Open Access
How to translate text using browser tools
30 September 2003 Reevaluation of some ungulate mammals from the Eocene Pondaung Formation, Myanmar
Takehisa Tsubamoto, SOE THURA TUN, Naoko Egi, Masanaru Takai, Nobuo Shigehara, Aung Naing Soe, AYE KO AUNG, Tin Thein
Author Affiliations +
Abstract

We reevaluate some fossil specimens of ungulate mammals from the uppermost middle Eocene Pondaung Formation (central Myanmar), describing some new materials. The taxa studied in this paper are Hsanotherium parvum (Ungulata), Asiohomacodon myanmarensis gen. et sp. nov. (Artiodactyla; Dichobunidae; Homacodontinae), Indomeryx (Artiodactyla; Ruminantia), Indolophus guptai (Perissodactyla; Tapiromorpha; Indolophidae), and Ceratomorpha fam., gen. et sp. indet. (Perissodactyla). (1) The lower molars of Hsanotherium show a similarity to those of Gobiohyus pressidens (Artiodactyla; Helohyidae), and its mesiodistally elongated and trilobed dP4 morphology recalls that of artiodactyls and macroscelideans. However, the unique molar and P4 morphologies of Hsanotherium indicate that Hsanotherium cannot confidently be classified into any present ungulate order, although it can be identified as belonging to the Ungulata because of its large, elongated, and posteriorly projecting hypoconulid on M3. (2) The molar size and morphology of Asiohomacodon recall primitive protoreodontine agriochoerids (Oreodontoidea) such as Protoreodon parvus and derived and agriochoerid-like homacodontine dichobunids such as Pentacemylus, both of which occur in Eocene North America. Asiohomacodon is classified not into the Protoreodontinae but into the Homacodontinae because of the lack of molar metastylid. The lower molar morphology of Asiohomacodon also resembles that of an unusual and agriochoerid-like anthracotheriid, Atopotherium, from Eocene Thailand, although the affinity between these two genera cannot be tested because of the lack of the P4 material of Asiohomacodon. (3) The Pondaung Indomeryx consists of large and small species, I. cotteri (including I. pilgrimi) and I. arenae (including I. minus). Dental morphology in each species of the Pondaung Indomeryx indicates relatively high variation, and the two species are not separable based on their dental morphology. Indomeryx shows many primitive characteristics among ruminants and lacks any critical derived features referable to any ruminant family. (4) Indolophus is referable to primitive tapiromorphs in having a somewhat lophodont dentition and in lacking lingual and buccal cingula and molar metaconule, paraconule, and metastylid. It differs from other tapiromorphs in having a smaller parastyle on the upper dentition and a unique P2–4 morphology with large protocone, high and acute preprotocristid, and no postprotocristid. (5) Although the material of the indeterminate ceratomorph is poorly preserved, its preserved tooth is not identical to any other ceratomorph from the Pondaung Formation, indicating an occurrence of an additional ceratomorph species in the Pondaung fauna.

Introduction

The Pondaung Formation of Myanmar is one of the richest Eocene fossil-bearing deposits of terrestrial mammals in Southeast Asia. The mammalian fossils from this formation were first described by Pilgrim and Cotter (1916), and since then numerous fossils of mammals (Pilgrim, 1925, 1927, 1928; Matthew, 1929; Colbert, 1937, 1938; Ba Maw et al., 1979; Ciochon et al., 1985, 2001; Holroyd and Ciochon, 1995, 2000; Pondaung Fossil Expedition Team, 1997; Jaeger et al., 1998, 1999; Takai et al., 1999, 2000, 2001, 2003; Chaimanee et al., 2000; Ducrocq et al., 2000a, b; Egi and Tsubamoto, 2000; Métais et al., 2000; Tsubamoto et al., 2000a, b, 2001, 2002a; Shigehara et al., 2002; Gunnell et al., 2002; Gebo et al., 2002), lower vertebrates (Hutchison and Holroyd, 1996), and microfossils (Swe Myint, 1999; Hla Mon, 1999; Thet Wai, 1999) have been reported.

In the present work, we reevaluate some fossil material of ungulate mammals from the Pondaung Formation, de-scribing several new specimens. Most of the fossil materials figured in this paper were also figured and mentioned by Tsubamoto et al. (2000a). However, the quality of the pictures of fossils in that reference is poor, and the paper contains only preliminary results. Therefore, we refigure the informative fossil materials reported by Tsubamoto et al. (2000a) with some new materials and revised discussion.

Abbreviations

AMNH = American Museum of Natural History, New York, USA. CM = Carnegie Museum of Natural History, Pittsburgh, USA. DMR = Department of Mineral Resources, Bangkok, Thailand. GSI = Geological Survey of India, Kolkata, India. IVPP = Institute of Vertebrate Paleontology and Paleoanthropology, Beijing, China. LK = Pondaung fossil specimens catalogued by the French paleontologists (Métais et al., 2000). NMMP-KU = National Museum, Myanmar, Paleontology-Kyoto University (Japan). NSM = National Science Museum, Tokyo, Japan. UCMP = Museum of Paleontology, University of California, Berkeley, USA.

Geologic setting and age

The Pondaung Formation is distributed in the western part of central Myanmar (Tsubamoto et al., 2000a, fig. 1, 2002a, fig. 1). It is a freshwater deposit associated with meandering rivers, marshes, and delta plains, and is subdivided into the “Lower” and “Upper” Members (Aye Ko Aung, 1999; Aung Naing Soe, 1999; Aung Naing Soe et al., 2002). All of the mammalian fossils occur in the lower half of the “Upper Member” (Stamp, 1922; Colbert, 1938; Bender, 1983; Aye Ko Aung, 1999).

The mammalian faunal correlation (Pilgrim and Cotter, 1916; Pilgrim, 1925, 1928; Colbert, 1938; Holroyd and Ciochon, 1994), nannoplankton assemblage (Hla Mon, 1999), and biostratigraphic relationships with adjacent marine units (Holroyd and Ciochon, 1994) indicate that most of the Pondaung Formation is Bartonian equivalent, which is now correlated to the upper middle Eocene (Berggren et al., 1995). In addition, the fission-track date of the “Upper Member” has been given as 37.2 ± 1.3 Ma (around middle-late Eocene boundary) by Tsubamoto et al. (2002b). Therefore, the age of the mammalian fossils from the Pondaung Formation is most likely the latest middle Eocene (Tsubamoto et al., 2003).

Systematic paleontology

The basic dental terminologies mostly follows those of Bown and Kraus (1979), Janis and Scott (1987), Gentry and Hooker (1988), and Hooker (1989). Several dental terminologies are indicated in Figure 1. Dental measure-ments of the NMMP-KU materials are shown in the Appendix.

Figure 1

Several dental terminologies. A–B. Hsanotherium. C–F. Primitive artiodactyl (modified from Gentry and Hooker, 1988). G–H. Primitive perissodactyl (modified from Hooker, 1989). A, C, G, left upper molar; B, E, F, H, right lower molar; D, right P4. Abbreviations: brme, buccal ridge of metacone; brpa, buccal ridge of paracone; cdo, cristid obliqua; cenc, centrocrista; Df, Dorcatherium-fold; ecph, ectoloph; ecs, ectostyle; ecsd, ectostylid; end, entoconid; ensd, entostylid; hy, hypocone; hyd, hypoconid; hyld, hypoconulid; hyldb, hypoconulid basin; hyfd, hypoflexid; hyphd, hypolophid; me, metacone; med, metaconid; mel, metaconule; meph, metaloph; mes, metastyle; mesd, metastylid; mss, mesostyle; pa, paracone; pacd, paracristid (preprotocristid); pad, paraconid; pal, paraconule (protoconule); pas, parastyle; pe, pericone; Pf, Palaeomeryx-fold; pocind, postcingulid; poencd, postentocristid; pohycd, posthypocristid; pomec, postmetacrista; pomecd, postmetacristid; pomelc, postmetaconulec-rista; popalc, postparaconulecrista; poprc, postprotocrista; pr, protocone; prcd, protocristid (postprotocristid); prd, protoconid; precind, precingulid; preened, preentocristid; premecd, premetacristid; premelc, premetaconulecrista; prepac, preparacrista; prepalc, preparaconulecrista; preprc, preprotocrista; prph, protoloph; prphd, protolophid; tadb, talonid basin; tctad, transverse crest on talonid; trdb, trigonid basin; Zf, Zhailimeryx-fold.

i1342-8144-7-3-219-f01.gif

Grandorder Ungulata Linnaeus, 1766

Order and family indeterminate

Genus Hsanotherium Ducrocq et al., 2000b

Type and only known species.—Hsanotherium parvum Ducrocq et al., 2000b.

Revised diagnosis.—Primitive and very small ungulate with low-crowned teeth. Upper molars with three main cusps (paracone, metacone, and protocone), four smaller cusps (parastyle, mesostyle, hypocone, and pericone), and two much smaller cusps (paraconule and metaconule) (metastyle absent). P4 simple premolariform and mesiodis-tally elongated with large protoconid, one distal cusp (hypoconulid?), and no talonid basin. DP4 mesiodistally elongated showing trilobed aspect. Lower molars with protoconid, metaconid, hypoconid, entoconid, hypoconulid (paraconid absent), large trigonid and talonid basins, three cristids descend from hypoconulid (cristid obliqua linking to trigonid, hypolophid linking to entoconid, and posthy-pocristid linking to hypoconulid), and hypoconulid basin (a basin surrounded by hypoconulid, posthypocristid, hypoconid, hypolophid, and entoconid). M1/M1 < M2/M2 < M3/M3.

Hsanotherium parvum Ducrocq et al., 2000b

Figures 23

Figure 2

Hsanotherium parvum Ducrocq et al., in occlusal view. A–C, A′–C. NMMP-KU 0035 (Bhn 10; Holotype), a right maxillary fragment with M1–3: A, A′, M1 (stereo pair); B, B′, M2 (stereo pair); C, C′ M3 (stereo pair). D, D′. NMMP-KU 0037, a right mandibular fragment with dP4M1–2 (stereo pair). E, E′. NMMP-KU 0031 (Bhn 11), a right maxillary fragment with M2–3 (stereo pair). F, F′. NMMP-KU 0033, a right mandibular fragment with M2 (stereo pair). Scale bar = 1 cm.

i1342-8144-7-3-219-f02.gif

Figure 3

Hsanotherium parvum Ducrocq et al., A, A′, C. NMMP-KU 0032, a left mandibular fragment with M3: A, A′, occlusal view (stereo pair); C, buccal view. B, B′, D. NMMP-KU 0036, a left mandibular fragment with P4-M3: B, B′, occlusal view (stereo pair); D, buccal view. E–G. NMMP-KU 0037, a right mandibular fragment with dP4M1–2: E, buccal view; F, lingual view; G, occlusal view of dP4. Upper and middle scale bars = 2 cm (upper scale corresponds to A–B, A′–B′, and middle scale corresponds to C–F). Lower scale bar = 5 mm (corresponds to G).

i1342-8144-7-3-219-f03.gif

Ungulata indet. D, Tsubamoto et al., 2000a, p. 39-42, 73-75, pls. 2-4.

Hsanotherium parvum Ducrocq et al., 2000b, p. 727, fig. 2.

Holotype.—NMMP-KU 0035 (Bhn 10), a right upper jaw fragment with M1–3.

Type locality.—Bhl (Yarshe Kyitchaung) locality, near Bahin Village, Myaing Township, central Myanmar (Tsubamoto et al., 2000a, fig. 5; Ducrocq et al., 2000b).

Referred material.—NMMP-KU 0031 (Bhn 11), 0032, 0033, 0034, 0036, 0037. (Tsubamoto et al., 2000a; Ducrocq et al., 2000b)

Diagnosis.—As for genus.

Description of dP4.—DP4 (in NMMP-KU 0037) (Figures 2D, D′, 3E–G) is mesiodistally elongated and shows trilobed structure: posterior talonid, central “trigonid,” and anterior lobe. Talonid is roughly similar to that of molars in morphology, having hypoconid, hypoconulid, and entoconid. The presence/absence of hypoconulid basin is unknown due to the poor preservation on that part. Hypoflexid is deeper than that of molars. In central “trigonid” there is a large protoconid. At metaconid region, there appears to be three small tubercles, making a single bulge. There is a lobe anterior to central “trigonid.” This anterior lobe makes a shallow and mesiodistally elongated basin. This basin is encircled by a low paracristid. There appear to be three tiny cuspules at the anterior margin of the anterior lobe (at paraconid region). These three capsules are much lower than the other cusps. Dental enamel is thinner than that of molars.

