THE PEARCE­SELLARDS Series Number 23 EARLY TERTIARY VERTEBRATE FAUNAS, VIEJA GROUP, TRANS-PECOS TEXAS: INSECTIVORA Michael J. Novacek February, 1976 Texas Memorial Museum/2400 Trinity/Austin, Texas 78705 W. W. Newcomb, Director material, ReferredType. for ERRATUM: 28, linesubstitute 3, Page The Pearce-Sellards Series is an occasional, miscellaneous series of brief reports of museum and museum associated field investigationsand other research. Its title seeks to commemorate the first two directors of the Texas Memorial Museum, both now deceased: J. E. Pearce and Dr. E. H. Sellards, professors of anthropology and geology respectively, of The University of Texas. A complete list of Pearce-Sellards papers, as well as all other publica­tions of the museum, will be sent upon request. INTRODUCTION This paper is another in a series of contributions dealing with the verte­brate faunas from the Vieja group. Early publications, primarily Stovall (1948), DeFord (1958), Wilson et al. (1968), and Wilson (1971 a), discussed the location, previous work, stratigraphy, and age of the Vieja group. For a stratigraphic section showing the positions of the local faunas, see Wilson (1971b, Fig. l,p. 4). METHODS The dental nomenclature employed here follows that of Rich (1971: 4), who slightly modified dental terminology proposed by Van Valen (1966:7-9). All specimens were measured on an Ehrenreich Photo Optical “Shop-scope.” Measurements were in millimeters and were rounded off to the near­est one-hundredth of a millimeter. The following orientations for measuring cheek teeth (Fig. 1) were used in this report: ANTEROPOSTERIOR AXIS (“A-P”) AXIS Lower posterior premolars long axis of tooth. Lower molars line drawn through the apices of the metaconid and the entoconid. Upper posterior premolars line extended from anteriormost point of the antero­labial lobe to the posteriormost point of the metastylar lobe. Upper molars (JvU-2) iine drawn through the apices of the paracone and metacone, except in the case of_Apternodus cf. brevirostris where criteria immediately below apply for all upper molars. Upper molars (M3) line drawn at right angles to a line which divides the tooth into - equal anterior and posterior halves. LENGTH - Lower posterior premolars total length of crown, i.e., greatest dimension measured parallel to “A-P” axis. - Lower molars total distance from the anterior face of the paraconid to the back of the talonid along a line parallel to the “A-P” axis (anterior cingulum not included in measurement). - Upper posterior premolars and upper molars greatest dimension parallel to the “A-P” axis of tooth. WIDTH Lower posterior premolars widest part of crown; greatest dimension measured at right angles to “A-P” axis. 1 Fig.I.—Occlusalviewof diagrammaticcheekteethshowingorientationsformeasurements:(a) lowerposteriorpremolar,(b)lowermolar,(c)upperposteriorpremolar,(d)uppermolar(MlorM2),(e)uppermolar(M3). Lower molars two width measurements: trigonid width, talonid width, widest part of each of these sections of the crown measured at right angles to “A-P” axis. Upper posterior premolars and upper molars two measurements: anterior width,distance from the labialmost point of the anterolabial corner of the crown to the lingualmost point of the protocone; posterior width, distance from posterolabialmostpoint of metastylar lobe to lingualmost point of protocone; both width measurements taken at right angles to the “A-P” axis. All specimens described in this report are conserved in the Texas Memorial Museum (TMM). Detailed descriptions of localities are on file at the Verte­brate Paleontology Laboratory, Balcones Research Center, The Universityof Texas at Austin. ACKNOWLEDGMENTS thank Dr. J. A. Wilson for generously offering me the Vieja insectivores for study. Dr. Malcolm C. McKenna helped me with the identification of Apternodus cf. brevirostris. Mrs. Margaret Stevens prepared all illustrations except Figure 1. I am very grateful to the following individuals and institutions for allow­ing me to study leptictid specimens under their care: Dr. William Turnbull, Chicago Field Museum of Natural History; Dr. Mary R. Dawson, Carnegie Museum of Natural History; Ms. Gay Vostreys and Dr. Dave Gillette, Phila­delphia Academy of Sciences; Dr. Malcolm C. McKenna, American Museum of Natural History; Dr. Parish Jenkins, Museum of Comparative Zoology at Harvard University; Dr. Donald Baird, Princeton Geological Museum; Dr. Robert J. Emry, United States National Museum of Smithsonian Institution; Drs. Robert W. Wilson and Morton Green, Museum of Geology, South Dakota School of Mines and Technology; Dr. J. T. Gregory, University of California, Museum of Paleontology; and Messrs. Bruce Lander, Ken Rose, and Craig Wood. I thank Drs. M. C. McKenna, D. E. Savage, J. A. Lillegraven,and W. A. Clemens, for their critical readings of earlier versions of this paper SYSTEMATIC PALEONTOLOGY No recent classification of the Insectivora has met with widespread accept­ance among students of the group (for conflicting views on insectivore tax­ onomy see Van Valen, 1966, 1967; Butler, 1956, 1972; and Novacek, in press). 