes 1968; Viranta 1996 ; Morales et al. 1998 , 2003 ; Peigné et al. 2006a , 2008 ; Jiangzuo et al. 2018 , 2019 ; Bastl et al. 2018 ). Here we join the majority and so recognize the species A. giganteus for the Arrisdrift and Moghra material. European A. giganteus differs from A. major (Middle Miocene of Europe) in being larger on average, lacking diastemata between its premolars, having p4 larger (see Morales et al. 2003 ), m1 with an entoconid that is narrower and tapers posteriorly, and m2 with a higher trigonid ( Jiangzuo et al. 2018 ). All of these characters also separate the African record of A. giganteus from A. major . ADDITIONAL SPECIMENS Three isolated upper teeth (P4: DPC 14532/1, M1: DPC 5426, M2: DPC 8981) from Moghra previously described as belonging to “ Cynelos sp. nov.” ( Morlo et al. 2007 ) are here provisionally attributed to A. giganteus . These same specimens have been discussed by other authors as perhaps belonging to Ysengrinia ginsburgi ( Morales et al. 2010: 48 ) , “ Afrocyon n. sp.” ( Morales et al. 2016 ), or Cynelos macrodon ( Adrian et al. 2018 ). The P4 resembles that of A. giganteus from La Barranca and Arroyo de Val ( Peigné et al. 2006b : pl. 2, fig. 1-2), but has the protocone placed slightly more anterior. It differs from C. macrodon and C. ginsburgi n. comb. in having a much larger protocone, a feature that is nearly absent in C. macrodon and is greatly reduced in C. ginsburgi n. comb. The M1 resembles A. giganteus from Farinheira ( Ginsburg & Antunes 1995 : pl. 1 fig. 2) and La Barranca ( Peigné et al. 2008 : pl. 2, fig. 5), but the talon of the Moghra specimen is shorter and narrower. In the former respect the Moghra M1 is reminiscent of the M1 of A. eppelsheimensis from the Late Miocene of Anjou ( Gagnaison et al. 2017 : fig. 2b). However, the two species are clearly different in that the M1 of A. eppelsheimensis displays a triangular occlusal outline ( Kuss 1965 , Morlo pers. observ.). The M1 from Moghra differs from members of Cynelos in having a mesiodistally very short talon relative to the trigon, with the protocone and metacone higher, more pronounced cingulae, and a small metaconule present. In these characters the Moghra specimen is unlike the holotype of C. macrodon from Rusinga (see Morales et al. 2016 : fig. 9, 2A), and is even more distant from the C. macrodon material from Kalodirr described by Adrian et al. (2018 : fig. 5.2-4). The short talon also distinguishes DPC 5426 from C. ginsburgi n. comb. ( Morales et al. 2016 : fig. 9, 5), C. euryodon , and C. lemanensis . As with P4 and M1, the isolated M2 DPC 8981 from Moghra resembles European specimens of A. giganteus , in this case from Bézian ( Ginsburg & Bulot 1982 : pl. 1, fig. 2), Olival de Suzana ( Ginsburg & Antunes 1995 : pl. 1, fig. 2), and La Barranca ( Peigné et al. 2008 : pl. 2, fig. 6). The Moghra specimen differs from the M 2 in C. euryodon ( Morales et al. 2016 : fig. 2, 4), C. lemanensis ( Peigné & Heizmann 2003 ), C. macrodon from Napak-I, and C. cf. macrodon from the Muruyur Formation, in having a more rectangular shape, possessing stronger labial and lingual cingulae, and with u-shaped, rather than v-shaped hypocone crests. Another African specimen possibly belonging to Amphicyon is the left mandible NHM M 82373 from Gebel Zelten, Libya , which has been discussed elsewhere as representing Afrocyon burolleti ( Morales et al. 2010 : fig. 2). However, the Gebel Zelten mandible differs from the poorly preserved holotype of Afrocyon burolleti ( Arambourg 1961 ; Morales et al. 2010 : fig. 1; Werdelin & Peigné 2010 : fig. 32.1) in possessing a single-rooted m3, and in having a shallower mandible and less acute coronoid angle. The mandible also differs from Cynelos in lacking premolar diastemata, having p4 shorter relative to m1 and m2, and a broader talonid in m1 and the m2 with a wider mesiolingual enlargement. In all these features, NHM M 82373 resembles Amphicyon giganteus , although given the Gebel Zelten specimen’s much smaller size, we refer the mandible to “cf. Amphicyon ”. REMARKS Fossil material of Amphicyon giganteus has been known since the time of Cuvier (1824) , but the range of morphological variation within the species remains an open discussion. Until now, A. giganteus was represented in Africa only by the single mandible reported from Arrisdrift, South Africa ( Morales et al. 1998 ) and some postcranial remains from Gebel Zelten, attributed to the species by Ginsburg (1980) and Ginsburg & Welcomme (2002) . With the addition of CUWM 53 from Moghra, the species is again interpreted as having a pan-African distribution ( Werdelin & Peigné 2010 ). The three isolated upper teeth from Moghra here attributed to A. giganteus are the first upper teeth described from A. giganteus in Africa. Even though there is little appropriate material available for comparison, the three specimens show features that are similar to those seen in A. giganteus from Europe ( Ginsburg & Bulot 1982 ; Ginsburg & Antunes 1995 ; Peigné et al. 2008 ) and which have not been documented in Cynelos , namely: P4 has a much stronger protocone, M1 has an extremely short talon, and M2 has a peculiar morphology with u-shaped hypocone crests and an enlarged lingual cingulum. Re-evaluation of the mandible NHM M 82373 from Gebel Zelten suggests that a second, smaller species of Amphicyon , cf. Amphicyon , is also present in Africa. The Gebel Zelten mandible resembles the Moghra material of A. giganteus much more than it does to specimens of Afrocyon burolleti or Cynelos . However, NHM M 82373 is much smaller than A. giganteus and its m2 shows a slight mesiolingual enlargement not observed among other published A. giganteus specimens. Amphicyon has already been reported from Gebel Zelten based on postcranial material ( Ginsburg 1980 ; Ginsburg & Welcomme 2002 ) that, however, is too large to belong to the same species as NHM M 82373. Large Amphicyon survived in Africa longer than on any other continent as is evidenced by “ Amphicyonidae species A” from Lothagam ( Werdelin 2003 ; Werdelin & Peigné 2010 ).