Discussion.—The upper molar dentition, NMMP-KU 0035 (Bhn 10) and NMMP-KU 0031 (Bhn 11), was originally described with the lower dentition and assigned to an indeterminate taxon within the Ungulata by Tsubamoto et al. (2000a). Based only on these examples of upper dentition, Ducrocq et al. (2000b) erected a new genus and species, Hsanotherium parvum, but they did not mention the lower dentition.

Ducrocq et al. (2000b) placed Hsanotherium among the most primitive anthracobunids (Ungulata; Tethytheria) because the following dental characteristics are shared by Hsanotherium and other anthracobunids: an increase in molar size from M1 to M3; the buccal notch of the upper molars between parastyle and mesostyle; the existence of six distinct cusps (paracone, metacone, protocone, hypo-cone, paraconule, and metaconule); the arrangement of the anterior cusps (paracone, paraconule, and protocone) in a somewhat convex curve; the lack of ectoloph; and the relatively great anterior breadth of the molars. Based on these characteristics, Ducrocq et al. (2000b) denied the relationship of Hsanotherium with artiodactyls, perissodactyls, phenacolophids (Condylarthra), and other condylarths.

Nevertheless, further comparison of the upper molar dentitions of Hsanotherium with those of anthracobunids casts doubt on the classification of Hsanotherium as a member of the Anthracobunidae. Although the upper molars of Hsanotherium resemble those of anthracobunids in the characteristics mentioned by Ducrocq et al. (2000b), the roughly triangular occlusal outline of the upper molars of Hsanotherium is reminiscent of that of P4 of anthracobunids, rather than of their upper molars (Ducrocq et al., 2000b). The upper molars of anthracobunids are more squared (Ducrocq et al., 2000b), have much larger paraconule and metaconule than those of Hsanotherium, and lack pericone. Therefore, the assignment of Hsanotherium to the Anthracobunidae is not definitive.

On the other hand, the upper molars of Hsanotherium are distinct from those of any other ungulate mammals. They are distinguished from those of perissodactyls and phenacolophids in lacking the pi (π) shaped arrangement of the six cusps (Ducrocq et al., 2000b). Hsanotherium has a molar metaconule so tiny that it cannot be assigned to the Artiodactyla, whose molar metaconule is large. The upper molars of Hsanotherium somewhat resemble those of Ectoconus, a periptychid condylarth, in having a similar arrangement of seven cusps (paracone, metacone, protocone, hypocone, pericone, paraconule, and metaconule), as well as in having two large styles with lobes. However, the parastylar lobe in Ectoconus extends buccally to paracone, while that in Hsanotherium does so mesiobuccally. Furthermore, the other stylar lobe in Ectoconus is located buccally to metacone, while that in Hsanotherium (i.e., mesostylar lobe) is located between paracone and metacone. Hsanotherium also lacks the lingually expanded protocone base, which is one of the diagnostic characteristics of periptychids (Archibald, 1998). Moreover, in condylarths, including periptychids, M3 is smaller than M2 and the posterior part of M3 is reduced buccolingually, while in Hsanotherium, M3 is larger than M2 and the posterior part of M3 is not reduced (Ducrocq et al., 2000b). Therefore, Hsanotherium is not assigned to either the Periptychidae or the Condylarthra.

The lower dental materials (NMMP-KU 0032, 0033, 0034, 0036, 0037) described with the type specimen of Hsanotherium parvum by Tsubamoto et al. (2000a) are assigned to this species. They come from the Bahin area of the Pondaung Formation, where the upper dental materials of this species were found (Tsubamoto et al., 2000a; Ducrocq et al., 2000b). The molar size and morphology of these lower dentitions provide a good match to the upper dentitions of H. parvum: the configurations of the pericone, protocone, and hypocone on the upper molars fit the trigonid, talonid, and hypoconulid basins of the corresponding lower molars, respectively. Because of this high corre-spondence, we have identified these lower dental materials as belonging to H. parvum. The lower dental morphology provides new evidence for the phyletic relationships of Hsanotherium.

The lower dentition of Hsanotherium is distinct from that of anthracobunids. The lower molars of Hsanotherium have a large trigonid basin, and their metaconid and entoconid are distal to protoconid and hypoconid, respectively. In contrast, the lower molars of anthracobunids have no or only a very narrow trigonid basin, and their metaconid and entoconid are located almost buccally to protoconid and hypoconid, respectively, creating a bilopho-dont structure. Hsanotherium has a molar posthypocristid linking to the hypoconulid, while anthracobunids have no such molar posthypocristid. P4 of Hsanotherium is simple and premolariform, lacking both paraconid and metaconid. In contrast, that of anthracobunids is more molariform, having distinct paraconid and metaconid, triangular trigonid in occlusal view, and wider talonid.

The lower dentition of Hsanotherium is distinct from that of condylarths, whose lower molars lack not only the hypolophid linking hypoconid and entoconid directly, but also lacks the hypoconulid basin on M1–2. Furthermore, condylarths generally have a much more molariform P4 and a less elongated hypoconulid on M3 than those of Hsanotherium.

The lower dentition of Hsanotherium is distinct from that of perissodactyls, although the lower molars of Hsanotherium and primitive perissodactyls such as Hyracotherium share hypolophid and large and mesiolingually open trigonid. The molar hypoconulid of primitive perissodactyls is much lower than the hypoconid and entoconid, while that of Hsanotherium is as tall as the hypoconid and entoconid. Perissodactyls lack hypoconulid basin on M1–2, and furthermore, they have a much more molariform P4 than that of Hsanotherium, showing large metaconid and large talonid basin.

The lower dentition of Hsanotherium is also distinct from that of primitive artiodactyls, such as Diacodexis. P4 of primitive artiodactyls has a small and shallow talonid basin, which is lacking in Hsanotherium, and it does not have such a large distal cuspid (hypoconulid?) as that of Hsanotherium. The lower molars of primitive artiodactyls also have a small paraconid appressed to the large and conical metaconid and no or only a very narrow trigonid basin, again unlike Hsanotherium. Additionally, M1–2 of primitive artiodactyls lack a hypoconulid basin.

Despite such a unique lower dental morphology, the lower molars of Hsanotherium are similar to those of Gobiohyus pressidens Matthew and Granger, 1925 (AMNH 20247) (Artiodactyla; Helohyidae). The molar trigonid of G. pressidens shows a paracristid that extends down mesially, and then turns distolingually, and finally extends up distally to the tip of the metaconid, forming a relatively large and lingually opened trigonid basin and a wide trigonid angle, like that of Hsanotherium. On M2 of G. pressidens, we also find posthypocristid and hypolophid, forming a kind of hypoconulid basin.

However, the dental morphology of Hsanotherium is also distinct from that of G. pressidens. In the lower molars of G. pressidens, there is a tiny paraconid appressed to a metaconid, and metaconid is located just lingually to a protoconid, while in those of Hsanotherium, there is no trace of paraconid, and metaconid is distal to protoconid. Moreover, in the lower molars of G. pressidens, both talonid and hypoconulid basins are open lingually, and cristid obliqua is very low, unlike those of Hsanotherium. Hypoconulid on M3 is much larger in G. pressidens than in Hsanotherium. Additionally, P4 of Gobiohyus is more molariform, having a large metaconid, and its upper molars are distinct from those of Hsanotherium in having much larger metaconule and in lacking pericone.

The most interesting morphology of Hsanotherium is its dP4 structure (Figure 3G). The mesiodistally elongated and trilobed dP4 morphology of Hsanotherium is reminiscent of that of artiodactyls and macroscelideans (Luckett and Hong, 1998). First, the trilobed dP4 morphology of artiodactyls has been well known since the 19th century, and has been treated as one of the diagnostic characters of the order (Cuvier, 1822; Blainville, 1839–1864; Weber, 1928; Sudre et al., 1983; Tobien, 1985; Gentry and Hooker, 1988; Sudre and Erfurt, 1996; Luckett and Hong, 1998). However, dP4 of Hsanotherium differs from that of a very primitive artiodactyl, Diacodexis, in having mesiodistally much longer anterior lobe and three cuspules at the metaconid region (Luckett and Hong, 1998, fig. 5). It also differs from dP4 of more derived artiodactyls, such as anthracotheres and haplobunodonts, in having and three cuspules at the metaconid region and in lacking better-derived and large two cusps (protoconid- and metaconid-like, respectively) on the anterior lobe (Tobien, 1985, fig. 4; Luckett and Hong, 1998, fig. 3). Second, the mesiodistally elongated dP4 of macroscelideans was mentioned by Butler (1969, 1984) and Luckett and Hong (1998). In particular, dP4 of Hsanotherium somewhat resembles that of the Miocene macro-scelidean genus Hiwegicyon (Butler, 1969, fig. 4B, 1984, fig. 8A) in having anteriorly elongated trigonid and three cuspules at the paraconid region. However, it differs from that of Hiwegicyon and other macroscelideans in having three cuspules at the metaconid region and much smaller and lower cuspules at the paraconid region (Butler, 1969, figs. 2D, 3D, 4B, 1984, fig. 8A). It further differs from Hiwegicyon in having more tapered anterior margin (Butler, 1969, fig. 4B; Luckett and Hong, 1998). In addition, the P4 and molar morphologies of Hsanotherium are distinct from those of macroscelideans: Hsanotherium has simple premolariform P4, and large and unreduced M3/3, whereas macroscelideans have molariform P4 and reduced M3/3 (Figures 12; Butler, 1969, figs. 2-3, 1984, figs. 2-7; Hartenberger, 1986; Simons et al., 1991, figs. 1-3, 5). On the other hand, the mesiodistally elongated dP4 can be also seen in Desmostylus (Tethytheria; Desmostylidae) (Uno, 2000, figs. 2, 10), although the dental morphology of Hsanotherium is quite different from that of Desmostylus. DP4 of anthracobunids shows bilobed morphology like lower molars (West, 1980, pl. 4, fig. 5), so that it is distinct from that of Hsanotherium.

In sum, the dental morphology of Hsanotherium strongly suggests that Hsanotherium is not an anthracobunid and cannot confidently be classified into any present ungulate order, although it can be identified as belonging to the Ungulata because of its large, elongated, and posteriorly projecting hypoconulid on M3 (Prothero et al., 1988; Prothero, 1993; Nessov et al., 1998). Although the dental morphology of Hsanotherium is quite unique, its lower molars show a similarity to those of Gobiohyus pressidens (Artiodactyla; Helohyidae), and its mesiodistally elongated dP4 morphology resembles that of artiodactyls and macroscelideans. Nevertheless, the overall unique dental morphology of Hsanotherium is distinguished from that of these two orders.

Order Artiodactyla Owen, 1848

Family Dichobunidae Turner, 1849

Subfamily Homacodontinae Marsh, 1894

Genus Asiohomacodon gen. nov.

Type and only known species.—Asiohomacodon myanmarensis sp. nov.

Diagnosis.—One of the most derived homacodonts with bunoselenodont dentition and small but distinct molar paraconule. Dental size and morphology similar to those of Protoreodon parvus (a primitive protoreodontine agriochoerid oreodont) and Pentacemylus (a derived and protoreodontine-like homacodont). Differs from protoreo-dontines in having larger molar paraconule, more erect molar paracone and metacone, weaker styles, less inflated molar mesostyle, more mesiodistally compressed molar talonid, and molar postmetacristid and preentocristid both of which link straight to each other in occlusal view, and in lacking molar metastylid. Differs from Pentacemylus in having more selenodont dentition, weaker molar parastyle, more mesiodistally compressed lower molars, sharper molar preentocristid, and M3 whose posterior part is much less reduced. Differs from other homacodonts in having more selenodont dentition and in lacking any trace of molar hypocone and paraconid. Differs from Atopotherium (an agriochoerid-like anthracotheriid) in being smaller, and in having shallower mandibular corpus and less mesiodistally compressed molar trigonid and talonid.

Etymology.—Asio-: Asia, where the type specimen was collected; -homacodon: homacodonts, in which this genus is included.

Asiohomacodon myanmarensis sp. nov.