1 follow here Van Valen’s (1967) designation of higher insectivore categories, pending further study and revision of the order. Order Insectivora Illiger, 1811 Suborder Proteutheria Romer, 1966 Family Leptictidae Gill, 1872 Subfamily Leptictinae Gill, 1872 Genus Leptictis Leidy, 1868 Leptictis wilsoni sp. Nov. Figs. 2, 3; Table 1 Etymology.—Named in honor ofDr. J. A. Wilson. Referred material.—TMM 40209-215, incomplete skull with P3-M 3 on right maxillary; Ml"3 and a fragment of P 3 on left maxillary. Stratigraphic position.—Upper part of Chambers Tuff, 60 feet below Bracks Rhyolite, Vieja Group. Reeves Bonebed, Presidio Co., Texas. Little Egypt local fauna, Chadronian (Early Oligocene). Diagnosis.-Skull and dentition significantly smaller than those referred toLeptictishaydeni,L. dakotensis,L.bullatus,and L. douglassi (secbelow), zygomatic arch narrow, no dorsoventral swelling ol the jugal; paroccipital process or squamosal flange not strongly produced ventrally; shallow ant-orbital fossa; depression in back of the zygomatic process of squamosal tor opening of the subsquamosal foramen very shallow; P3 tricuspid with lingual and anterior spurs; molariform P4 longer but not quite as wide as M 1 with hypocone and precingulum; M^' 3 with well developed hypocones, precingula, and labial spurs; molar cusps conical but not swollen at their bases or bulbous; M 3 not greatly reduced relative to Ml"3 with stronger metacone than in most species of Leptictis ; length P 3-M 3 14.52 mm. TABLE 1.-Measurements of skull and teeth (in mm) of Leptictis wilsoni sp. Nov., TMM 40209-215. Measurements were taken on the right dentition only. Skull width at orbital constriction 11.13 Skull width at P­ 10.13 Palate width at P­ 6.65 Length P3-M3 Length M1 -3 14.52 7.62 P3 Length3.66 Anterior Width 2.90 Posterior Width - p4 Ml 3.35 3.05 3.63 4.03 3.92 4.13 2.91 4.10 3.94 M3 2 1 2 3.41 2.99 DESCRIPTION The dentition in TMM 40209-215 is complete from the through M-3 in the right maxillary. Only Ml‘3 and a fragment of are preserved in the left maxillary. P-3 is a three-rooted, vaguely triangular tooth dominated by a high paracone which is broken not far above its base. There appears to have been a metacone but the area of its location in the crown has been subject to breakage and heavy wear. A small parastylar spur projects direct­ly anteriorly from the base of the paracone. The protocone, better preservedin the left P3, is a prominent cusp but lower than the metacone. A small paraconule is situated just lingual to the base of the paracone at the labial termination of the preprotocrista. The posterior base of the protocone is bordered by a very narrow precingulum. There is no hypocone. p4 is a three-rooted molariform tooth, considerably larger than P3 and less transversely wide than Ml and The paracone and metacone are broken just above their bases. The labial face of the paracone is nearly con­tinuous with the labial edge of the crown, and there is a very narrow ecto­cingulum external to the metacone. There is an anteriorly projecting para­stylar spur but details of its morphology are obscured by damage and wear. The paraconule and metaconule are well developed bulbous cuspules situated labially on the crown. The paraconule lies nearly at the base of the paracone;the metaconule is separated from the metacone by a shortly extending post­metaconule wing. There is no premetaconule wing or postparaconule wing.The protofossa is deep, its lowest point situated directly between the para­conule and the metaconule. The protocone is a prominent cusp with a steepprevallum and postvallum. The postcingulum extends from the base of the crown below the metaconule terminating in a small hypocone. There is a narrow precingulum. Ml is more transversely elongate and less anteroposteriorly long than p4. A narrow ectocingulum borders the labial faces of the metacone and para­ cone. The paracone is considerably higher than the metacone and well separated from the latter. There is no paracrista. The parastylar spur is smaller than that on p4. A short, low metacrista runs to a point just labial to the posterolabial corner of the crown. The paraconule and metaconule are subequal in size and are situated much closer to the paracone and meta-cone respectively