Metatelmatherium parvum Granger and Gregory, 1943

Mihlbachler, Matthew C., 2008, Species Taxonomy, Phylogeny, and Biogeography of the Brontotheriidae (Mammalia: Perissodactyla), Bulletin of the American Museum of Natural History 311 (1), pp. 1-475 : 345-351

Metatelmatherium parvum Granger and Gregory, 1943

HOLOTYPE: AMNH 20168 , a left mandible fragment with p3–p4.

TYPE LOCALITY: Irdin Manha Formation, Inner Mongolia, China.

AGE: Middle Eocene (Irdinmanhan land mammal ‘‘age’’).

DETERMINATION: Nomen dubium, holotype lacks diagnostic characters.

REMARKS

The type of Metatelmatherium parvum is a mandible fragment with only two teeth (p3– p4) (AMNH 20168) that do not exhibit a diagnostic set of characters (fig. 174). In size, it is most similar to Metatelmatherium ultimum , but it has a more lingually oriented paralophid than the one available Asian specimen of M. ultimum (AMNH 26411). In this respect, the specimen more closely resembles Protitan grangeri , although the specimen seems somewhat smaller than fossils that belong to that species.

Hyotitan thomsoni Granger and Gregory, 1943

HOLOTYPE: AMNH 26401 , a nearly complete mandible in two pieces with right i2–c, p–m3, and left i1–m3.

TYPE LOCALITY: Camp Margetts, ‘‘Houldjin’’ Formation, Inner Mongolia.

AGE: Middle Eocene (Irdinmanhan land mammal ‘‘age’’).

DETERMINATION: Nomen dubium. This species could be synonymous with Qufutitan zhoui .

DESCRIPTION

The holotype of Hyotitan thomsoni consists of a nearly complete and very large mandible (fig. 175) with only a few missing teeth. The symphysis is unusually long and narrow with a long postcanine diastema. The ventral margin of the symphysis is angled about 45 °. The symphysis extends about to the midpoint of p3.

The incisor row is slightly arched anteriorly. The incisors are all similar in size. Overall, they are small and nearly featureless; however, this is partly an artifact of heavy wear.

The worn crowns show evidence of having had a semiglobular or subcaniniform shape. Traces of lingual cingulids are still apparent on the incisors, but there is no evidence of a labial incisor cingulid.

Two of the most distinctive features of Hyotitan are the very large canines and the relatively elongate postcanine diastema. The tusklike canines are heavily worn, but they are clearly much larger than other brontotheres of similar body size. The postcanine diastema (76 mm) is more than double the length of p2.

The premolar crowns are broad and tall and closely resemble those of ‘‘ Protitan ?’’ cingulatus . The crown of p1 sits well below the crown of p2 and consists of a single large cusp and a very short cusplike talonid heel. The trigonid of p2 is longer than the talonid, but on p3 and p4, the trigonids and talonids are of similar length. The trigonid and talonid of p2 are nearly equal in width, while the trigonids of p3 and p4 are narrower than the talonids. The paralophid of p2 arches only slightly lingually forming a minor lingual notch in the trigonid. The p2 protolophid is straight and angled slightly lingually. The p2 lacks a discernable metaconid. On the p3, the paralophid is angled somewhat more lingually and the p3 protolophid extends distolingually from the protoconid, forming a broad lingual notch in the trigonid. A distinct p3 metaconid is present at the junction of the protolophid. On p4, the paralophid and protolophid are arched strongly lingually, forming a nearly molariform trigonid basin. A large p4 metaconid is positioned mostly lingually from the protoconid. The talonid basin of p2 is small with a short cristid obliqua and hypolophid, creating a small talonid basin. The cristids obliqua and hypolophids of p3 and p4 are longer and create increasingly broader talonid basins.

The lower molars of Hyotitan thomsoni are relatively broad with shallow trigonid and talonid basins and very weak molar ribs. The m3 is among the shortest among brontotheres. Its length/width proportions resemble Desmatotitan tukhumensis . However, the hypoconulid of m3 is remarkably narrow. Finally, there is a strong labial cingulid on all cheek teeth posterior to p1.

REMARKS

Granger and Gregory (1943) based Hyotitan thomsoni on a very large mandible, AMNH 26401. It shows a unique combination of very large canines, small but not vestigial incisors, and a very long postcanine diastema with an elongate symphysis. The canines of the holotype of H. thomsoni are larger than any other similarly sized brontothere except for a partial mandible (AMNH 20014) described by Colbert (1938) from the Pondaung deposits of Myanmar that is possibly Sivatitanops birmanicum . However, the symphysis of that specimen is much shorter than that of Hyotitan and the m3 is more elongate. Other large brontotheres with relatively elongate symphyses include Pollyosbornia altidens , Gnathotitan berkeyi , and Diplacodon elatus . The p3 metaconid and wide, well-developed p2 talonid clearly distinguishes Hyotitan thomsoni from Pollyosbornia altidens . Gnathotitan berkeyi has a differently shaped mandible than H. thomsoni and has much larger incisors and a more elongate m3. Specimens of Diplacodon elatus are much smaller than Hyotitan thomsoni , the m3 is more elongate, and the posterior margin of the symphysis is more posterior. The premolars of Hyotitan thomsoni most closely resemble those of ‘‘ Protitan ?’’ cingulatus , particularly in relative dimensions, strong labial cingulid, anteroposteriorly compressed p1, and p3 metaconid. However, Hyotitan thomsoni is much larger than ‘‘ Protitan ?’’ cingulatus , has a more elongate symphysis, and lacks the labial incisor cingulids of that species.