Figures 4, 5A, B

Figure 4

Asiohomacodon myanmarensis gen. et sp. nov. A, A′. NMMP-KU 0026, a right M3 in occlusal view (stereo pair). B, B′, C, C′. NMMP-KU 0713 (Holotype), a left maxillary fragment with P4–M3, in occlusal view: B, B′ P4M1 (stereo pair); C, C′, M2–3 (stereo pair). D, D′. NMMP-KU 0027, a right mandibular fragment with M2 3, in occlusal view (stereo pair). E, E′. NMMP-KU 0028, a right mandibular fragment with M3, in occlusal view (stereo pair). F, F′. NMMP-KU 0264, a left mandibular fragment with M2–3, in occlusal view (stereo pair). G, G′ H, I. NMMP-KU 0029, a right mandibular fragment with M1–2: G, G′, occlusal view (stereo pair); H, lingual view; I, buccal view. Scale bar = 2 cm.

i1342-8144-7-3-219-f04.gif

Figure 5

A, B. Asiohomacodon myanmarensis gen. et sp. nov. A, NMMP-KU 0027, a right mandibular fragment with M2–3, in buccal view; B, NMMP-KU 0028, a right mandibular fragment with M3 in lingual view. C, D, D′. Cf. Asiohomacodon myanmarensis gen. et sp. nov., NMMP-KU 0030, a right mandibular fragment with M1?: C, buccal view; D, D′, occlusal view (stereo pair). Scale bars = 2 cm (upper scale corresponds to A–C, lower scale corresponds to D, D′).

i1342-8144-7-3-219-f05.gif

?Agriochoeridae indet. E, Tsubamoto et al., 2000a, p. 45-48, 80-81, pls. 9, 10 A-C.

Holotype.—NMMP-KU 0713, a left upper jaw fragment with P4-M3 (new material).

Type locality.—Kd2 locality (21°49′24.0″N, 94°35′25.2″E), Kyudaw (near Thidon Village in Bahin area), Myaing Township, central Myanmar (Tsubamoto et al., 2000a, fig. 4).

Referred material.—NMMP-KU 0026, 0027, 0028, 0029, 0068, 0264 (Tsubamoto et al., 2000a).

New material.—NMMP-KU 0714, a right mandibular fragment with broken M3 (this specimen probably belongs to the same individual as NMMP-KU 0713).

Locality of new material.—As for the holotype.

Diagnosis.—As for genus.

Etymology.—Named after the Union of Myanmar, the country where the type specimen was collected.

Description.—The dental materials show small, primitive, brachyodont, and bunoselenodont artiodactyl condition. Mandible is as deep as in Pentacemylus (Dichobunidae; Homacodontinae) and Protoreodon (Agriochoeridae; Protoreodontinae), and is much deeper than those of primitive ruminants, such as Indomeryx and Archaeomeryx. Dental enamel is somewhat wrinkled.

P4 is premolariform, bearing two distinct cusps, paracone and protocone, and an incipient trace of hypocone at the distobuccal base of the protocone. Crown is triangular in occlusal view and is wider than long. Paracone is much larger and taller than protocone. Paracrista is distinct and somewhat selenodont shaped. Metastyle is small but distinct. Parastylar region is only weakly developed. Cingulum extends from the distolingual base of the metastylar region and surrounds the distal and lingual margins of the crown, disappearing at the mesial base of the protocone.

Upper molars are quadrilateral in occlusal view, bearing large paracone, metacone, protocone, and metaconule, and small but distinct paraconule (there is no trace of hypocone). M1–3 are very similar to one another in morphology, although metaconule on M3 is proportionally only slightly smaller and lower than that on M1–2. Upper molar sizes are: M1 < M2 = M3 Paracone and metacone are of nearly equal size and are conical and erect. Paracone has strong buccal ridge, while metacone has weaker one. Ectoloph is distinct and somewhat W-shaped in occlusal view. Parastyle is small but distinct. A strong crista from tip of parastyle extends down distally, disappearing at buccal face of paracone. Mesostyle and metastyle are only weakly developed. Lingual margin of metaconule is much more buccally located than that of protocone. Postprotocrista reaches mesial wall of metaconule, making protocone incompletely selenodont. A cingulum that originates below the mesiolingual base of the parastyle is continuous through the mesial and lingual base of the crown, disappearing at the distal base of the metaconule. Ectostyle is absent. In NMMP-KU 0713 (type), a weak buccal cingulum exist on the buccal face of the metacone, linking mesostyle and metastyle.

Lower molars show primitive selenodont morphology. M1 and M2 are almost identical in morphology, although M1 is only slightly less mesiodistally compressed than M2. M3 is also almost identical to M2 in morphology, except for having posteriorly elongated hypoconulid. The molar sizes increase gradually from M1 to M3. Trigonid is as wide as, roughly as long as, and nearly as tall as talonid. Metaconid and entoconid are somewhat buccolingually compressed, and the latter is more compressed than the former. Paraconid, metastylid, and entostylid are absent. Protocristid extends down lingually and then extends up to tip of metaconid, making a V-shaped notch between protoconid and metaconid. Cristid obliqua originates below the notch between protoconid and metaconid. Posthypocristid extends lingually, and then turns distolingually at the distobuccal base of entoconid, linking to hypoconulid. Preentocristid extends down mesially and links to postmetacristid, making a V-shaped talonid notch. Postentocristid extends down distally and stops at distal base of entoconid: it connects neither to hypoconulid nor to posthypocristid. Precingulid extends from mesial base of metaconid, disappearing at mesiobuccal base of protoconid. Buccal cingulum occurs between protoconid and metaconid, but ectostylid is absent. Lingual cingulum is absent. Postcingulid extends from distolingual base of entoconid and disappears at distal base of hypoconid on M1–2, bearing tiny hypoconulid. On M3, hypoconulid is large and posteriorly elongated, and bears single cusp and single hypoconulid loop. The inner cristid of hypoconulid loop stops just distal to postentocristid, making crenulation there. Its outer cristid connects to posthypocristid.

Discussion.—The overall dental morphology of Asiohomacodon recalls those of a derived homacodont, Pentacemylus (Bunomerycini), and Asiohomacodon is classified with the Homacodontinae. Asiohomacodon and Pentacemylus share the following characteristics: similar dental size, bunoselenodont dentition, incipient trace of hypocone on P4, small but distinct molar paraconule and mesostyle, enlarged molar metaconule, no trace of molar hypocone, no molar paraconid, neither molar metastylid nor entostylid, mesiodistally compressed molar trigonid and talonid, and deeper mandibular corpus than that of primitive ruminants. Based on these characteristics, Asiohomacodon can be referred to one of the derived homacodonts. Homacodonts have been found mostly in Eocene North America, and one genus has been recorded from Eocene Europe (McKenna and Bell, 1997). A new homacodont was recently found in the Eocene Shanghuang fauna of central China (Qi et al., 1996), but it has not been described yet.

Asiohomacodon is distinct from all other homacodonts. It differs from Pentacemylus in having a more selenodont dentition, slightly weaker molar parastyle, more mesiodistally compressed lower molars, sharper molar preentocristid, P4 whose lingual part (protocone part) is mesiodistally much shorter than the buccal part (paracone part), and M3 whose posterior part is almost unreduced (i.e., M3 metaconule is nearly as large as that of M2, and the posterior part of M3 is nearly as wide as that of M2). Asiohomacodon, with its more selenodont dentition, may be more similar to the new homacodont genus B of Stucky (1998) than to Pentacemylus, judging from Stucky's short description (1998, p. 368). In any case, Asiohomacodon differs from the new homacodont genus B in lacking any trace of molar metastylid and ectostylid. On the other hand, it differs from the other homacodonts, such as Homacodon, in having more selenodont dentition and in lacking molar hypocone and paraconid.

It is believed that Pentacemylus-like homacodonts gave rise to the Protoreodontinae (Oreodontoidea; Agriochoeridae), which occurs only in Eocene North America (Lander, 1998), so that the dental morphology of Asiohomacodon might suggest a transitional morphology from homacodonts to protoreodontines. In fact, the dental morphologies of Asiohomacodon, Pentacemylus, and the new homacodont genus B (Stucky, 1998, p. 368) are also similar to that of primitive protoreodontines such as Protoreodon parvus. However, both Asiohomacodon and Pentacemylus differ from protoreodontines in having less selenodont upper molars and smaller molar paraconule, and in lacking molar metastylid. The lack of molar metastylid in Asiohomacodon and Pentacemylus specifically indicates that these two genera are more primitive than protoreodontines and therefore cannot be assigned to the Protoreodontinae. In this respect, the new genus B, which has an incipient molar metastylid (Stucky, 1998, p. 368), is more derived and more protoreodontine-like than Asiohomacodon and Pentacemylus. Compared to Pentacemylus, Asiohomacodon is more derived and more protoreodontine-like in having more selenodont dentition and much less reduced M3 metaconule but is more primitive in having slightly smaller molar parastyle.

It is interesting to note that, in its lower molar morphology, Asiohomacodon also resembles an agriochoerid-like anthracotheriid, Atopotherium bangmarkensis, from the upper Eocene Krabi basin of Thailand (Ducrocq et al., 1996). Atopotherium is based only on the single specimen, lower mandibular fragments with right P2-M2 and left P4 (DMR TF 2908) (Ducrocq et al., 1996). Both mammals have selenodont lower molars with mesiodistally compressed trigonid and talonid. Compared to Asiohomacodon, however, Atopotherium is much larger, and has a much deeper mandibular corpus and more mesiodistally compressed molar trigonid and talonid. Atopotherium has been assigned to the Anthracotheriidae on the basis of its P4 morphology, although Atopotherium and agriochoerids are similar to each other in their lower molar morphology (Ducrocq et al., 1996). The affinity of these two genera cannot be tested because of the lack of the P4 material of Asiohomacodon.

Cf. Asiohomacodon myanmarensis sp. nov.

Figure 5C, D, D′

Cf. ?Agriochoeridae indet. E, Tsubamoto et al., 2000a, p. 48, 81, pi. 10 D-F.

Material.—NMMP-KU 0030 (Tsubamoto et al., 2000a).

Discussion.—As mentioned by Tsubamoto et al. (2000a), the dental size and morphology of the preserved molar in NMMP-KU 0030 is nearly identical to that of M1, of Asiohomacodon myanmarensis. However, it differs in the following features from A. myanmarensis: the mandibular is much more slender (it is as slender as that of primitive ruminants); the molar protocristid extends down lingually and stops at the distal base of the metaconid, whereas that of A. myanmarensis extends down lingually and then extends up to the tip of the metaconid, making a V-shaped notch.

Suborder Ruminantia Scopoli, 1777

Family indeterminate

Genus Indomeryx Pilgrim, 1928

Type species.—Indomeryx cotteri Pilgrim, 1928.

Referred species.—Indomeryx arenae Pilgrim, 1928.

Revised diagnosis.—Small and very primitive ruminant with brachyodont and bunoselenodont dentition. Upper molars with weak styles and no distinct ectostyle. Short diastema between P2 and P3. P4 relatively simple having metaconid twinned with protoconid and small talonid basin, and lacking transverse crest on talonid. Lower molars with tiny paraconid twinned with metaconid in some specimens, paracristid and slight premetacristid both not so projecting anteriorly, Zhailimeryx-fold (double preentocristid; entoconidian groove), and lingually closed hypoconulid loop on M3, and without ectostylid, Palaeomeryx-fold, distinct metastylid, and distinct Dorcatherium-fold. Differs from Archaeomeryx, Gelocus, Gobiomeryx, Notomeryx, and Prodremotherium in having Zhailimeryx-fold, less anteriorly projecting molar paracristid and premetacristid, and P4 metaconid twinned with protoconid, and in lacking transverse crest on P4 talonid. Further differs from the latter three genera (Prodremotheriidae) in having smooth and undepressed distal face of molar entoconid, and in lacking distinct molar metastylid and ectostylid and sharp molar postentocristid. Further differs from Gobiomeryx in lacking accessory cusp on M3 hypoconulid. Further differs from Archaeomeryx in having lingually closed and larger hypoconulid loop on M3. Differs from lophiomerycids in having much weaker molar styles, P4 metaconid twinned with protoconid, and molar premetacristid, and in lacking lingually opened and large molar trigonid basin. Differs from tragulids in having P4 metaconid and in lacking distinct Dorcatherium-fold. Further differs from the basal tragulid, Archaeotragulus, in having molar trigonid as wide as talonid and in lacking anteriorly projecting and stronger paracristid, and in lacking lingually opened and large molar trigonid basin. Differs from Xinjiangmeryx in having less developed upper molar cingulum and styles, and less molariform P4.