than to the protocone. The premetaconule and postpara­conule wings are poorly developed. A narrow metacingulum runs from the metaconule to the posterior border of the crown just labial to the meta­crista. The protocone is higher than the metacone but not as tall as the paracone. A long postcingulum terminates lingually in a heavily worn hypo-cone. The narrow precingulum extends along the base of the crown from a point below the preprotocrista to a point below the anterior face of the protocone short of the anterolingual corner of the crown. M-has a deep ectoflexus whose most lingual point is external to the meta-cone. There is a prominent metastylar spur. The metaconule is much lower and smaller in basal dimensions than the paraconule. The precingulum is broader in M-than in Ml. In all other features M-is nearly identical to M 1. M 3 differs from in its more transverse outline, and in having a strongly anterolabially projecting parastylar spur, a wide ectocingulum labial to the paracone, a posteriorly canted metacone that is considerably reduced relative to the paracone, and a paraconule about midway between the proto­cone and the base of the paracone. Unlike MIM3 virtually lacks a meta­stylar spur and has a much narrower postcingulum with a vestigial hypocone. Wear on the molars is most extensive along the prevallum: a wear tacet on the protocone is confluent with that running along the preprotocrista and that on the paraconule. Wear is less evident on the postprotocrista and on the apices of the metacone and paracone. The skull is incomplete, missing much of the facial portion, the rostrum, the left side of the basicranial and squamosal region, and the occipital region. Much of the remaining portions are badly damaged and deformed. The top of the skull in the region ot the parietals is fragmentary. The base of the skull posterior to the palatine has been compressed and twisted sinistrally, leaving many structures, including the left epipterygoid process and the alisphenoid, severely damaged or miss­ing. What follows is a description of the major cranial features which are reasonably well preserved. The anterior opening of the infraorbital canal is large and is situated directly above the Ml. There is a distinct depression running along the an­terior border of the orbit, presumably for the origin of the snout muscles (levator labii superioris, levator alae nasi, zygomaticus). The anterior border of the orbit is crested, and the lacrimal foramen opens within the orbit. The lacrimal does not have a distinct facial extension. The jugal is a long, large bone with a furcate anterior end. Its lower an­terior branch contacts the short zygomatic process of the maxillary while the upper branch runs up the anterior border of the orbit, terminating just below the lacrimal-maxillary suture. The jugal continues tor a long distance along the zygomatic arch gradually tapering posteriorly and moving below the zygomatic process of the squamosal. There is no postorbital process on the jugal nor is there one on the frontal, and no demarcation exists between the orbit and the temporal fossa. Fig. 2.—Leptictis wilsoni sp. Nov. TMM 40209-215. Partial skull with posterior premolar,molar dentition: (a) ventral view, (b) view of right side of skull. The anterior root of the zygoma arises opposite the M~'3. A small, shallow fossa on the zygomatic process of the squamosal sur­rounds the subsquamosal foramen. The large postglenoid foramen is located at the internal posterior base of the partially preserved postglenoid process. Details of the tympanic chamber are obscured by damage, but it is pos­sible to discern a few important structures. There is no auditory bulla, but this has probably been lost in fossilization, as is the case in many specimensof Leptictis. A flange of the periotic at the posterior wall of the tympanicchamber is lacking and the fenestra rotunda is fully exposed. The stylo­mastoid foramen is not isolated from the tympanic chamber by the tym­panohyal. The promontorium (incompletely preserved) is well developedwith a “high knob” above the foramen rotunda. Grooves on the promon­torium for the passage of the lateral internal carotid artery and its branches are not preserved. Between the mastoid and paroccipital processes is a gutter in the ventral surface of the mastoid bone, presumably for the facial nerve. The small paroccipital process is situated slightly posterior to the suture between the mastoid and occipital bones. The mastoid exposure faces directly backward and does not have a lateral component. Fig. 3.