icop.3 ifS :i m UN' RSITYOF ILLIiwiS LIBRARY AT URBANA-CHAMPAIGN NATURAL HIST. SURVEY FIELDIANA • GEOLOGY Published by CHICAGO NATURAL HISTORY MUSEUM Volume 10 June 19, 1953 No. 15 A NEW AND PRIMITIVE EARLY OLIGOCENE HORSE FROM TRANS-PECOS TEXAS Paul O. McGrew Department of Geology, the University op Wyoming Fragmentary remains of Oligocene mammals have been found sporadically in Trans-Pecos Texas for some twenty-five years. Real knowledge of these interesting occurrences dates from 1938 and 1940, when parties from the Museum of the University of Oklahoma collected in the Porvenir-Pilares area, Presidio County (Stovall, 1948). In 1946, a Chicago Natural History Museum expedition, composed of Bryan Patterson, James H. Quinn and John M. Schmidt, working in collaboration with the Texas Memorial Museum and the Bureau of Economic Geology, the University of Texas, made extensive collections in the same area. The Equidae obtained by this expedition were entrusted to me for study, for which I wish to express my sincere thanks. The description of the new and very interesting form here recorded has been extracted from my manu- script and is being published now in order to facilitate work by others, in which reference to it is desirable. Family Equidae Subfamily Hyracotheriinae Haplohippus gen. nov. Type species. — Haplohippus texanus sp. nov. Known distribution. — Earliest Oligocene, western Texas. Diagnosis. — Height of cheek teeth approximately as in Epihip- pus; lower premolars without and lower molars with external cingula; P2 not molariform, with single anterior cusp and poorly distinguished posterior cusps; P^ partially molariform with incomplete lophids; P4 and molars with distinct entoconids and low parastylids; sym- physis long. No. 707 167 168 FIELDIANA: GEOLOGY, VOLUME 10 Haplohippus texanus sp. nov. Type. — C.N.H.M. PM 17, nearly complete lower jaw with sym- physis and both rami, lacking coronoids and condyles. Hypodigm. — Type and C.N.H.M, PM 29, portion of left ramus with talonid of My, M^.^. Horizon and locality. — Vieja Formation, near the base of a series of maroon tuffs and greenish-gray conglomerate 180 feet thick; this series overlies the basal lava flows of the formation. Both specimens were found at the base of a conspicuous promontory, known locally as Big Cliff, 3^4 miles east-northeast of the mouth of Van Horn Creek and 2^i miles northwest of the old Quinn Ranch Spring. This appears to be the spot where the greater part of the University of Oklahoma material was obtained (Patterson, personal communi- cation).^ Diagnosis. — As for the genus. Description. — Py is a simple tooth only slightly separated from Po, It is compressed laterally, with an antero-posterior crest ex- tending the entire length of the tooth. It has a rather prominent posterior shoulder and a slight postero-internal cingulum. The tooth is double-rooted; that of Mesohippus is single-rooted. Pa is very primitive, and, in sharp contrast to known contemporary horses, not molariform. There is a single, high, rather sharp cusp with no sign of separation into metaconid and protoconid. Running along the anterior side of this cusp is a vertical ridge corresponding with the parastylid of the posterior premolars and the molars. Behind the anterior cusp is a broad, shallow valley with a low ridge extend- ing posteriorly from the crest of the cusp. Posteriorly, the hypoconid is a high, pointed central cusp. Sloping internally from the hypoconid is a small entoconid. Behind the entoconid and separated from it by a narrow groove is a postero-internal ridge that probably is the hypoconulid. F2 has reached approximately the same stage of evo- lution as P^ of Hyracotherium. P^ is sub-molariform. Anteriorly, there are two rather closely appressed cusps, separated by a narrow groove, that represent the metaconid and the protoconid. A ridge slopes anteriorly and slightly internally from the protoconid, but is little more than an incipient parastylid. The hypoconid and the entoconid are separated by a 1 It must be noted that the scale given by Stovall (1948, fig. 2) is misleading. The portion of the map reproduced (U.S.G.S. Texas, Presidio County, San Carlos Sheet) is reduced to two-thirds, whereas the scale in miles is reduced to one-half. Fig. 65. Haplohippus texanus gen. et sp. nov. C.N.H.M. PM 17; X 1 Drawing by Miss Maidi Wiebe. 