Classification of the Pondaung species.—To date, four species of Indomeryx from the Pondaung Formation have been described. Pilgrim (1928) elected the genus Indomeryx and described two species from the Pondaung Formation: the first, Indomeryx cotteri (the type species), based on a mandibular fragment with P4-M3 (GSI B768); and the second, Indomeryx arenae, based only on a mandibular fragment with M3 talonid (GSI B769) (Pilgrim, 1928). Pilgrim (1928) distinguished I. arenae from I. cotteri by the following four features: (1) its smaller size; (2) a lesser degree of concavity of the lower border of the mandibular corpus behind M3; (3) the absence of buccal cingulum on M3; and (4) molar preentocristid showing a double ridge (Zhailimeryx-fold; Figure 1; Guo et al., 2000). On the other hand, Colbert (1938) considered I. arenae to be possibly synonymous with I. cotteri by the following reasons: (1) the size differences between the specimens were not large enough to separate them into two species; (2) the difference in the shapes of the lower border of the corpus of the two species could be attributed to the difference between immature material (I. cotteri) and fully adult material (I. arenae); and (3) the difference in the degree of cingular development could be a result of individual variation. According to Colbert (1938), the only significant difference between I. arenae and I. cotteri is the molar preentocristid showing a double ridge (Zhailimeryx-fold) in the case of the former. Métais et al. (2000) has recently described two new species of Indomeryx from the Pondaung Formation, Indomeryx pilgrimi and Indomeryx minus. According to them, I. pilgrimi differs from I. cotteri in its larger size, retention of molar paraconid, a groove on the mesial side of molar entoconid (Zhailimeryx-fold), weak Dorcatherium-fold, and stronger postcingulum on M1–2; while I. minus differs from I. cotteri in its smaller size, Zhailimeryx-fold, and weak Dorcatherium-fold, and differs from I. pilgrimi again in its smaller size, thin buccal cingulum on the lower molars, and in lacking molar paraconid.

The new materials and a reexamination of previously described materials (Figures 68) lead us to conclude that there is no distinct dental morphology, except for differences in M1 size, between the four species of the Pondaung Indomeryx. Our reasons are summarized in the following five paragraphs:

  1. The double-ridged preentocristid (Zhailimeryx-fold), one of the diagnoses of I. arenae cited by Pilgrim (1928) and of I. pilgrimi and I. minus by Métais et al. (2000), exists on M1–3 of all well preserved lower dental materials of the Pondaung Indomeryx including I. cotteri (Figure 7). This structure is also seen in I. cotteri from the upper Eocene Naduo Formation of China, so that it is one of the identifying characteristics of this genus.

  2. As for the morphology of the mandibular corpus of I. arenae mentioned by Pilgrim (1928), the suggestion by Colbert (1938) is appropriate: the difference found between I. cotteri and I. arenae is due to the expected difference between immature material (I. cotteri) and fully adult material (I. arenae). Moreover, there is individual variation in the degree of concavity of the lower border of the mandibular corpus among the Pondaung Indomeryx (Figure 8).

  3. Colbert (1938) is also correct in his evaluation of the development of the cingulum in I. arenae (and also in I. pilgrimi and I. minus): its difference results from individual variation (Figure 7).

  4. The presence or absence of molar paraconid of Indomeryx mentioned by Métais et al. (2000) is also considered to depend on individual variation. Most of the Indomeryx materials lack a molar paraconid, however, in NMMP-KU 0015 (the type of I. pilgrimi), a very tiny molar paraconid can be observed on M1, though not on M2. On its M3, the mesiobuccal part of the tip of the metaconid is broken, so that it is unclear whether or not there is a paraconid. In some other materials as well, such as M3 of NMMP-KU 0013 and of AMNH 20023, a very tiny paraconid can be observed.

  5. The presence or absence of Dorcatherium-fold and metastylid of Indomeryx mentioned by Métais et al. (2000) is also considered to depend on individual variation. Most of the Indomeryx materials seem to lack both distinct molar Dorcatherium-fold and metastylid, however, in some materials, such as M1–2 of NMMP-KU 0015 and M2 of NMMP-KU 0013, there are structures which might be treated as incipient Dorcatherium-fold and metastylid. However, the structure is so indistinct that it is unclear whether it is a true Dorcatherium-fold or not.

Figure 6

A–C, A′-C′. Indomeryx cotteri Pilgrim, in occlusal view: A, A′, NMMP-KU 0008 (Mgg 2), a left maxillary fragment with M1–3 (stereo pair); B, B′ NMMP-KU 0009 (Mgg 14), a left maxillary fragment with M2–3 (stereo pair); C, C′ NMMP-KU 0010, a left maxillary fragment with dP1–4M1 (stereo pair). D, D′. Indomeryx arenae Pilgrim, NMMP-KU 0007 (Bhn 1115), a right maxillary fragment with M1–3, in occlusal view (stereo pair). E, E′. Cf. Indomeryx cotteri, NMMP-KU 0025, a right M2?, in occlusal view (stereo pair). Scale bar = 2 cm.

i1342-8144-7-3-219-f06.gif

Figure 7

A–F, A′–F′. Indomeryx cotteri Pilgrim, in occlusal view: A, A′, NMMP-KU 0015 (Bhn 911, the type of Indomeryx pilgrimi Métais et al.), a left mandibular fragment with M1–3 (stereo pair); B, B′ NMMP-KU 0019, a right mandibular fragment with P3–M3 (stereo pair); C, C′ NMMP-KU 0018, a left mandibular fragment with M2–3 (stereo pair); D, D′, NMMP-KU 0021, a right mandibular fragment with P4 (stereo pair); E, E′, NMMP-KU 0289, a right mandibular fragment with M3 (stereo pair); F, F′, NMMP-KU 0017 (Mgg 5), a right mandibular fragment with M3, (stereo pair). G, G′, H, H′. Indomeryx arenae Pilgrim, in occlusal view: G, G′, NMMP-KU 0013, a right mandibular fragment with P4–M3 (stereo pair); H, H′, NMMP-KU 0011 (Bhn 3, the type of Indomeryx minus Métais et al.), a left mandibular fragment with P4–M3 (stereo pair). Scale bar = 2 cm.

i1342-8144-7-3-219-f07.gif

Figure 8

A–D. Indomeryx cotteri Pilgrim. A, NMMP-KU 0015 (Bhn 911), a left mandibular fragment with M1–3, in buccal view; B, NMMP-KU 0289. a right mandibular fragment with M3, in buccal view; C, NMMP-KU 0019, a right mandibular fragment with P3–M3, in lingual view; D, NMMP-KU 0021, a right mandibular fragment with P4, in buccal view. E–F. Indomeryx arenae Pilgrim. E, NMMP-KU 0011 (Bhn 3), a left mandibular fragment with P4–M3, in lingual view; F, NMMP-KU 0013, a right mandibular fragment with P4–M3, in buccal view. Scale bar = 2 cm.

i1342-8144-7-3-219-f08.gif

On the other hand, the dental sizes of lower molars of the Pondaung Indomeryx materials are also variable (Figure 9). However, the M1 size can be readily divided into two groups, large (I. cotteri and I. pilgrimi) and small (I. arenae and I. minus). Because M1 has been considered to express less size variation and to correlate very closely to the body size of mammals compared to other tooth classes (Tsubamoto et al., 2002a), the dental size distribution (Figure 9) suggests that the Pondaung Indomeryx can be grouped into two size categories.

Figure 9

Size distribution of the lower molars of the Pondaung Indomeryx.

i1342-8144-7-3-219-f09.gif

In conclusion, there is no distinct dental morphology by which to distinguish the species of the Pondaung Indomeryx. The M1 sizes indicate that they can be grouped into two (large and small) categories. Although Colbert (1938) and Tsubamoto et al. (2000a) suggested that the two groups probably indicate the sexual dimorphism of a single species, it is difficult to determine this issue without knowing the morphological and size variation of the canines, so that we tentatively consider that the two categories are in-dicative of two species in this paper. The larger species is referred to as I. cotteri Pilgrim, 1928 (including I. pilgrimi Métais et al., 2000), while the smaller one is referred to as I. arenae Pilgrim, 1928 (including I. minus Métais et al., 2000).

Phyletic position.—The classification and phyletic relationships of primitive ruminants are complicated (e.g., Scott and Janis, 1993), so that the familial position of Indomeryx has been, and remains, controversial. Pilgrim (1928) erected the genus Indomeryx and tentatively assigned it to the Tragulidae. Colbert (1938) and Qiu (1978), on the other hand, assigned it to the Hypertragulidae, while Sudre (1984) and Holroyd and Ciochon (1995) treated it as a representative of the Gelocidae, which is now considered to be a polyphyletic taxon (Janis, 1987; Janis and Scott, 1987, 1988). McKenna and Bell (1997) assigned Indomeryx to the Leptomerycidae. Guo et al. (1999) described Indomeryx from the upper Eocene Naduo Formation of China and classified Indomeryx into a new family Prodremotheriidae (Pecora; Ruminantia) together with Prodremotherium, Notomeryx, and Gobiomeryx. Métais et al. (2000) assigned Indomeryx to an undetermined family within the Pecora.

Among ruminants, Indomeryx is very primitive in having bunoselenodont dentition, brachyodont teeth, relatively conical cusps, relatively simple P4, and very tiny molar paraconid in some samples. The existence of Zhailimeryx-fold distinguishes Indomeryx from all other ruminants except for primitive lophiomerycids, such as Zhailimeryx and Krabimeryx, and primitive tragulids, such as Archaeotragulus, Dorcatherium, Dorcabune, and Siamotragulus (Guo et al., 2000; Tsubamoto et al., 2000a; Métais et al., 2000, 2001). Indomeryx is phyletically close neither to Zhailimeryx from Eocene China nor to Krabimeryx from Eocene Thailand and is not assigned to the Lophiomerycidae because it lacks a lingually opened and large trigonid basin, one of the identifying characteristics of the Lophiomerycidae. Also, it is distinguished from the tragulids in having P4 metaconid and in lacking distinct Dorcatherium-fold. It is further distinguished from the basal tragulid from Eocene Thailand, Archaeotragulus, in having molar trigonid as wide as talonid and in lacking anteriorly projecting and stronger paracristid, and in lacking lingually opened and large molar trigonid basin (Métais et al., 2001). Therefore, Indomeryx is not referable to the Tragulidae. According to Métais et al. (2000, 2001), Zhailimeryx-fold is also a symplesiomorphic characteristic among Asian primitive ruminants such as lophiomerycids and tragulids. The only derived feature of Indomeryx among ruminants is a lingually closed hypoconulid loop and relatively large hypoconulid on M3, but this characteristic is seen in many primitive ruminant taxa.

Although Indomeryx shows a certain resemblance to primitive ruminants, such as Gelocus, Archaeomeryx, Gobiomeryx, Prodremotherium, and Notomeryx, in both size and dental morphology (Pilgrim, 1928; Colbert, 1938; Guo et al., 1999), it nevertheless remains distinct from these ruminants. Indomeryx is distinct from Prodremotherium, Notomeryx, and Gobiomeryx, in lacking enlarged and more distally located P4 metaconid, transverse crest on P3–4 talonid, depressed distal face of molar entoconid (double-ridged postentocristid in Notomeryx), sharp molar postentocristid, and distinct molar metastylid and ectostylid. Because of having these characteristics, these latter three genera can be grouped as the Prodremotheriidae (Guo et al., 1999). According to Guo et al. (1999), the Chinese Indomeryx (Indomeryx cotteri, the same species as one of the Pondaung Indomeryx) has some of those characteristics of the Prodremotheriidae and is referable to this family. However, as mentioned above, the Pondaung Indomeryx lacks those distinct prodremotheriid characteristics and is so primitive that it cannot be referred to the family with confidence. On the other hand, Indomeryx is distinguished from Archaeomeryx in having a lingually closed hypoconulid loop on M3, and therefore the former is more derived than the latter on that point. However, Indomeryx is more primitive than Archaeomeryx and also Gelocus, Notomeryx, Prodremotherium, and pecorans in lacking a transverse crest on P3–4 talonid and enlarged and more distally located P4 metaconid. The dental morphology of Indomeryx shows many primitive characteristics among ruminants and lacks any critical derived features referable to any ruminant family.