-Leptictis wilsoni sp. Nov. TMM 40209-215. Right p 3-M 3. DISCUSSION The premolars in the above described specimen may be deciduous, but certain observations vitiate against this possibility. Although the last upperpremolar is molariform, permanent molariform premolars are diagnosticof the Leptictidae in general. The protocone, hypocone, and conules in the last upper right premolar show less wear than homologous cusps in Ml-2. The low relief of the paracone and metacone in the fourth premolar, and the protocone and metacone in the third, is probably a result of breakagerather than wear. Unlike the in known specimens referable to Leptic­tis (Lillegraven, 1969), the third upper premolar does not have a strongposterior cingulum and well developed conules. West (1972) observed a general tendency among individuals of this genus for late retention of DP3-DP4. Almost all of the Leptictis specimens I have examined, however, 7 show that the presence of a permanent p3, p4, and the phenomenon of late replacement seems more characteristic of Leptictis accutidens than other species of the genus. The genus Leptictis was first described by Leidy (1868) from the Brule formation of the White River Group in the Badlands of South Dakota. The type and only specimen of Leptictis haydeni was a beautifully preserved skull. Leidy also described a fragmentary skull in the same paper which he made the type of Ictops dakotensis. Leptictis haydeni was distinguished from Ictops dakotensis in having a without an inner cusp (protocone),and in the alignment of and the anterior premolars with the lingual,instead of the labial, border of p4. Despite such minor differences, recog­nition of both genera persisted in the literature until Van Valen’s (1967) synonomy of Ictops dakotensis with Leptictis haydeni. There seems little case for retaining the name Ictops as these ditferences hardly warrant sepa­ration at the generic level, but whether or not L. haydeni and I. dakotensis are conspecific is a matter worth further consideration. Van Valen (1967) claimed that the P3 characters in L. haydeni were of minor taxonomic significance as the morphology of this tooth showed such a high degree ot intraspecific variation. 1 have examined approximately 100 skulls and den­titions of Oligocene leptictids which show, contrary to Van Valen’s observa­tion, that the is fairly consistent in morphology, and, in fact, is quiteuseful taxonomically. The condition of in the type of Leptictis haydeni is not known in any other specimens referable to the genus. Thus, it one accepts Van Valen’s arrangement, the most atypical Leptictis specimen is the type specimen. Although it has not been noted previously, this specimenis distinctive in features other than morphology. The skull is much larger than the original type of Ictops dakotensis and most specimensreferred to the latter species, and the zygomatic arch is deep. The latter condition cannot easily be attributed to the large size of this specimen, as certain specimens identified in collections as I. dakotensis (e.g. CM 2149) are as large as the type skull but show the presence of the very narrow zygo­matic arch typical of the hypodigm. Thus, the possibility that the type of L. haydeni represents a species distinct from that represented by the original type of/, dakotensis cannot easily be dismissed. 1 propose the following taxonomic arrangement as a partial solution to the problem. Leptictis haydeni should be regarded as a species distinct from Leptictis dakotensis, the latter for which Leidy’s type of Ictops dakotensis would serve as the type specimen. The hypodigm of L. haydeni would then consist of a single specimen, but such is far from the rare case in fossil taxa where the typological concept of a species predominates in lieu of adequate­ly large samples. If the diagnostic features in L. haydeni can be attributed to intraspecific variation based on well documented evidence, this species can be synonymized with L. dakotensis. Pending such a study it seems more useful to recognize L. haydeni as a distinct species. The diagnoses for Ictopsdakotensis provided by Leidy (1868) and Scott and Jepsen (1936) would thus stand as the diagnosis for Leptictis dakotensis. A further consideration of this matter is presented in a formal revision of the Leptictidae now in preparation by this author. Leptictis wilsoni from the Vieja differs from Leptictis haydeni Leidy(1868) in having a P3 with a metacone and a protocone, and in that the labial border of P3 is in line with the external margin of the molars and P4 as part of a gradual arc curving in an anterolingual direction. These differ­ences correspond to those, originally cited by Leidy (1838), between Leptic­tis (= letups) dakotensis and Leptictis haydeni. Leidy’s type of letups dakutensis was so fragmentary that it was not