Duchesneodus primitivus ( Lambe, 1908 )

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 : 330-336

Duchesneodus primitivus ( Lambe, 1908)

HOLOTYPE: NMC 6421, a partial mandible with right i1–m3, left i1–c, and p2.

TYPE LOCALITY: Cypress Hills Formation, Saskatchewan, Canada.

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

DETERMINATION: Nomen dubium, possibly a synonym of either Protitanops curryi or Notiotitanops mississippiensis .

REMARKS

Lambe (1908) based Duchesneodus primitivus on a mandible (NMC 6421), originally referring it to the genus Megacerops (fig. 165). However, this specimen retains three small globular incisors and a long postcanine diastema, two characters that actually distinguish it from Megacerops (sensu Mihlbachler et al., 2004b) . Osborn (1929a) reassigned the species to the genus Teleodus because it shared with Teleodus avus three globular lower incisor pairs. However, Lucas and Schoch (1982) and Lucas (2004) convincingly argued that Teleodus was an invalid genus. They subsequently reassigned Teleodus primitivus to Duchesneodus , a genus whose type species, D. uintensis , retains three lower globular incisors. Lucas and Schoch (1989b) differentiated D. primitivus from D. uintensis based on the following: a relatively large postcanine diastema and vertical labial faces of the lower cheek teeth. However, this diagnosis does not justify a separate species. At least one mandible of Duchesneodus uintensis (CMNH 11761) has a postcanine diastema, and although it is shorter than that of NMC 6421, postcanine diastema length tends to be variable in brontothere species. Secondly, I was unable to confirm that the labial faces of the lower cheek teeth were of a different angle as suggested by Lucas and Schoch (1989b) from Duchesneodus uintensis or any other similar brontothere.

I can find no morphological evidence that NMC 6421 is distinct from Duchesneodus uintensis , except, perhaps, that the inferior margin of the symphysis is rounded in NMC 6421, while in mandibles of D. uintensis , the inferior margin of the symphysis tends to be straight or convex. This is weak evidence, at best, for arguing two species. Moreover, there are other large brontotheres that retain three globular incisor pairs and a postcanine diastema. Among these is Eubrontotherium clarnoensis ; however, the anterior end of the symphysis of E. clarnoensis is not narrowed as in NMC 6421. Moreover, E. clarnoensis does not have a p3 metaconid, in contrast to NMC 6421. E. clarnoensis can therefore be ruled out. However, Protitanops curryi and Notiotitanops mississippiensis cannot be ruled out. The anterior portions of the mandibles of these species are not known, although each of these species has a long upper postcanine diastema similar in length to the lower postcanine diastema of NMC 6421. It is therefore possible that D. primitivus is either a synonym of Protitanops curryi or Notiotitanops mississippiensis . Lucas et al. (2004) suggested that Duchesneodus primitivus is more likely to have been synonymous with Protitanops curryi rather than Notiotitanops mississippiensis , but this assessment was based on size similarity. However, N. mississippiensis and P. curryi are each known only from holotypes and although the type of N. mississippiensis is somewhat smaller than that of P. curryi , the differences are minor and individuals of these species probably overlapped in size. Moreover, NMC 6421 is only slightly larger than the few known mandibles of Duchesneodus uintensis . Therefore, because D. primitivus could belong to (1) Duchesneodus uintensis , (2) Protitanops curryi , or (3) Notiotitanops mississippiensis with nearly equal probability, this species is a nomen dubium.

Sivatitanops birmanicum ( Pilgrim and Cotter, 1916)

LECTOTYPE: GSI C315 (in part), lingual fragment of a premolar.

TYPE LOCALITY: Myaing Area, Myanmar ( Burma).

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

SYNONYMS: Sivatitanops cotteri Pilgrim, 1925 .

REFERRED SPECIMENS: (From ‘‘Myaing township of Pakokku district, 6.5 furlongs distant from hill 1258 and in a direction 9° West of South from it, Pondaung’’ [ Pilgrim 1925]) GSI C332, a left P3 (?) (lectotype of Sivatitanops cotteri Pilgrim, 1925 ); GSI C 334, a right P4; (from near the Black Sea coast of Bulgaria) AMNH 108191 (cast), a right P4; AMNH 141258 (cast), a right P4. The numbers listed above for the last two specimens are numbers of casts in the AMNH; actual specimen numbers were not reported by Nikolov and Heissig (1985). Several other specimens discussed below may belong to Sivatitanops birmanicum , but they are not directly referable.