Indomeryx cotteri Pilgrim, 1928

Figures 6A–C, A′–C, 7A–F, A′–F, 8A–D

Indomeryx cotteri Pilgrim, 1928, p. 33-35, pl. 4, figs. 7, 7a, 9, 9a; Colbert, 1938, p. 393-397, figs. 55-56; Tsubamoto et al., 2000a (in part), p. 49-53, 82-85, pls. 11A-C, G-I, 12G-I, 13, 14B-D.

Undescribed ruminant, Matthew, 1929, p. 516, fig. 40.

Indomeryx pilgrimi Métais et al., 2000, p. 808-810, fig. 2A-C.

Holotype.—GSI B768, a right mandibular corpus with P4-M3.

Type locality.—Seven furlongs E.S.E. of Sinzwe Village (in Bahin area), Myaing Township, central Myanmar (Pilgrim, 1928; Tsubamoto et al., 2000a, figs. 4-5).

Referred material.—AMNH 20023, 32521; NMMP-KU 0008 (Mgg 2), 0009 (Mgg 14), 0010, 0015 (Bhn 911), 0016, 0017 (Mgg 5), 0018, 0019, 0020, 0021, 0022, 0024, 0201, 0266, 0289, 0290; LK 6. (Colbert, 1938; Tsubamoto et al., 2000a; Métais et al., 2000)

New material.—NMMP-KU 0716, a left mandibular fragment with M1–2, upper and lower molar fragments, and bone fragments; NMMP-KU 0720, right and left maxillary fragments with right and left M1–2.

Locality of new material.—NMMP-KU 0716 is from Kd2 locality (21°49′24.0″N, 94°35′25.2″E), Kyudaw (near Thidon Village in Bahin area), Myaing Township, central Myanmar; NMMP-KU 0720 is from PA1 locality (21°46′24.0″N; 94°36′30.8″E) near Sinzwe Village (in Bahin area), Myaing Township, central Myanmar (Tsubamoto et al., 2000a, figs. 4-5).

Revised diagnosis.—Body size larger than I. arenae based on the size of M1.

Discussion.—Qiu (1978, p. 9, line 13 from the bottom) mentioned that the lower dental materials of Indomeryx cotteri described and figured by Colbert (1938, p. 394, fig. 55), AMNH 20023 and 32521, belong to Notomeryx besensis (Ruminantia), which was originally described from the upper Eocene Naduo Formation of south China. However, the sizes and dental morphologies of AMNH 20023 and 32521 are identical to that of I. cotteri, and definitively differ from those of Notomeryx from China (Tsubamoto et al., 2000a). On the other hand, AMNH 32521 was described as a left mandibular fragment with M1–2 of Indomeryx cotteri by Colbert (1938). However, the posterior part of its posterior molar is broken, so that it is difficult to determine whether the posterior molar is M2 or M3 based only on its morphology. On the basis of size the two molars preserved in AMNH 32521 are M2 and M3.

Cf. Indomeryx cotteri Pilgrim, 1928

Figure 6E, E′

Cf. Indomeryx cotteri Pilgrim, 1928. Tsubamoto et al., 2000a, p. 53-54, 85, pl. 14E.

Material— NMMP-KU 0025 (Tsubamoto et al., 2000a).

Discussion.—As described by Tsubamoto et al. (2000a), the size and morphology of NMMP-KU 0025 are nearly identical to those of M2 of Indomeryx cotteri, except that: paracone and metacone are more conical; overall shape in occlusal view is less diagonal and somewhat wider; and a distinct buccal cingulum buccal to metacone exists. Also, in this specimen, ectoloph is less developed, and protocone and metaconule are more distally located (just lingually to protocone and metacone, respectively) compared to Indomeryx cotteri.

Indomeryx arenae Pilgrim, 1928

Figures 6D, D′, 7G, G′, H, H′, 8E–F

Indomeryx arenae Pilgrim, 1928, p. 35-36, pl. 4, figs. 10, 10a; Colbert, 1938, p. 393-397, fig. 55-56.

Indomeryx cotteri Pilgrim, 1928 (in part). Tsubamoto et al., 2000a (in part), p. 49-53, 82-83, 85, pls. 11D-F, J-L, 12A-F, 14A.

Indomeryx minus Métais et al., 2000, p. 810, fig. 2D-F.

Holotype.—GSI B769, a left mandibular fragment with talonid of M3.

Type locality.—0.25 mile west of Pangan Village (probably PGN2), Myaing Township, central Myanmar (Pilgrim, 1928; Tsubamoto et al., 2000a, fig. 6).

Referred material.—NMMP-KU 0007 (Bhn 1115), NMMP-KU 0011 (Bhn 3), 0012 (Bhn 4), 0013, 0014 (Bhn 6), 0222; LK 3. (Tsubamoto et al., 2000a; Métais et al., 2000)

New material.—NMMP-KU 0669, a left mandibular fragment with M2–3.

Locality of new material.—Bhl (Yarshe Kyitchaung) locality, near Bahin Village, Myaing Township, central Myanmar (Tsubamoto et al., 2000a, fig. 5).

Revised diagnosis.—Body size smaller than I. cotteri based on the size of M1.

Order Perissodactyla Owen, 1848

Suborder Tapiromorpha Haeckel, 1866 sensu Froehlich, 1999

Family Indolophidae Schoch, 1984

Genus Indolophus Pilgrim, 1925

Type and only known species.—Indolophus guptai Pilgrim, 1925.

Revised diagnosis.—Medium-sized and primitive tapiromorph. P2–4 with triangular aspect in occlusal view, large and subcorneal paracone and metacone, large and centrally placed protocone with high and acute preprotocrista, and very small parastyle, and lacking postprotocrista, metaconule, and hypocone: P2 longer than wide with conical protocone and an incipient paraconular bulge; P3–4 wider than long with large, buccolingually flattened, and posteriorly elongated protocone, and lacking paraconule. Upper molars with conical paracone and metacone, somewhat lophodont protocone and hypocone, small parastyle, straight centrocrista, and low and weakly developed ectoloph, lacking paraconule and metaconule. M3 posterior part reduced buccolingually with somewhat reduced metacone. Four lower premolars with a diastema between C1 and P1. P1 one-rooted. P4 nearly as large as M1. Lower molars with small hypoconulid on M1–2, distinct hypolophid, and distinct cristid obliqua originating below the notch between protoconid and metaconid, lacking metastylid. Differs from other tapiromorphs in having smaller parastyle on the upper dentition and very large and centrally placed P2–4 protocone with a strong and sharp preprotocrista and without postprotocrista. Further differs from ceratomorphs and Kalakotia in having less lophodont dentition, more conical cusps, and low and weak molar ectoloph. Differs from primitive hippomorphs (such as Hyracotherium and Propalaeotherium) and further differs from basal tapiromorphs (Orientolophus, Cymbalophus, and Systemodon) in having more developed molar protolophid and hypolophid, and smaller hypoconulid on M1–2, and in lacking lingual and buccal cingulum and molar paraconule, metaconule, and metastylid.

Indolophus guptai Pilgrim, 1925

Figure 10

Figure 10

Indolophus guptai Pilgrim. A, A′. NMMP-KU 0265, a left M3, in occlusal view (stereo pair). B, B′. NMMP-KU 0040, a left mandibular fragment with M2, in occlusal view (stereo pair). C–G, C′–D′. NMMP-KU 0623, a left mandible with P4–M2: C, C′ P4M1, in occlusal view (stereo pair); D, D′, M2, in occlusal view (stereo pair); E, occlusal view; F, lingual view; G, buccal view. Scale bars = 2 cm (upper scale corresponds to A–D, A′–D′, lower scale corresponds to E–G).

i1342-8144-7-3-219-f10.gif

Indolophus guptai Pilgrim, 1925, p. 22-25, pl. 2, figs. 8a-8d; Matthew, 1929, p. 515, fig. 39; Colbert, 1938, p. 346-348, fig. 39; Radinsky, 1965, p. 235-236, fig. 22, pl. 4, fig. 6; Tsubamoto et al., 2000a, p. 56-59, 92, pl. 21.

Holotype.—GSl C347, a left maxilla with P2–4M1.

Type locality.—1.25 miles north of Konywa Village (in Mogaung area), Palé Township, central Myanmar (Pilgrim, 1925; Tsubamoto et al., 2000a, fig. 4).

Referred material.—NMMP-KU 0040, 0041, 0265 (Tsubamoto et al., 2000a).

New material.—NMMP-KU 0623, a left mandible with P4-M2.

Locality of new material.—Pk2 locality, near Paukkaung Village, Myaing Township, central Myanmar (Tsubamoto et al., 2000a, fig. 5).

Diagnosis.—As for genus.

Description of new material.—NMMP-KU 0623 is a subadult specimen, of which M3 (this tooth was not collected) was probably unerupted. Mandible is slightly more slender than that of NMMP-KU 0040. Five alveoli are preserved anterior to P4, indicating one rooted P1 and two rooted P2–3. There is a diastema between C1 and P1. This diastema seems to be relatively long, judging from the broken anterior margin of the mandible and broken alveolus for C1. P1 alveolus is oval in occlusal view, elongating distolingually. P2 alveoli are conical in occlusal view, suggesting P2 is premolariform. Anterior P3 alveolus is conical, and posterior one is somewhat widened buccolingually, suggesting that P3 talonid is somewhat molariform. Judging from preserved anterior part of mandible, posterior margin of mandibular symphysis seems to be anterior to P1 alveolus. There are four mental foramina: the largest one is located below anterior P2 alveolus; two are located below the posterior P2 alveolus; and remaining one is located near anteriormost part of the preserved mandible.

P4 is nearly as large as M1 and is molariform but is distinct from M1 in having mesiodistally oriented (more buccally oriented) paracristid and lower entoconid, and in lacking hypoconulid and hypolophid.

M1–2 shows a typical primitive tapiromorph structure with some lophodonty and without paraconid and metastylid. Metaconid and entoconid are distal to protoconid and hypoconid, respectively. Small hypoconulid exists on postcingulid and links neither to hypoconid nor to entoconid. Paracristid extends down mesiolingually from protoconid. Protocristid (protolophid) makes a notch between protoconid and metaconid. Trigonid basin is open lingually. Posterior trigonid wall is nearly perpendicular to the mandibular extension in occlusal view, and vertically diagonal in lateral view. Talonid is as wide as trigonid. Cristid obliqua originates below the notch between protoconid and metaconid. Hypolophid is not notched and extends slightly distolingually. Talonid basin is open lingually. Lingual and buccal cingulids are absent. Precingulid extends from mesial base of trigonid, disappearing at mesiobuccal base of protoconid. M2 is larger than M1.

Identification of NMMP-KU materials.—The morphology of M3 specimen, NMMP-KU 0265, is almost identical to that of M1 of the type of Indolophus guptai: conical paracone and metacone, straight and low centrocrista, lophodont protocone and hypocone, relatively small parastyle, no paraconule, and no metaconule. This M3 material is slightly larger in size than M1 of Indolophus guptai and this is congruent with the primitive perissodactyl condition. The morphology of the lower dental materials, NMMP-KU 0040, 0041, and 0623, is referable to primitive tapiromorphs, such as Isectolophus and Orientolophus, and their size is congruent with the upper dentition of Indolophus guptai. The cusp morphology and lophid configuration of these lower dental materials provide a good match to the upper dentitions of Indolophus guptai: the cusps display a slightly conical aspect and the directions of protolophid and hypolophid are congruent with those of protoloph and metaloph, respectively. Because of these high correspondences, we have identified these NMMP-KU materials as belonging to Indolophus guptai.

Phyletic position.—The phyletic position of Indolophus has been uncertain because of the unique morphology of its upper dentition (Pilgrim, 1925; Matthew, 1929; Colbert, 1938; Radinsky, 1963, 1965, 1969; Schoch, 1984, 1989). Pilgrim (1925) assigned Indolophus into the Tapiridae, while Matthew (1929) and Colbert (1938) assigned it to the Isectolophidae (= Parisectolophidae). Both of these researchers believed that Indolophus was closely related to the North American isectolophids, such as Homogalax and Isectolophus. Radinsky (1963, 1965, 1969) reported that the differences between Indolophus and the isectolophids were great enough to remove Indolophus from the Isectolophidae. He also mentioned that Indolophus resembled cf. Breviodon acares (AMNH 81751) (Lophialetidae) from the middle Eocene of China, although the phyletic relationship between them was uncertain (Radinsky, 1965). In any case, all these researchers considered Indolophus to be a primitive “tapiroid” (= primitive tapiromorph). On the other hand, Schoch (1984, 1989) considered Indolophus to be not a tapiromorph but a hippomorph, and Schoch (1984) erected a new family Indolophidae for Indolophus.