cer­tain whether it belonged to either one of the two variants from the Brule formation. Because Matthew (1903) referred the larger and more robust of these to his Leptictis (= letups) bullatus, Scott and Jepsen (1936) restricted the name Leptictis (= letups) dakutensis to the smaller and more slender specimens from the Brule. It is possible, as Scott and Jepsen recognized, that L. dakutensis and L. bullatus merely represent different growth stages of the same species. Nevertheless, the latter is distinctive in the very small size of m 3 and its specific separation may be valid. In Leptictis wilsoni, 4/3 is not - greatly reduced relative to M l 2. Skulls referred to Leptictis (= letups) dakutensis differ from Leptictiswilsoni in; (1) their larger size; (2) having more transverse molars with much narrower ectocingulum on Mi-2; (3) a vestigial metacone on m3; (4) less prominently projecting parastylar spurs on P3-M3; (5) more bulbous molar cusps; and (6) the presence of a much deeper and larger fossa in the zygo­matic process of the squamosal for the opening of the subsquamosal fora­ men. Leptictis thumpsuni (Matthew, 1903) is easily distinguished from all other Leptictis species, including L. wilsoni, in having extremely transverse teeth with sharper, more piercing cusps, a more projecting anterior spur on P3, in lacking precingula on the upper molars, and a postcingulum and hypocone on M 3 and p4. L. thumpsuni is also the smallest of the species of Leptictis. Showing the closest resemblance to Leptictis wilsoni is L. acutidens de­scribed by Douglass (1901, 1905) from Pipestone Springs, Montana, Chad­ronian fauna. The only differences observed between the two species are (1) the smaller relative size of M 3 and P4; (2) the more transverse uppermolars; (3) the more labial position of the molar precingula; and (4) the weaker labial lobes, particularly the parastylar lobes on and m 3 in L. acutidens. Matthew (1903) stated that L. acutidens is in some ways struc­turally intermediate between Palaeietups and Oligocene leptictids. I agreewith his assertion. The dental resemblance between L. acutidens and Palae­ietups tauricinerei is strong. Douglass also described Leptictis montanus and L. tenuis in his 1905 monograph on the Tertiary of Montana. The types and only known speci­mens referred to these species were both from the Chadronian McCarty’sMtn. fauna. Comparisons of these specimens lead me to believe that L. montanusis aseniorsynonymofL.tenuisandthatL.montanusisreferable to a new undescribed leptictid genus from the Bates Hole of Wyoming. A diagnosis and description of this new genus will be provided elsewhere. L. montanus (= L. tenuis) is easily set apart from Leptictis wilsoni in having (1) a very deep and short zygomatic arch; (2) a strongly produced paroc­cipital process and squamosal flange; (3) a very deep anteorbital fossa; (4) relatively much smaller than and (5) a more projecting anterior spur on p3. 9 There remain for comparison three poorly represented species of Lep­tictis whose specific status may not be valid. Leptictis intermedins (Douglass,1905), also from the McCarty’s Mtn. fauna, looks like a smaller individual referable to L. montanus. Leptictis major (Douglass, 1905), the largestspecies from the McCarty’s Mtn. locality, is probably a nomen dubium. Den­tal characters, other than greater size, do not readily distinguish it from L. montanus. L. major differs from L. wilsoni in its larger size and in havingless transverse upper molars with weaker labial lobes, a vestigial metacone on M3, and a more projecting anterior spur on p3. Cope’s species Leptictis (= Mesodectes) caniculus is identical to Leptictis dakotensis except for the lack of a posteroexternal cusp on p3. The species is a nomen dubium best referred to Leptictis dakotensis (Scott and Jepsen 1936, p. 22). To sum up, Leptictis wilsoni seems different enough from all recognizedspecies of Leptictis to warrant its specific recognition. This Vieja leptictidshows the closest morphological similarity to Leptictis acutidens from the early Oligocene (Chadronian) of Montana. Leptictis douglassi sp. Nov. Figs. 4, 5; Table 2 - Etymology.Named in honor of Earl Douglass, a vertebrate paleontologistof the late 19th and early 20th centuries. Type. TMM 40688-6, both halves lower jaw; left ramus with P3-4-Ml-2,right ramus with P4-M3; upper maxillary fragment with P3. Stratigraphic position. Rifle Range Hollow, Presidio Co., Texas. Approxi­mately 200 feet above the base of Chambers Tuff, Vieja Group. Porvenir local fauna, Chadronian (early Oligocene). Diagnosis. —Significantly larger than all other known species of Leptictisp3 with trenchant outer row of cusps, a prominently projecting anterior; spur, a large anterior accessory cusp, and a reduced protocone; P4 paraconidwell developed and bulbous, with apex positioned anterior to the lowest point of the protolophid. TABLE 2.