DETERMINATION: Nomen dubium, this species could be a synonym of Brachydiastematherium transylvanicum .

DESCRIPTION

Noteworthy features of the lectotype upper premolar fragment (GSI C315, in part) include a very prominent anterolingual cingular cusp and a posterolingual cingular cusp (fig. 166c). Other aspects of this fragment include a single, large protocone, and a discontinuous lingual cingulum. An anteropossibly belong to this species are discussed below.

lingual cingular cusp on the premolars is seen in occasional specimens of Megacerops and Duchesneodus although it is larger and more developed in premolars of Sivatitanops birmanicum . Other premolars described by Pilgrim (1925) such as GSI C334 (fig. 166e) and GSI C332 (fig. 166f) also have a large anterolingual cusp and posterolingual cingular cusp. In addition, two upper premolars from Bulgaria described by Nikolov and Heissig (1985) show the same distinctive characteristics (fig. 166a, b). No other specimens are directly referable to Sivatitanops birmanicum , although several specimens that

REMARKS

The brontothere fossils collected from the Pondaung Formation of Myanmar seem to represent at least four species: Sivatitanops birmanicum , ‘‘ Metatelmatherium (?)’’ lahirii (in part), ‘‘ Sivatitanops (?)’’ rugosidens, and Bunobrontops savagei . Among these only Bunobrontops savagei , with its unique set of molar characters, is a clearly valid species name. The remaining three seem to represent distinct species that occur in the Pondaung, but their names are technically nomina dubia because the scrappy nature of the known materials limits comparison with species from other areas.

Pilgrim and Cotter (1916) based Sivatitanops birmanicum on five upper cheek-tooth fragments that were all assigned to GSI C315. Pilgrim and Cotter (1916) originally referred this species the genus ‘‘ Telmatherium (?)’’. Later, Pilgrim (1925) named a new genus and species, Sivatitanops cotteri , from a series of isolated upper teeth. Four specimens were referred to as types: a M1 or M2 (GSI C330), a P2 (GSI C331), a P3 (GSI C332) (fig. 166f), and an upper canine (GSI C333). In addition to these four specimens, Pilgrim (1925) referred a P4 (GSI C334) (fig. 166e), an incisor (GSI C335), a P1(?) (GSI C335), and a jaw fragment with a premolar and partial canine (GSI C338) to S. cotteri . The P3 (?) and P4 of S. cotteri are similar to the lectotype premolar of ‘‘ Telmatherium (?)’’ birmanicum (fig. 166c), particularly in the possession of prominent anterolingual and posterolingual cingular cusps. These similarities led Pilgrim (1925) to reassign birmanicum to Sivatitanops . However, S. birmanicum and S. cotteri continued to be considered separate species.

Whether the five specimens all assigned to GSI C315 originally referred to Sivatitanops birmanicum belong to the same individual or even the same species is open to question. I consider the lingual fragment of the premolar (fig. 166c) the lectotype because it expresses apparently diagnostic characters (anterolingual and posterolingual cingular cusps). Nor was it was never made clear which of the four specimens of Sivatitanops cotteri was intend- ed to be the holotype of that species. The P3 (?) (GSI C332) is here considered the lectotype of S. cotteri . That specimen also has prominent anterolingual and posterolingual cingular cusps. Therefore, S. cotteri is a junior synonym of S. birmanicum . The remaining specimens that have been referred to S. birmanicum and S. cotteri lack these diagnostic features and cannot be certainly referred to S. birmanicum or any other species.

In addition to the specimens discussed above, Pilgrim (1925) assigned a partial skull (GSI C329) as the cotype of Sivatitanops birmanicum (fig. 167). This specimen consists of an anterior portion of skull lacking the nasal processes and with poorly preserved teeth. Using this skull, Pilgrim (1925) distinguished S. birmanicum from S. cotteri with a broader molar, and a more prominent anterolingual cingular cusp. Differences in molar proportions were commonly used in the past to diagnose brontothere taxa but without recognizing that differential molar wear significantly affects the length/width proportions (see remarks under Epimanteoceras formosus for clarification). Pilgrim (1925) admitted that his assignment of this skull to S. birmanicum was arbitrary. Therefore, the identity of this skull as S. birmanicum is questionable.