Indolophus is a tapiromorph (sensu Hooker, 1984, 1989, and Froehlich, 1999), not a hippomorph. Schoch (1984, 1989) suggested that the M1 metaloph of Indolophus is very low at the ectoloph and this characteristic is referable to the hippomorphs. However, this characteristic is also seen in the type specimen of the basal tapiromorph, Orientolophus (Ting, 1993, fig. 4A). Therefore, SchocH′s (1984, p. 16, 1989, p. 312-313) suggestion of a hippomorph affinity for Indolophus is invalid. Indolophus differs from primitive hippomorphs, such as Hyracotherium and Propalaeo-therium, in having more buccally oriented P4 paracristid and more developed molar lophs and lophids, and in lacking lingual and buccal cingulum and molar paraconule, metaconule, and metastylid. Judging from these characteristics (Ting, 1993), Indolophus is assigned into the Tapiromorpha.

Among tapiromorphs, Indolophus displays relatively primitive characteristics in having rather conical cusps, straight and low centrocrista, small parastyle, and distinct and diagonally oriented cristid obliqua. Therefore, Indolophus lacks any derived features of primitive cerato-morphs, such as helaletids, lophialetids, and deperetellids, and is not assigned to the Ceratomorpha (sensu Froehlich, 1999). However, Indolophus is more derived than the basal tapiromorphs (Orientolophus, Cymbalophus, and Systemodon) (Ting, 1993; Froehlich, 1999; Maas et al., 2001) in having larger size, more developed molar protolophid and hypolophid, and smaller hypoconulid on M1–2, and in lacking lingual and buccal cingulum and molar paraconule, metaconule, and metastylid.

Indolophus is distinct from other tapiromorphs in having smaller parastyle on the upper dentition and a unique P2–4 morphology: paracone and metacone are very large and conical; protocone is centrally placed with distinct and strong preprotocrista lacking postprotocrista; protocone is relatively large and conical on P2 and is very large, buccolingually flattened, and posteriorly elongated on P3–4 (Radinsky, 1965). These characteristics indicate that Indolophus is not assignable into the North American isectolophids (Radinsky, 1963, 1965; Schoch, 1989). Among primitive tapiromorphs, however, the upper premolar dentition of Orientolophus from China is not known (Ting, 1993), so that the possibility of a close phyletic relationship between Indolophus and Orientolophus still remains.

The molar morphology of Indolophus shows a mosaic structure with reference to the primitive tapiromorphs. In lacking molar paraconule and metaconule, it is more derived than that of some isectolophids [Cardiolophus, Homogalax, Sastriolophus, Karagalax, and unnamed isectolophid from the Wutu basin of China (Tong and Wang, 1998)], a questionable very primitive lophialetid from China (Ampholophus Wang and Tong, 1996), and basal tapiromorphs. In lacking metastylid, it is also more derived than Cardiolophus, Homogalax, and basal tapiromorphs. On the other hand, in having less sharp and low molar ectoloph with no or only slightly developed postmetacrista, it is more primitive than that of some isectolophids (Isectolophus, Sastriolophus, and Karagalax), Kalakotia (including Aulaxolophus) from Indo-Pakistan (Ranga Rao, 1972), and ceratomorphs. Such mosaic characteristics make it difficult to determine the phyletic position of Indolophus among tapiromorphs.

Hooker (1989) performed a cladistic analysis of primitive tapiromorphs and indicated that the isectolophids are placed at the base of the tapiromorph lineage, more basally so than the ancylopods (chalicotherioids and lophiodonts). Froehlich (1999) also performed a cladistic analysis of the primitive tapiromorphs and further indicated that isectolophids are paraphyletic. However, Hooker (1989) did not include some important primitive Asian Eocene tapiromorphs such as Sastrilophus from Indo-Pakistan (Sahni and Khare, 1971), Homogalax wutuensis from China (Chow and Li, 1965), and Indolophus from Myanmar (and also the later-described taxa including Orientolophus Ting, 1993 from China); while Froehlich (1999) also did not include Sastrilophus, Homogalax wutuensis, Indolophus, and other important Asian Eocene tapiromorphs such as Kalakotia from Indo-Pakistan, Ampholophus from China, lophialetids, and deperetellids [and also the later-described taxa such as Karagalax Maas et al., 2001, unnamed isectolophid from China (Tong and Wang, 1998), and unnamed primitive tapiromorph from the Akasaki Formation of Japan (Miyata and Tomida, 1998)] (Maas et al., 2001). We could determine the phylogenetic relationship of Indolophus among the primitive tapiromorphs by undertaking a cladistic analysis combining all the known anatomical characteristics of the above-mentioned primitive tapiromorphs, but the lack of materials representing important characteristics for some taxa, such as M3 of Indolophus and premolar dentition of Orientolophus, might obstruct a determination of the phyletic position of Indolophus.

Infraorder Ceratomorpha Wood, 1937 sensu Froehlich, 1999

Ceratomorpha fam., gen. et sp. indet.

Figure 11

Figure 11

Ceratomorpha fam., gen. et sp. indet., NMMP-KU 0058, a left maxillary fragment with a tooth which is assumed here as P3 in occlusal view (stereo pair). Scale bar = 2 cm.

i1342-8144-7-3-219-f11.gif

Material.—NMMP-KU 0058, a left maxillary fragment with a lingual half of a tooth (P3?).

Locality.—Pk2 locality, near Paukkaung Village (in Bahin area), Myaing Township, central Myanmar (Tsubamoto et al., 2000a, fig. 5).

Description.—Only a lingual part of a tooth is preserved in this left maxillary fragment. We will assume this preserved tooth to be P3 here for the purpose of description. Judging from P3 and preserved alveoli for other broken teeth, there were originally at least two teeth anterior to P3 without diastema and there were also at least two teeth posterior to it: P2 is narrower and shorter than P3 and P1 is much narrower than P2; and P4 is wider and longer than P3 Infraorbital foramen is located just above the anterior part of P3.

P3 is wider than long and shows bilophodont structure with complete protoloph and metaloph, which is reminiscent of ceratomorph molars. Both lophs are separated lingually by a groove, and not perpendicular but somewhat diagonal to the tooth row. Protoloph is slightly more diagonal to the tooth row than metaloph at the lingual part. Mesial cingulum exists and continuous to lingual cingulum, disappearing at mesiolingual base of metaloph. Distal cingulum originates at distolingual base of metaloph. Buccal structure cannot be described because this part is broken.

Discussion.—Although the present material is poorly preserved, the dental morphology and size of the preserved tooth is not identical to any other ceratomorph species described from the Pondaung Formation to date. This material indicate an occurrence of an additional ceratomorph species in the Pondaung fauna.

Acknowledgments

We are grateful to Brigadier General Than Tun and Major Bo Bo (International Affairs and Research Department, Office of the Chief of Military Intelligence, Ministry of Defence, Myanmar), to the personnel of the Myanmar-Japan (Kyoto University) Joint Fossil Expedition Team, and to the curators of the National Museum of Myanmar for their guidance and help in the field and museum. P.A. Holroyd (UCMP), M.C. McKenna (AMNH), J. Alexander (AMNH), K.C. Beard (CM), D. Haldar (GSI), Y. Chaimanee (DMR), and Y. Tomida (NSM) helped us in working at their respective institutions, and engaged in critical discussions. This manuscript was improved by T. Kase (NSM) and two anonymous referees. This research was supported by the MEXT Overseas Scientific Research Funds (No. 14405019 to N. Shigehara), the MEXT Grant-in-Aid for the 21st Century COE Program (A2 to Kyoto University, Leader T. Nishida), and the JSPS Research Fellowships for Young Scientists (No. 04836 to T. Tsubamoto).

Appendices

Appendix.

Dental measurements (in mm) of the NMMP-KU specimens studied in this paper. Abbreviations: L, anteroposterior length; W, buccolingual width; TRW, trigonid with; TAW, talonid width; *, estimate.

i1342-8144-7-3-219-ta01.gif

References

1.

J. D. Archibald 1998. Archaic ungulates (“ondylarthra”). In C. M. Janis, et al , editor. eds.. Evolution of Tertiary Mammals of North America. Volume I: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. p. 292–331.Cambridge University Press. Cambridge. Google Scholar

2.

Aung Naing Soe 1999. Sedimentary facies of the upper part of the Pondaung Formation (in central Myanmar) bearing late Middle Eocene anthropoid primates. In. Pondaung Fossil Expedition Team ed.. Proceedings of the Pondaung Fossil Expedition Team152–178.Office of Strategic Studies, Ministry of Defence. Yangon. Google Scholar

3.

Aung Naing Soe, Myitta, Soe Thura Tun, Aye Ko Aung, Tin Thein, B. Marandat, S. Ducrocq, and J-J. Jaeger . 2002. Sedimentary facies of the late Middle Eocene Pondaung Formation (central Myanmar) and the palaeoenvironments of its anthropoid primates. Comptes Rendus Palevol vol. 1:p. 153–160. Google Scholar

4.

Aye Ko Aung 1999. Revision on the stratigraphy and age of the primates-bearing Pondaung Formation. In. Pondaung Fossil Expedition Team ed.. Proceedings of the Pondaung Fossil Expedition Team131–151.Office of Strategic Studies, Ministry of Defence. Yangon. Google Scholar

5.

Ba Maw, R. L. Ciochon, and D. E. Savage . 1979. Late Eocene of Burma yields earliest anthropoid primate, Pondaungia cotteri. Nature vol. 282:p. 65–67. Google Scholar

6.

F. Bender 1983. Geology of Burma. viii + 293. Gebrüder Borntraeger. Berlin. Google Scholar

7.

W. A. Berggren, D. V. Kent, C. C. Swisher III, and M-P. Aubry . 1995. A revised Cenozoic geochronology and chronostrati-graphy. In W. A. Berggren, et al , editor. eds.. Geochronology, Time Scales and Global Stratigraphic Correlation. SEPM Spec. Publ. #54. 129–212.SEPM. Oklahoma. Google Scholar

8.

H. M. D. de Blainville 1839–1864. Ostéographie ou Description Iconographique Comparée du Squelette et du Systéme Dentaire des Mammiféres Récents et Fossiles pour Servir de Base á la Zoologie et á la Géologie. Tome Quatriéme, J.B. Bailliére et Fils. Paris. Google Scholar

9.

T. M. Bown and M. J. Kraus . 1979. Origin of the tribosphenic molar and metatherian and eutherian dental formulae. In J. A. Lillegraven, Z. Kielan-Jaworowska, and W. A. Clemens , editors. eds.. Mesozoic Mammals: The First Two-Thirds of Mammalian History. p. 172–181.University of California Press. Berkeley. Google Scholar

10.

P. M. Butler 1969. Insectivores and bats from the Miocene of East Africa: New material. In L. S. B. Leakey , editor. ed.. Fossil Vertebrates of Africa. p. 1–37.Academic Press. New York. Google Scholar

11.

P. M. Butler 1984. Macroscelidea, Insectivora, and Chiroptera from the Miocene of East Africa. Palaeovertebrata vol. 14/no. 3:p. 117–200. Google Scholar

12.

Y. Chaimanee, Tin Thein, S. Ducrocq, Aung Naing Soe, M. Benammi, Than Tun, Thit Lwin, San Wai, and J-J. Jaeger . 2000. A lower jaw of Pondaungia cotteri from the Late Middle Eocene Pondaung Formation (Myanmar) confirms its anthropoid status. Proceedings of the National Academy of Sciences of the United States of Americavol. 97/no. 8:p. 4102–4105. Google Scholar

13.

R. L. Ciochon, P. D. Gingerich, G. F. Gunnell, and E. L. Simons . 2001. Primate postcrania from the late middle Eocene of Myanmar. Proceedings of the National Academy of Sciences of the United States of Americavol. 98:p. 7672–7677. Google Scholar

14.

R. L. Ciochon, D. E. Savage, Thaw Tint, and Ba Maw . 1985. Anthropoid origins in Asia? New discovery of Amphipithecus from the Eocene of Burma. Science vol. 229:p. 756–759. Google Scholar

15.