—Measurements of mandible and teeth of Leptictis douglassi sp. Nov. Depth of mandible below M 3 8.30 Length P4-M3 I 8.86 Length M ]-M^ I 3.46 Length P^ 5.99 Width P3 3.92 Length Trigonid Width Talonid Width |54 5.92 3.2 I 3.32 M \ 4.58 3.53 3.86 Mi 4.36 3.67 3.60 m3 4.37 3.28 2.64 DESCRIPTION TMM 40688-6 is a robust mandible (depth of jaw below alveolar border of M 3 is 8.60 mm.) with a well preserved series of teeth from the P4-M3 in the right ramus and P4-M2 in the left. The left P3 is only partially preserved.It is a trenchant two-rooted tooth with a prominent posterior cusp posi­tioned directly above the posterior root. The anterior cusp appears to have been still larger but it is broken near its base. A narrow cingulum arcs along the posterior and lingual bases of the posterior cusp. The labial termination of this cingulum is marked by a minute cuspule. The cingulum is separatedfrom the lingual base of the posterior cusp by a shallow fossa. Fig. 4.-Leptictis douglassi sp. Nov. TMM 40688-6. Left ramus with 3-4, M 1-2; right P M 3 shown reversed, (a) lingual view, (b) occlusal view, (c) labial view. ?4 is a large molariform tooth, longer anteroposteriorly than the lower molars. The paraconid is robust, somewhat bulbous, and subequal in heightwith the hypoconid. The apex of the paraconid is anterior to the lowest point of the protolophid. There is no anterior cingulum on the prevallid of the tooth. A deep prefossid opens lingually, separating the paraconid and metaconid nearly at their bases. The paralophid is a sharp crest, V-shapedin labial profile, which diverges from its lowest point steeply up the proto­coled and the metaconid faces. The latter two cusps are subequal in heightbut the protoconid is slightly more bulbous, robust, and is situated more anteriorly on the crown than the metaconid. The steep, but not sheer, post­vallid is formed by the posterior faces of the protoconid and metaconid. The talonid is well developed, with a basin and three cusps. The crista obliquaterminates at the postvallid wall at a point below the apex of the proto­lophid. The hypoconid is a prominent cusp, much larger and higher than the hypoconulid and entoconid. The apex of the hypoconulid is situated sub-equidistantly between and slightly posterior to the entoconid and hypo­conid. A minute entoconulid is situated immediately behind the deep, narrow groove for the lingual opening of the talonid basin. Unlike P4, the lower molars have very reduced crest-like paraconids, givingthe teeth a more quadrate outline in occlusal view. There is a narrow anterior cingulum that runs for a short distance dorsolingually up the prevallid below the protolophid. The apices of the protoconid and metaconid are nearlyopposite each other. There does not appear to be a small entoconulid in anyof the lower molars. M] is almost identical to M 1 except for its basal dimen­sions (Table 2). M 3 is distinguished by its transversely narrower dimensions, more elongate talonid, and very prominent hypoconulid. Differential wear occurred on the molars, as the metaconid is worn extensively relative to the protoconid. A confluent wear facet joins the apices of the metaconid and paraconid. The latter is worn to a flat shelf, no higher than the basin of the prefossid and the lowest part of the protolophid. The deepest part of the jaw is below the talonid of M-A mental foramen is present about midway on the jaw below the paraconid of P4. The three-rooted is an anteroposteriorly elongate tooth with a tren­chant outer row of cusps and a very weak protocone: most of the crown pro­jects anterior to the protocone and metacone. The anterior spur is composedof a prominent anterior accessory cusp and a large paracone, easily the most dominant cusp on the crown. The metacone is considerably lower and smaller than the paracone but distinctly higher than the protocone. A small cingulum rises up the lingual posterior base of the metacone terminating in a minute cuspule in the posterolabial corner of the crown. The protoconeis a cone-shaped cusp lacking either a pre-or postcingula along its bases. DISCUSSION Despite the current problems and complexities in the intrageneric tax­onomy of Leptictis, there seems little doubt that Leptictis douglassi is a distinct species. Its considerably larger size alone easily distinguishes it from other species of the genus. I know of no other member of Leptictis that has Fig. s.