Despite the uncertain identity of the skull (GSI C329), it is the best available brontothere specimen from the Pondaung Formation that might represent Sivatitanops birmanicum and an abbreviated description is given here, highlighting some of the more important characters noted by Pilgrim (1925). The dorsal surface of the cranium is broad and flat. The anterior rim of the posterior nares is located next to M3. The incisors are known only from roots, but they appear to form an arched tooth row. P1 is apparently absent and there is no diastema between C and P2. The crowns of the premolars are badly damaged. The molars have an anterolingual cingular cusp and central molar fossa. M3 has a small hypocone. The lack of a postcanine diastema is reminiscent of Brachydiastematherium transylvanicum , Metatitan , Duchesneodus , and Megacerops . However, the skulls of Metatitan , Duchesneodus , and Megacerops all differ from GSI C329. This leaves the possibility that S. birmanicum is a junior synonym of Brachydiastematherium transylvanicum (see below).

Colbert (1938) referred a large mandible (AMNH 20014) to S. cotteri (5 S. birmanicum ) although there is no evidence, other than its general similarity in size, that this mandible is correctly identified. The mandible is massive and only the p4–m3 are completely preserved. The teeth are stereotypical in their morphology. The crown of the canine is missing, but its root is one of the largest canines of any known brontothere specimen.

Although the best brontothere material from the Pondaung Formation (skull and jaw) cannot be taxonomically identified with any certainty, several lines of evidence seem to link Sivatitanops birmanicum and Europe- an brontotheres. Mentioned earlier was the consistency of the skull (GSI C329) with what might be expected of the skull of B. transylvanicum . More compelling evidence comes from several brontothere teeth from upper Eocene strata near the Black Sea coast of Bulgaria that were described by Nikolov and Heissig (1985). This material includes two right P4s (AMNH cast 108191, AMNH cast 141258), a partial upper molar fragment, a lower left m2 (AMNH cast 141255), and a lower right p4 (AMNH cast 141257).

Nikolov and Heissig (1985) assigned these teeth to ‘‘ Menodus (?)’’ rumelicus. The similarities of the premolars with those of Sivatitanops led Nikolov and Heissig (1985) to reassign ‘‘ Menodus (?)’’ rumelicus to the genus ‘‘ Sivatitanops (?)’’. Lucas and Schoch (1989a) were skeptical of the specific identification by Nikolov and Heissig (1985) due to the dubious nature of the taxon ‘‘ Menodus (?)’’ rumelicus, whose holotype may actually be a fossil imported from North America ( Osborn, 1929a; Lucas and Schoch, 1989a). The specimens described by Nikolov and Heissig (1985) are of interest not only because they are the second definite occurrence of brontotheres in Europe (second to the holotype mandible of Brachydiastematherium transylvanicum ), but because they clearly resemble the southeast Asian species Sivatitanops birmanicum in the prominent anterolingual cingular cusp on the premolars and the posteriorly situated hypocones that are connected to the posterior cingulum. The p4 described by Nikolov and Heissig (1985) (fig. 166d) is similar in size to the p4 of the holotype jaw of Brachydiastematherium transylvanicum . However, the p4 of Brachydiastematherium transylvanicum has an unusual crest of enamel extending posteriorly from the middle of the cristid obliqua (fig. 98) that is not seen in the p4 pictured in fig. 176. This suggests that the Sivatitanops -like material from Europe described by Nikolov and Heissig may not be Brachydiastematherium transylvanicum , but a second European brontothere. At any rate, the available material is too incomplete to make a definitive conclusion regarding synonymy of Brachydiastematherium and Sivatitanops . Therefore, Sivatitanops birmanicum must be considered a nomen dubium.

‘‘ Metatelmatherium (?)’’ lahirii ( Pilgrim, 1925)

LECTOTYPE: GSI C342 , a right jugalmaxillary fragment with P4 and M1.

TYPE LOCALITY: Pondaung Formation, 1 mile east-southeast of Sinzwe village, Myanmar ( Burma).