M. Chow and C. Li . 1965. Homogalax and Heptodon of Shantung. Vertebrata PalAsiatica vol. 9/no. 1:p. 15–21. pl. 1. (in Chinese with English summary). Google Scholar

16.

E. H. Colbert 1937. A new primate from the upper Eocene Pondaung Formation of Burma. American Museum Novitates no. 951:p. 1–18. Google Scholar

17.

E. H. Colbert 1938. Fossil mammals from Burma in the American Museum of Natural History. Bulletin of the American Museum of Natural History vol. 74:p. 255–436. Google Scholar

18.

G. Cuvier 1822. Recherches sur les ossemens fossiles. Nouvelle edition. vol. 3:p. 398–403.d'Ocagne. Paris. Google Scholar

19.

S. Ducrocq, Aung Naing Soe, Aye Ko Aung, M. Benammi, Bo Bo, Y. Chaimanee, Than Tun, Tin Thein, and J-J. Jaeger . 2000a. A new anthracotheriid artiodactyl from Myanmar, and the relative ages of the Eocene anthropoid primate-bearing localities of Thailand (Krabi) and Myanmar (Pondaung). Journal of Vertebrate Paleontology vol. 20/no. 4:p. 755–760. Google Scholar

20.

S. Ducrocq, Aung Naing Soe, Bo Bo, M. Benammi, Y. Chaimanee, Than Tun, Tin Thein, and J-J. Jaeger . 2000b. First record of an Anthracobunidae (Mammalia, TTethytheria) from the Eocene of the Pondaung Formation, Myanmar. Comptes Rendus de l'Académie des Sciences, Paris, Sciences de la Terre et des Planètes/Earth and Planetary Sciences vol. 330:p. 725–730. Google Scholar

21.

S. Ducrocq, Y. Chaimanee, V. Suteethorn, and J-J. Jaeger . 1996. An unusual anthracotheriid artiodactyl from the late Eocene of Thailand. Neues Jahrbuch für Geologie und Paläontologie Monatshefte H.7. 389–398. Google Scholar

22.

N. Egi and T. Tsubamoto . 2000. A preliminary report on carnivorous mammals from Pondaung fauna. In. Oversea Research Report of Studies on “Origin and Evolution of Hominoids in East Asia”. p. 103–114. 1997-99 Grant-in-Aid for Scientific Research Reports for Oversea Scientific Survey, “Asian Paleoprimatology, vol. 1, 2000”. Primate Research Institute. Kyoto University, Japan. Google Scholar

23.

D. J. Froehlich 1999. Phylogenetic systematics of basal perissodactyls. Journal of Vertebrate Paleontology vol. 19/no. 1:p. 140–159. Google Scholar

24.

D. L. Gebo, G. F. Gunnell, R. L. Ciochon, M. Takai, T. Tsubamoto, and N. Egi . 2002. New eosimiid primate from Myanmar. Journal of Human Evolution vol. 43/no. 4:p. 549–533. Google Scholar

25.

A. W. Gentry and J. J. Hooker . 1988. The phylogeny of the Artiodactyla. In M. J. Benton , editor. ed.. The Phylogeny and Classification of the Tetrapods, Volume 2: Mammals. p. 235–272.Clarendon Press. Oxford. Google Scholar

26.

G. F. Gunnell, R. L. Ciochon, P. D. Gingerich, and P. A. Holroyd . 2002. New assessment of Pondaungia and Amphipithecus (Primates) from the late middle Eocene of Myanmar, with a comment on ‘Amphipithecidae’. Contributions from the Museum of Paleontology, the University of Michigan vol. 30/no. 13:p. 337–372. Google Scholar

27.

J. Guo, M. R. Dawson, and K. C. Beard . 2000. Zhailimeryx, a new lophiomerycid artiodactyl (Mammalia) from the late middle Eocene of Central China and the early evolution of ruminants. Journal of Mammalian Evolution vol. 7/no. 4:p. 239–258. Google Scholar

28.

J. Guo, T. Qi, and H. Sheng . 1999. A restudy of the Eocene ruminants from Baise and Yongle basins, Guangxi, China, with a discussion of the systematic positions of Indomeryx, Notomeryx, Gobiomeryx and Prodremotherium. Vertebrata PalAsiatica vol. 37/no. 1:p. 18–39. in Chinese with English summary. Google Scholar

29.

E. Haeckel 1866. Generelle Morphologie der Organismen. clx + 462. 8 pls. Georg Reimer. 2.Berlin. Google Scholar

30.

J-L. Hartenberger 1986. Hypothèse paléontologique sur l'origine des Macroscelidea (Mammalia). Comptes Rendus de l'Académie des Sciences, Paris, Sér. II vol. 302:p. 247–249. Google Scholar

31.

Hla Mon 1999. Nanopaleontological analysis of the rock samples collected by the Pondaung Fossils Expedition Team. In. Pondaung Fossil Expedition Team ed.. Proceedings of the Pondaung Fossil Expedition Team94–111.Office of Strategic Studies. Ministry of Defence, Yangon. Google Scholar

32.

P. A. Holroyd and R. L. Ciochon . 1994. Relative ages of Eocene primate-bearing deposits of Asia. In J. G. Fleagle and R. F. Kay , editors. eds.. Anthropoid Origins. p. 123–141.Plenum Press. New York. Google Scholar

33.

P. A. Holroyd and R. L. Ciochon . 1995. A new artiodactyl (Mammalia) from the Eocene Pondaung Sandstones, Burma. Annals of Carnegie Museum vol. 64/no. 3:p. 177–183. Google Scholar

34.

P. A. Holroyd and R. L. Ciochon . 2000. Bunobrontops savagei: a new genus and species of brontotheriid perissodactyl from the Eocene Pondaung fauna of Myanmar. Journal of Vertebrate Paleontology vol. 20/no. 2:p. 408–410. Google Scholar

35.

J. J. Hooker 1984. A primitive ceratomorph (Perissodactyla, Mammalia) from the early Tertiary of Europe. Zoological Journal of Linnean Society of London vol. 82:p. 229–244. Google Scholar

36.

J. J. Hooker 1989. Character polarities in early perissodactyls and their significance for Hyracotherium and infraordinal relationships. In D. R. Prothero and R. M. Schoch , editors. eds.. The Evolution of Perissodactyls. p. 79–101.Oxford University Press. Oxford. Google Scholar

37.

J. H. Hutchison and P. A. Holroyd . 1996. Preliminary report on the lower vertebrate fauna of the late middle Eocene Pondaung Sandstones. Journal of Vertebrate Paleontology 16/3:43A. Google Scholar

38.

J-J. Jaeger, Tin Thein, M. Benammi, Y. Chaimanee, Aung Naing Soe, Thit Lwin, Than Tun, San Wai, and S. Ducrocq . 1999. A new primate from the middle Eocene of Myanmar and the Asian early origin of anthropoids. Science vol. 286:p. 528–530. Google Scholar

39.

J-J. Jaeger, U. Aung Naing Soe, U. Aye Ko Aung, M. Benammi, Y. Chaimanee, R-M. Ducrocq, Than Tun, U. Tin Thein, and S. Ducrocq . 1998. New Myanmar middle Eocene anthropoids. An Asian origin for catarrhines. Comptes Rendus de l'Académie des Sciences, Sér. Ill Sciences de la Vie/Life Sciences vol. 321:p. 953–959. Google Scholar

40.

C. M. Janis 1987. Grades and clades in hornless ruminant evolution: the reality of the Gelocidae and the systematic position of Lophiomeryx and Bachitherium. Journal of Vertebrate Paleontology vol. 7/no. 2:p. 200–216. Google Scholar

41.

C. M. Janis and K. M. Scott . 1987. The interrelationships of higher ruminant families with special emphasis on the members of Cervoidea. American Museum Novitates no. 2893:p. 1–85. Google Scholar

42.

C. M. Janis and K. M. Scott . 1988. The phylogeny of the Ruminantia (Artiodactyla, Mammalia). In M. J. Benton , editor. ed.. The Phylogeny and Classification of the Tetrapods, Volume 2: Mammals. p. 273–282.Clarendon Press. Oxford. Google Scholar

43.

B. Lander 1998. Oreodontoidea. In C. M. Janis, K. M. Scott, and L. L. Jacobs , editors. eds.. Evolution of Tertiary Mammals of North America. Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. p. 402–420.Cambridge University Press. Cambridge. Google Scholar

44.

C. Linnaeus 1766. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species cum characteribus, differentiis, synonymis, locis. Editio duodecimo, reformata. 532. p. Stockholm. Laurentii Salvii. 1.pt. 1. Google Scholar

45.

W. P. Luckett and N. Hong . 1998. Phylogenetic relationships between the orders Artiodactyla and Cetacea: A combined assessment of morphological and molecular evidence. Journal of Mammalian Evolution vol. 5/no. 2:p. 127–182. Google Scholar

46.

M. C. Maas, S. T. Hussain, J. J. M. Leinders, and J. G. M. Thewissen . 2001. A new isectolophid tapiromorph (Perissodactyla, Mammalia) from the early Eocene of Pakistan. Journal of Paleontology vol. 75/no. 2:p. 407–417. Google Scholar

47.

O. C. Marsh 1894. Description of Tertiary artiodactyles. American Journal of Science ser. 3vol. 48:p. 259–274. Google Scholar

48.

W. D. Matthew 1929. Critical observations upon Siwalik mammals. Bulletin of the American Museum of Natural History 56:art. 7. 437–560. Google Scholar

49.

W. D. Matthew and W. Granger . 1925. New mammals from the Irdin Manha Eocene of Mongolia. American Museum Novitates no. 198:p. 1–10. Google Scholar

50.

M. C. McKenna and S. K. Bell . 1997. Classification of Mammals, above the Species Level. xii + 631. Columbia University Press. New York. Google Scholar

51.

G. Méteis, M. Benammi, Y. Chaimanee, J-J. Jaeger, Than Tun, Tin Thein, and S. Ducrocq . 2000. Discovery of new ruminant dental remains from the Middle Eocene Pondaung Formation (Myanmar): reassessment of the phylogenetic position of Indomeryx. Comptes Rendus de l'Académie des Sciences, Paris, Sciences de la Terre et des Planètes/Earth and Planetary Sciences vol. 330:p. 805–811. Google Scholar

52.

G. Méteis, Y. Chaimanee, J-J. Jaeger, and S. Ducrocq . 2001. New remains of primitive ruminants from Thailand: evidence of the early evolution of the Ruminantia in Asia. Zoologica Scripta vol. 30:p. 231–248. Google Scholar

53.

K. Miyata and Y. Tomida . 1998. Eocene mammals from the Akasaki Formation, western Japan, and their biostratigraphic significance. Journal of Vertebrate Paleontology 18:supplement to no. 3. 64A. Google Scholar

54.

L. A. Nessov, J. D. Archibald, and Z. Kielan-Jaworowska . 1998. Ungulate-like mammals from the Late Cretaceous of Uzbekistan and a phylogenetic analysis of Ungulatomorpha. Bulletin of Carnegie Museum of Natural History no. 34:p. 40–88. Google Scholar

55.

R. Owen 1848. Description of teeth and portions of jaw of two extinct anthracotherioid quadrupeds…discovered…in the Eocene deposits on the N. W. coast of the Isle of Wight. Quarterly Journal of the Geological Society of London vol. 4:p. 103–141. pls. 7-8. Google Scholar

56.

G. E. Pilgrim 1925. The Perissodactyla of the Eocene of Burma. Palaeontologia Indica, New Series 8:memoir no. 3. 1–28. pls. 1-2. Google Scholar

57.

G. E. Pilgrim 1927. A Sivapithecus palate and other primate fossils from India. Palaeontologia Indica, New Series vol. 14:p. 1–26. pl. 1. Google Scholar

58.

G. E. Pilgrim 1928. The Artiodactyla of the Eocene of Burma. Palaeontologia Indica, New Series vol. 13:p. 1–39. pls. 1-4. Google Scholar

59.

G. E. Pilgrim and G. de P. Cotter . 1916. Some newly discovered Eocene mammals from Burma. Records of the Geological Survey of India vol. 47:p. 42–77. pls. 1-6. Google Scholar

60.

Pondaung Fossil Expedition Team 1997. Report on Work Achieved by the Pondaung Fossil Expedition Team. Office of Strategic Studies. Ministry of Defence, Yangon. in Burmese, partly in English. Google Scholar

61.