—Leptictis douglassi sp. Nov. TMM 40688-6. P 3 (left): (a) labial view, (b) oc­clusal view, (c) lingual view. such an exaggerated development of the anterior spur of the althoughthis condition is approached in L. montanus and L. acutidens. The mor­phology in this animal is interesting because it suggests a condition struc­turally intermediate between the simple elongate P-3 lacking a protocone in the type of Leptictis haydeni and the condition seen in the p3s of most other known Leptictis individuals. Suborder Zalambdodonta Gill, 1884 Super Family Tenrecoidea Gray, 1821 Family Tenrecidae Gray, 1821 Subfamily Apternodontinae Matthew, 1910 Genus Apternodus Matthew, 1903 Apternodus cf. brevirostris (Schlaikjer, 1934) Fig. 6, Table 3 Referred material—TMM 40492-9, partial skull with right Ml-3, left p4,M I'2, and fragment of M^. Stratigraphic position.—Red Mound, approximately one-half mile north of Big Cliff, Presidio Co., Texas. Approximately 100 feet above Buckshot Ignimbrite Formation. Porvenir local fauna, Chadronian (early Oligocene). TABLE 3. -Measurements of the skull and teeth of Apternodus cf. brevirostris, TMM 40429-29 Palate width at M-7.22 Length P4-M3 8.30 Lengt h Width 1 p4 2.49 2.45 Ml 2.57 3.54 m: 2.32 3.46 M3 1.68 3.21 1Width = the greatest dimension taken perpendicular to the anteroposterior axis of the tooth. The morphology of these teeth precluded measurement of both anterior and posterior width with accuracy. DESCRIPTION The cheek teeth preserved in TMM 40492-9 are all highly specialized for a shearing and piercing mode of occlusion. The crown is dominated by a single sharp cusp. On the molars, two distinct ridges diverge from the apexof the high central cusp (probably homologous with a paracone;see Butler, 1972) and run to either corner on the labial edge of the tooth, lending to the crown a V-shaped appearance when viewed in a horizontal plane. Such a condition is commonly known as “zalambdodonty.” p4 is a three-rooted, triangular tooth consisting essentially of a high cen­tral cusp and narrow basal anterior and external cingula. A strong crest runs from the apex of the central cusp to the posterolabial corner of the crown. The prevallum surface of the central cusp, formed by its anterior, labial, and part of its lingual faces, is rounded. The postvallum is a sheer wall, set ob­ liquely in a posterolingual direction in plan view. The height of the post­ vallum is much greater than the height of the prevallum. The upper molars are separated from each other by extremely deep inter-dental embrasures. A narrow basal cingulum runs completely around the pos­terior, lingual, and anterior faces of the paracone, connecting the metastylar spur with the parastylar spur. The parastylar spur is prominent and cuspi­date. The preparacrista is a sharp ridge running from the apex of the para­cone to a well developed stylocone. The metastyle, also well developed, is situated more labially than the stylocone. The ectoflexus is deep on all the upper molars. M-differs from M 1 in its more transverse outline, in its weak­er metastylar spur, and in its stronger parastylar spur. is nearly identical toM-in structure except for its smaller dimensions. The skull is only partially preserved. The rostrum and much of the post-orbital region of the cranium is entirely missing. The large anterior openingof the infraorbital canal is above P4. The very large lacrimal foramen is not hidden in lateral view by a flange from the anteorbital ridge. The anterior rim of the lacrimal is thick and torus-like. The posterior opening of the in­fraorbital canal is an extensive orifice with the outline of a vertically orient­ed ellipse. The lack of a depression on the maxillary immediately anterior to the orbit suggests the snout muscles were not well developed. The maxillaryhas no zygomatic process; instead, a small boss situated above the M-is the final buttress of the descending maxillary ridge. The widest part of the palate is between the second molars. This is also the region of the transverse portion of the maxillary-palatine suture. The ventral openings of posterior palatine canals are located on the maxillary-palatine suture immediately in front of the postpalatine torus. The maxilla does not extend backward along the side of the pterygoid lamina for any great distance. A postpalatine torus is only moderately developed. The ptery­goid process is a low, rounded protuberance. There is no evidence of a dis­tinct hamular process. There appears to be a large rounded sphenopalatine foramen in the orbital component of the palatine. A small opening located below and slightly pos­terior to the sphenopalatine foramen is probably the dorsal orifice of the postpalatine canal. A prominent torus overhangs the anterior lacerate fora­men (sphenorbital fissure). Fig. 6.