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

DETERMINATION: Nomen dubium, possibly a synonym of Sivatitanops birmanicum .

REMARKS

Pilgrim (1925) named ‘‘ Eotitanotherium (?)’’ lahirii from a few fragments including a right maxillary and jugal fragment with a heavily worn M1 and a very badly damaged P4 (GSI C 342) (fig. 168a), an ectoloph of an upper molar (GSI C341) (fig. 168c), and an upper left premolar (GSI C340) (fig. 168b). None of these was designated as the holotype, although Colbert (1938) selected GSI C342 as the lectotype and questionably reassigned this species to Metatelmatherium .

Pilgrim (1925) differentiated ‘‘ Metatelmatherium (?)’’ lahirii from other Pondaung species that he had referred to Sivatitanops by the following features of the M1 of the lectotype specimen: the forward position of the protocone, the less pronounced anterior cingulum ridge, and the absence of a protoconule. Unfortunately, the P4 of the lectotype is too damaged to compare to other specimens. The apparent differences of the molars relate to the fact that the molar of GSI C432 is more heavily worn (including substantial interstitial wear) than molars that assigned to lahirii seems to represent a species of brontothere that is distinct from Sivatitanops birmanicum and ‘‘ Sivatitanops (?)’’ rugosidens. This premolar is relatively advanced with two distinct and completely disconnected lingual cusps (protocone and hypocone). More notably, it does not exhibit the anterolingual cingular cusp that is well developed in S. birmanicum . This tooth is insufficient to pin down its true taxonomic identity, but the differences between this premolar and those of S. birmanicum and ‘‘ Sivatitanops (?)’’ rugosidens seems to exceed the normal degree of intraspecific variation seen in the premolars of other brontothere species.

were referred to Sivatitanops by Pilgrim (1925). The molar of GSI C342 is otherwise similar to those referred to Sivatitanops by Pilgrim (1925) with distinct central molar fossae, and a prominent anterolingual cingular cusp. These molar characters are typical of several brontothere species and the taxonomic identity of the lectotype of ‘‘ Metatelmatherium (?)’’ lahirii (GSI C432) is uncertain, although ‘‘ Metatelmatherium (?)’’ lahirii could be a synonym of Sivatitanops birmanicum .

The species name, lahirii, is undoubtedly invalid, due to the dubious identity of the lectotype specimen. However, an isolated premolar (GSI C340) that was originally

‘‘ Sivatitanops (?)’’ rugosidens Pilgrim, 1925

HOLOTYPE: GSI C339 , a partial P4 and M1.

TYPE LOCALITY: Pondaung Formation, 1 mile east-southeast of Sinzwe village, Myanmar ( Burma).

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

DETERMINATION: Nomen dubium, holotype is best identified as cf. Rhinotitan sp.

REMARKS

Pilgrim (1925) based yet another species of Pondaung brontothere, ‘‘ Sivatitanops (?)’’ rugosidens, on a lingual portion of a P4 (fig. 169a) and a lingual portion of a M1 (fig. 169b) that are reported to belong to the same individual (GSI C339). Colbert (1938) accepted this species but was of the opinion that the material was of little value. This material is different from both Sivatitanops birmanicum and ‘‘ Metatelmatherium (?)’’ lahirii and seems to represent yet another species of brontothere in the Pondaung Formation, possibly Rhinotitan .

The premolar is rectangular, has only a very tiny pinpoint of enamel that could be interpreted as a rudimentary hypocone, and it lacks anterolingual and posterolingual cingular cusps. Therefore, this premolar is inconsistent with Sivatitanops birmanicum as well as the premolar (GSI C340) that was referred by Pilgrim (1925) to ‘‘ Metatelmatherium (?)’’ lahirii (see above). Pilgrim (1925) described the crowns of the teeth as low. Additionally, the area surrounded by the premolar protocone is rugose. Pilgrim’s (1925) description and illustrations indicate specimens that compare well with Rhinotitan , particularly with the very low, small premolar protocone and the crenulated lingual premolar enamel. It is possible this specimen is Rhinotitan , although its specific identity ( R. kaiseni , R. andrewsi , or another species) is uncertain. ‘‘ Sivatitanops (?)’’ rugosidens is a nomen dubium, but its holotype is possibly a species of Rhinotitan .