D. R. Prothero 1993. Ungulate phylogeny: molecular versus morphological evidence. In F. S. Szalay, M. J. Novacek, and M. C. McKenna , editors. eds.. Mammal Phylogeny—Placentals. p. 173–181.Springer-Verjag. New York. Google Scholar

62.

D. R. Prothero, E. M. Manning, and M. Fischer . 1988. The phylogeny of the ungulates. In M. J. Benton , editor. ed.. The Phylogeny and Classification of the Tetrapods, Volume 2: Mammals. p. 201–234.Clarendon Press. Oxford. Google Scholar

63.

T. Qi, K. C. Beard, B. Wang, M. R. Dawson, J. Guo, and C. Li . 1996. The Shandong mammalian fauna, middle Eocene of Jiangsu: history of discovery and significance. Vertebrata PalAsiatica vol. 34/no. 3:p. 202–214. in Chinese with English summary. Google Scholar

64.

Z. Qiu 1978. Late Eocene hypertragulids of Baise Basin, Kwangsi. Vertebrata PalAsiaticq vol. 16/no. 1:p. 7–12. pls. 1-2. (in Chinese with English abstract). Google Scholar

65.

L. B. Radinsky 1963. Origin and early evolution of North American Tapiroidea. Bulletin of Peabody Museum of Natural History, Yale University vol. 17:p. 1–106. pls. 1-4. Google Scholar

66.

L. B. Radinsky 1965. Early Tertiary Tapiroidea of Asia. Bulletin of the American Museum of Natural History vol. 129:p. 181–264. Google Scholar

67.

L. B. Radinsky 1969. The early evolution of the Perissodactyla. Evolution vol. 23:p. 308–328. Google Scholar

68.

Ranga Rao 1972. New mammalian genera and species from the Kalakot Zone of Himalayan foot hills near Kalakot, Jammu and Kashmir State, India. Special Paper of the Directorate of Geology, Oil, and Natural Gas Commission, Dehra Dun, India l:1–22. Google Scholar

69.

A. Sahni and S. K. Khare . 1971. Three new Eocene mammals from Rajauri District, Jammu and Kashmir. Journal of the Paleontological Society of India vol. 16:p. 41–53. Google Scholar

70.

R. M. Schoch 1984. Two unusual specimens of Helaletes in the Yale Peabody Museum collections, and some comments on the ancestry of the Tapiridae (Perissodactyla, Mammalia). Postilla no. 193:p. 1–20. Google Scholar

71.

R. M. Schoch 1989. A review of the tapiroids. In D. R. Prothero and R. M. Schoch , editors. eds.. The Evolution of Perissodactyls. p. 298–320.Oxford University Press. Oxford. Google Scholar

72.

G. A. Scopoli 1777. Introductio ad historiam naturalem, sistens genera lapidum, plantarum et animalium hactenus detecta, caracteribus essentialibus donata, in tribus divisa, subinde ad leges naturae. 506. p. Prague. Gerle. Google Scholar

73.

M. K. Scott and C. M. Janis . 1993. Relationships of the Ruminantia (Artiodactyla) and an analysis of the characters used in ruminant taxonomy. In F. S. Szalay, M. J. Novacek, and M. C. McKenna , editors. eds.. Mammal Phylogeny — Placentals. p. 282–302.Springer-Verlag. New York. Google Scholar

74.

N. Shigehara, M. Takai, R. F. Kay, Aye Ko Aung, Aung Naing Soe, Soe Thura Tun, T. Tsubamoto, and Tin Thein . 2002. New dental and facial remains of the middle Eocene primate Pondaungia cotteri from Central Myanmar. Journal of Human Evolution vol. 43/no. 2:p. 143–166. Google Scholar

75.

E. L. Simons, P. A. Holroyd, and T. M. Bown . 1991. Early elephant-shrews from Egypt and the origin of the Macroscelidea. Proceedings of the National Academy of Sciences of the United States of Americavol. 88:p. 9734–9737. Google Scholar

76.

L. D. Stamp 1922. An outline of the Tertiary geology of Burma. The Geological Magazine vol. 59/no. 11:p. 481–501. Google Scholar

77.

R. K. Stucky 1998. Eocene bunodont and bunoselenodont Artiodactyla (“dichobunids”). In C. M. Janis, K. M. Scott, and L. L. Jacobs , editors. eds.. Evolution of Tertiary Mammals of North America. Volume 1: Terrestrial Carnivores, Ungulates, and Ungulatelike Mammals. p. 358–374.Cambridge University Press. Cambridge. Google Scholar

78.

J. Sudre 1984. Cryptomeryx Schlosser, 1886, Tragulidé de l'Olig ocène d'Europe; relations du genre et considérations sur l'ori gine des ruminants. Palaeovertebrata vol. 14/no. 1:p. 1–31. Google Scholar

79.

J. Sudre and J. Erfurt . 1996. Les Artiodactyles du gisement Yprésien terminal de Prémontré (Aisne, France). Palaeovertebrata vol. 25:p. 391–414. Google Scholar

80.

J. Sudre, D. E. Russell, P. Louis, and D. E. Savage . 1983. Les Artiodactyles de l'Éocène infédrieur d'Europe, Première partie. Bulletin du Muséum National d'Histoire Naturelle ser. 4sec. C. vol. 5:p. 281–333. Google Scholar

81.

Swe Myint 1999. Foraminiferal fossil analysis of the rock samples collected by the Pondaung Fossils Expedition Team. In. Pondaung Fossil Expedition Team ed.. Proceedings of the Pondaung Fossil Expedition Team78–93.Office of Strategic Studies. Ministry of Defence, Yangon. Google Scholar

82.

M. Takai, N. Shigehara, Aye Ko Aung, Soe Thura Tun, Aung Naing Soe, T. Tsubamoto, and Tin Thein . 2001. A new anthropoid from the latest middle Eocene of Pondaung, central Myanmar. Journal of Human Evolution vol. 40/no. 5:p. 393–409. Google Scholar

83.

M. Takai, N. Shigehara, N. Egi, and T. Tsubamoto . 2003. Endocranial cast and morphology of the olfactory bulb of Amphipithecus mogaungensis (latest middle Eocene of Myanmar). Primates vol. 44/no. 2:p. 137–144. Google Scholar

84.

M. Takai, N. Shigehara, T. Tsubamoto, N. Egi, Aye Ko Aung, Tin Thein, Aung Naing Soe, and Soe Thura Tun . 2000. The latest middle Eocene primate fauna in the Pondaung area, Central Myanmar. In. Oversea Research Report of Studies on “Origin and Evolution of Hominoids in East Asia”. p. 7–28. 1997-99 Grant-in-Aid for Scientific Research Reports for Oversea Scientific Survey, “Asian Paleoprimatology, vol. 1, 2000”. Primate Research Institute. Kyoto University, Japan. Google Scholar

85.

M. Takai, T. Tsubamoto, Y. Kunimatsu, and N. Shigehara . 1999. Pondaung primates and its geological age: late Eocene mammal fauna in Myanmar. Primate Research vol. 15:p. 17–38. in Japanese with English summary. Google Scholar

86.

Thet Wai 1999. Palynological analysis of the rock samples collected by the Pondaung Fossils Expedition Team. In. Pondaung Fossil Expedition Team ed.. Proceedings of the Pondaung Fossil Expedition Team112–121.Office of Strategic Studies. Ministry of Defence, Yangon. Google Scholar

87.

S. Ting 1993. A preliminary report on an early Eocene mammalian fauna from Hengdong, Hunan Province, China. Kaupia, Dar mstädter Beiträge zur Naturgeschichte Heft 3. 201–207. Google Scholar

88.

H. Tobien 1985. Zur Osteologie von Masillabune (Mammalia, Artiodactyla, Haplobunodontidae) aus dem Mitteleozän der F ossilfundstätte Messel bei Darmstadt (S-Hessen, Bundesre-publik Deutschland). Geologisches Jahrbuch Hessen vol. 113:p. 5–58. Google Scholar

89.

Y. Tong and J. Wang . 1998. A preliminary report on the early Eocene mammals of the Wutu fauna, Shandong Province, China. Bulletin of Carnegie Museum of Natural History no. 34:p. 186–193. Google Scholar

90.

T. Tsubamoto, N. Egi, M. Takai, N. Shigehara, Aye Ko Aung, Tin Thein, Aung Naing Soe, and Soe Thura Tun . 2000a. A preliminary report on the Eocene mammals of the Pondaung fauna, Myanmar. In. Oversea Research Report of Studies on “Origin and Evolution of Hominoids in East Asia”. p. 29–101. 1997-99 Grant-in-Aid for Scientific Research Reports for Oversea Scientific Survey, “Asian Paleoprimatology, vol. 1, 2000”. Primate Research Institute. Kyoto University, Japan. Google Scholar

91.

T. Tsubamoto, P. A. Holroyd, M. Takai, N. Shigehara, Aye Ko Aung, Tin Thein, Aung Naing Soe, and Soe Thura Tun . 2000b. Upper premolar dentitions of Deperetella birmanica (Mammalia: Perissodactyla: Deperetellidae) from the Eocene Pondaung Formation, Myanmar. Paleontological Research vol. 4/no. 3:p. 183–189. Google Scholar

92.

T. Tsubamoto, H. Suzuki, N. Egi, M. Takai, and N. Shigehara . 2001. Excavation of mammalian fossils in Myanmar. Journal of the Geological Society of Japan 107/9:XVII-XVIII. in Japanese. Google Scholar

93.

T. Tsubamoto, M. Takai, N. Egi, and N. Shigehara . 2003. Mammalian faunal change in Eocene Asia and the Pondaung mammal fauna of Myanmar. Primate Research vol. 19/no. 1:p. 43–64. in Japanese with English summary. Google Scholar

94.

T. Tsubamoto, M. Takai, N. Egi, N. Shigehara, Soe Thura Tun, Aye Ko Aung, Aung Naing Soe, and Tin Thein . 2002a. The Anthracotheriidae (Mammalia; Artiodactyla) from the Eocene Pondaung Formation (Myanmar) and comments on some other anthracotheres from the Eocene of East Asia. Paleontological Research vol. 6/no. 4:p. 363–384. Google Scholar

95.

T. Tsubamoto, M. Takai, N. Shigehara, N. Egi, Soe Thura Tun, Aye Ko Aung, Maung Maung, T. Danhara, and H. Suzuki . 2002b. Fission-track zircon age of the Eocene Pondaung Formation, Myanmar. Journal of Human Evolution vol. 42/no. 4:p. 361–369. Google Scholar

96.

H. N. Turner Jr. 1849. On the evidence of affinity afforded by the skull in the ungulate Mammalia. Proceedings of the Zoological Society of London147–158. Google Scholar

97.

H. Uno 2000. The evaluation of the morphological variation of deciduous teeth in Desmostylus hesperus Marsh, 1888. Bulletin of Ashoro Museum of Paleontology no. 1:p. 153–158. in Japanese with English abstract. Google Scholar

98.

J. Wang and Y. Tong . 1996. A new lophialetid perissodactyl (Mammalia) from the early Eocene of Wutu basin, Shandong Province. Vertebrata PalAsiatica vol. 34/no. 4:p. 312–321. pi. 1. (in Chinese with English summary). Google Scholar

99.

M. Weber 1928. Die Säugetiere. Zweite Auflage, Gustav Fischer. Jena. Google Scholar

100.

R. M. West 1980. Middle Eocene large mammal assemblage with Tethyan affinities, Ganda Kas Region, Pakistan. Journal of Paleontology vol. 54/no. 3:p. 508–533. Google Scholar

101.

H. E. Wood 1937. Perissodactyl suborders. Journal of Mammalogy vol. 18/no. 1:p. 106. Google Scholar
Takehisa Tsubamoto, SOE THURA TUN, Naoko Egi, Masanaru Takai, Nobuo Shigehara, Aung Naing Soe, AYE KO AUNG, and Tin Thein "Reevaluation of some ungulate mammals from the Eocene Pondaung Formation, Myanmar," Paleontological Research 7(3), 219-243, (30 September 2003). https://doi.org/10.2517/prpsj.7.219
Received: 29 May 2002; Accepted: 1 April 2003; Published: 30 September 2003
KEYWORDS
Asiohomacodon myanmarensis
Eocene
Hsanotherium
Indolophus
Indomeryx
Pondaung Formation
Back to Top