Hyotitan thomsoni could be a synonym of Qufutitan zhoui , a species known only from a partial skull. Both are of similar size and they share a remarkable number of peculiar similarities that are discussed further with remarks with Qufutitan zhoui . Hyotitan thomsoni is therefore classified here as dubious. However, it should be noted that H. thomsoni Granger and Gregory, 1943 , has priority over Qufutitan zhoui Wang and Wang, 1997 . In this instance, I part from the rule of priority and accept Q. zhoui because its holotype retains much more character data. If one adheres to ICZN rules, H. thomsoni is valid whereas Q. zhoui is dubious.

‘‘ Protitan ?’’ cingulatus Granger and Gregory (1943)

HOLOTYPE: AMNH 26412 , a right mandibular ramus with P1–M3.

TYPE LOCALITY:?‘‘Houldjin’’, 10 miles west of Camp Margetts, Inner Mongolia, China.

REFERRED SPECIMENS: (From the ‘‘Houldjin’’ beds of Camp Margetts, Inner Mongolia) AMNH 26403, a right mandibular ramus with p1–m3; (from the Irdin Manha Formation, Inner Mongolia) AMNH 20110, a complete mandible missing only the left i2.

AGE: Middle Eocene (Irdinmanhan land mammal ‘‘age’’).

DETERMINATION: Nomen dubium; it could be a synonym of Epimanteoceras formosus .

DESCRIPTION

In addition to the partial holotype jaw (AMNH 26412) (fig. 176) and another partial mandible from the?‘‘Houldjin’’ of Camp Margetts (AMNH 26403), there is a virtually complete mandible (AMNH 20110) with less heavily worn teeth from the Irdin Manha Formation (fig. 177). AMNH 20110 is slightly larger than the Camp Margetts ‘‘ Protitan ?’’ cingulatus material, but it has congruent cheektooth morphology. The following description is based primarily on AMNH 20110. The cheek teeth of the holotype are more worn that that of AMNH 20110, but they match the description given below, except where noted.

The horizontal ramus of AMNH 20110 is deep and not nearly as slender as that of Protitan grangeri . The coronoid process is moderately curved and extends higher than the condyle. The inferior margin of the symphysis is steep and strongly curved. The symphysis extends to the p3 talonid. The incisor row is narrow and forms a semicircular arch anterior to the canines. The incisors of AMNH 20110 are similar in size to those of Protitan grangeri and Epimanteoceras formosus . The incisors are all similar in size and the i2 is not enlarged. The crowns are more or less subcaniniform. The apex of the right i1 is worn off, but it appears to have been similar in shape to the i2. The crown of the right i2 is damaged. The left i2 crown is short and conical, with a bluntly rounded apex. The i3 is somewhat more pointed. The i3 is round and not as elongate as in Protitan grangeri . Each incisor has a distinct lingual cingulid. Additionally, there are distinct labial cingulids on each incisor. The canines of AMNH 20110 are rather small, are almost perfectly rounded in cross section, and they are erect. A precanine diastema is absent. The postcanine diastema is slightly longer than p2.

The premolars of ‘‘ Protitan ?’’ cingulatus are much taller, broader, and more molariform than those of Protitan grangeri and most closely resemble those of Hyotitan thomsoni in their general size, proportions, and degree of development of labial cingulids. The p1 is very small with a single cusp and a shortened talonid heel. The trigonid of p2 is much longer than the talonid. The p3 and p4 trigonids are only slightly longer than their corresponding talonids. The trigonid and talonid of p2 are of similar width, while the p3 and p4 trigonids are slightly narrower than the talonids. The paralophid of p2 curves slightly lingually, creating a small lingual notch in the trigonid. The p2 protolophid is lingually positioned, but it is straight. The paralophid of p3 arches more strongly lingually and the protolophid arches fully lingually, creating a much broader lingual notch in the trigonid. The paralophid and protolophid of p4 arch fully lingually and form a molarlike trigonid basin. The p2 lacks a metaconid, but large, lingually positioned, and molariform metaconids are present on p3 and p4. The talonid of p2 includes a distinct but short cristid obliqua and a hypolophid that create a small lingual-talonid notch. The cristids obliqua and hypolophids of p3 and p4 are longer and form more nearly molariform basins. The angulations and lengths of the various lophids of the premolars of AMNH 26412 resemble those of AMNH 20110, but the metaconids are worn flat. Lingual premolar cingulids are absent, but the labial premolar cingulids of all cingulatus specimens are very strong. The labial premolar cingulids are somewhat stronger in AMNH 26412 and somewhat weaker on AMNH 20110.