169 170 FIELDIANA: GEOLOGY, VOLUME 10 distinct, sharp groove and would become connected only with con- siderable wear. Extending antero-internally from the hypoconid is a ridge that slopes to the base of the metaconid. A distinct meta- stylid, such as occurs on the completely molariform P^ of Meso- hippus, is not present. P4 is molariform, although the lophids are not well developed. The entoconid remains distinct. The premo- lars are without external cingula. The molars are primitive in that the metastylids are poorly developed and the anterior portions of the protolophids are low. A prominent external cingulum is present and the entoconid is a dis- tinct cusp on all. In PM 29, the external cingulum rises to a cusp between the hypoconid and the hypoconulid. Compared with Meso- hippus, all cheek teeth are low-crowned; the ramus is much deeper and the symphysis much longer. MEASUREMENTS (In Millimeters) PM 17 Pt P2 P3 P4 Mt M2 M3 PM 29 My Mg M3 L 5.0 8.0 9.5 9.0 9.5 9.7 13.2 L ... 9.5 14.3 W 2.9 4.2 5.7 6.7 6.7 7.0 6.8 W 6.6 7.3 6.8 PM 17 PM 29 Length of symphysis 40 . 6 .... Length of diastema, C-Py . . 36.2 Depth of ramus under Pj . . . 20-5 .... Depth of ramus under M3 . . 26.4 25.0 Discussion.- — After so many years of extensive field collecting and intensive research on fossil horses, it is most interesting to dis- cover at this time a new Early Tertiary phylum. Several phyla of horses are well known in the Miocene, Pliocene, and Pleistocene, but, until now, it has been believed that in the Eocene and Oligocene there occurred only a single, approximate line of descent: Hyraco- therium-Orohippus-Epihippus-Mesohippus-Miohippus. Haplohippus, an Early Oligocene form retaining many of the dental characters of the Mid Eocene Orohippus, from which it may well have descended directly, shows that at least two phyla were in existence in the Eocene and Oligocene. Both apparently stemmed from Orohippus; the one culminated, so far as known, in the Chadronian Haplohippus; the other passed through the Uintan and Duchesnean Epihippus to the Oligocene and Miocene Anchitheriinae. To judge from its molar structure, there can be little doubt that Haplohippus should be in- cluded in the Hyracotheriinae, hitherto unknown subsequent to Eo- cene time. Epihippus, a form with more advanced molar structure McGREW: AN EARLY OLIGOCENE HORSE 171 and on or near the line leading to Mesohippus, should, I believe, be transferred from this subfamily to the Anchitheriinae. We thus have in the earlier Tertiary a situation comparable to those obtain- ing later on, when browsing anchitheriines were contemporaneous with grazing equines and three- toed hipparions lived side by side with one-toed forms. Horse evolution, it would seem, has been a complicated, branching affair almost from the beginning. These surviving phyla of the later Cenozoic enjoyed a wide geo- graphic range in North America, and several of them made their way to the Old World. This may not have been true of the line represented by Haplohippus, no trace of which has yet appeared in northern Chadronian deposits or in those of Uintan and Duchesnean age. It is possible that the phylum was largely restricted to the southern portion of the continent, but the rarity of horse remains in the Uinta, Duchesne River, and lower Chadron prevents assur- ance on this point. REFERENCE Stovall, J. W. 1948. Chadron vertebrate fossils from below the Rim Rock of Presidio County, Texas. Amer. Jour. Sci., 246, pp. 78-95, figs. 1-3, pis. 1, 2. a^^B^Bti * f mm 1