The molars of ‘‘ Protitan ?’’ cingulatus are typical with relatively thin lingual enamel, shallow trigonid and talonid basins, and an elongate m3. There are no lingual cingulids, but like the premolars, the labial molar cingulids are very strong.

REMARKS

Granger and Gregory (1943) erected a new species, ‘‘ Protitan ?’’ cingulatus , based on AMNH 26412. In the same paper they referred the same specimen to another species, Protitan robustus (considered, here, to be a junior synonym of P. grangeri ). Presumably, AMNH 26412 was intended to serve as the holotype of ‘‘ Protitan ?’’ cingulatus , and the referral of AMNH 26412 to P. robustus is mostly likely an accident. Therefore I consider AMNH 26412 the holotype of ‘‘ Protitan ?’’ cingulatus . Granger and Gregory (1943) questionably referred this species to the genus Protitan and distinguished it from other species of Protitan by the heavy external cingulids on p2–m3. Although no other specimens were referred to P. cingulatus by Granger and Gregory (1943), one additional specimen (AMNH 26403) from Camp Margetts resembles the holotype of cingulatus . Additionally, a nearly complete mandible (AMNH 20110) from the Irdin Manha Formation, referred to Protitan robustus by Granger and Gregory (1943) is far more consistent with cingulatus .

The referral of this species to ‘‘ Protitan ?’’ is probably incorrect. The specimens referred to cingulatus by Granger and Gregory (1943) feature a distinctive set of morphological characters. ‘‘ Protitan ?’’ cingulatus differs from other large brontotheres, such as Protitan grangeri , Rhinotitan , Aktautitan , and Protitanotherium in having a relatively molariform p3 with a large metaconid. The premolars and quite broad and tall with very strong labial cingulids; these traits distinguish them from Diplacodon and Gnathotitan . Yet, there is a long postcanine diastema, large subcaniniform incisors, and the symphysis does not extend past p3; these traits rule out a variety of large advanced brontotheres such as Metatitan , Embolotherium , Eubrontotherium . The premolar morphology is essentially indistinguishable from Hyotitan , although ‘‘ Protitan ?’’ cingulatus is much smaller, with a shorter postcanine diastema, and a more elongate m3 with a wider hypoconulid.

In considering those brontotheres whose mandibles are unknown, a possible synonymy with Epimanteoceras formosus , a similarly size species from the same region and time period, seems likely. The mandibles of ‘‘ Protitan ?’’ cingulatus are consistent with what one might anticipate for the mandibles of E. formosu s. Epimanteoceras formosus upper premolars are partially molarized with occasional hypocones; this degree of molarization is roughly equivalent to the degree of molarization seen in the lower premolars of cingulatus . Epimanteoceras formosus also has similar large subcaniniform upper incisors, a similar postcanine diastema, and it is virtually the same size. ‘‘ Protitan ?’’ cingulatus must be considered a nomen dubium because of the possible synonymy with Epimanteoceras formosus .

Epimanteoceras praecursor Yanovskaya, 1953

HOLOTYPE: Collection of the Laboratory of Paleozoology of the Zoological Institute of the Academy of Sciences Kazak SSR 3367a/ 51G, a right maxilla fragment with P4–M2 (the same specimen is referred to as KAN-Z-369/MP-61 by Emry et al. [1998]).

TYPE LOCALITY: 90 km to the northeast of Agadyr’ in the region of the headwaters of the Sary-Su River, on a stream also called Aksoran ( Russell and Zhai, 1987); Kiin Kerish, probably from locality K18 or K 20, either from Kusto or Aksyir Svita of Kazakstan ( Emry et al., 1998).

AGE: Late Eocene (Ulangochuian land mammal ‘‘age’’).

DETERMINATION: Nomen dubium: holotype lacks diagnostic features.

REMARKS

Yanovskaya (1953) referred to the holotype of Epimanteoceras praecursor as SSR 3367a/51G. Later, Emry et al. (1998: fig. 6c) published a photo of an identical specimen that they identified as Epimanteoceras sp. with a different number KAN-Z-369/MP-61. Emry et al. (1998) did not recognize this specimen as the holotype of E. praecursor , although comparison with the original figure from Yanovskaya (1953) leaves little doubt that the specimen reported by Emry et al. (1998) is indeed the holotype of E. praecursor .

The holotype is a maxillary fragment with P4, M1, and M2 (fig. 178). The general shape, size, proportions, and characteristics of these teeth are consistent with a number of other brontotheres. The teeth are relatively advanced. P4 has a small hypocone and each of the molars has a small anterolingual cusp and a shallow central molar fossa. Yanovskaya’s (1953) identification of this specimen as Epimanteoceras is certainly wrong. The anterolingual cingular cusps of the holotype are not consistent with Epimanteoceras formosus . The holotype of E. praecursor appears to represent a more derived brontothere. Asian brontotheres that are of similar size and have a P4 hypocone and a small anterolingual cingular cusp include Parabrontops gobiensis , Dianotitan lunanensis , Eubrontotherium clarnoensis , Pachytitan ajax , Nasamplus progressus . A number of North American species have the same traits. Epimanteoceras praecursor is a nomen dubium and its holotype could belong to one of a number of species.