—Apternodus cf. brevirostris TMM 40492-9. Partial skull with posterior premolar,molar dentition: (a) view ofleft side of skull, (b) ventral view. DISCUSSION Of the recognized species of Apternodus, only two, A. brevirostris and A. gregoryi, are represented by more than upper and lower dentitions. 1 he hypodigm of the type species, A. mediaevus (Matthew, 1903), consists ot a lower jaw with ?2, P4, M2, and M3, thus precluding its comparison with the Vieja apternodontine. Matthew (1910) described a beautifully preserved skull and jaws and re­ferred them to Apternodus mediaevus. Schlaikjer (1934), however, allocated this material to a new species, A. brevirostris. He distinguished A. breviros­tris from A. gregoryi (Schlaikjer, 1934) by a number of features in the skull and upper dentition. The Vieja apternodontine differs from Apternodusgregoryi in; (1) the less acutely triangular outline of the P4; (2) the postero­lingual, rather than anterolingual, slant of the lingual corner of the P4 crown; (3) the larger size of the P4 relative to the upper molars; (4) the much narrower basal cingula on the upper molars; (5) the deeper ectoflexi, particularly on m3; and (6) the more prominently projecting parastylar spurof m3. In these dental features, the Vieja apternodontine compares much more closely with A. brevirostris than with A. gregoryi. Unfortunately, the cranial features, including the diagnostic snout region, which Schlaikjer (1934) used for differentiating the latter two species, are not preserved in the Vieja skull. CONCLUSIONS Wilson (1971a, b, and various papers) has recognized the Porvenir and Little Egypt local faunas of the Vieja Group as early Oligocene in age. New evidence derived from fossil rodents (Wood, 1974) further supports this in­ terpretation. Wood (1974) observed a great similarity in rodent composition between the Porvenir local fauna and the McCarty’s Mtn. fauna ofMontana (Chadronian) and between the Little Egypt local fauna and the Pipestone Springsfauna ofMontana(Chadronian; somewhatyoungerthanthatofthe McCarty’s Mtn.). The small sample of Vieja insectivores described in this report provides minor biostratigraphic information, but they certainly do not conflict with interpretations for a Chadronian age for the Porvenir and Little Egypt local faunas. Leptictis wilsoni shows the closest morphological resemblance to L. acutidens from the Pipestone Springs locality, thus sup­ porting Wood’s (1974) contention for age correlation of the Little Egypt with Pipestone Springs. Apternodus cf. brevirostris from the Little Egypt local fauna does not resemble an undescribed species of Apternodus from McCarty’s Mtn. (McKenna, personal communication) but is closely similar to Apternodus brevirostris from Bates Hole locality (Chadronian) of Wyo­ ming. 17 REFERENCES BUTLER, F. M., 1956. The skull of Ictopsand the classification of the Insectivora. Proc. Zool. Soc. London, v. 126, pp. 453-481. , 1972. The problem of insectivore classification. In Studies in vertebrate evolution K. A. Joysey and T. S. Kemp (eds.), Winchester, New York, 284 pp. DEFORD, R. K., 1958. Tertiary formations of Rim Rock Country, Presidio County, Trans-Pecos Texas. Texas Journ.Sci., v. 10, no. 1,pp. 87-108. DOUGLASS, E., 1901. Fossil mammalia from the White River Beds of Montana. Trans. Amer. Philos. Soc., v. 20, pp. 237-279. , 1905. TheTertiary of Montana. Mem. Carnegie Mus., 11, pp. 146-199. LEIDY, J., 1868. Notice of some remains of some extinct Insectivora from Dakota. Proc. Acad. Nat. Sci. Phila., 1868, pp. 315-316. LILLEGRAVEN, J. A., 1969. Latest Cretaceous mammals of the upper part of the Edmonton Forma­tion of Alberta, Canada, and review of the marsupial-placental dichotomy in mammalian evolu­tion. Paleont. Contrib. Univ. Kansas, v. 50, p. 1-122. MATTHEW, W. D., 1903. The fauna of the Titanotherium beds of Pipestone Springs, Montana. Bull. Amer. Mus. Nat. Hist., v. 19, pp. 197-226. , 1910. On the skull of Apternodus and the skeleton of a new artiodactyl. Bull. Amer. Mus. Nat. Hist., v. 28, pp. 33-42. NOVACEK, M. J. (In Press). Insectivora and Proteutheria (Mammalia) of the later Eocene (Uintan)of San Diego County, California. Los AngelesCounty Mus. Nat. Hist. Contributions. RICH, T. H. V., 1971. Deltatheridia, Carnivora, and Condylarthra (Mammalia) of the early Eocene, Paris Basin, France. Univ. Calif. Publ. Geol. Sci., v. 88, pp. 1-72. SCHLAIKJER, E. M., 1934. A new fossil zalambdodont insectivore. Amer. Mus. Novit., no. 698,pp.'l-8. SCOTT, W. 8., and G. L. JEPSEN, 1936. The mammalian fauna of the White River Oligocene. Part I. Insectivora and Carnivora. Trans. Amer. Philos. Soc.;v. 28, pp. 1-287. STOVALL, J. W., 1948. Chadron vertebrates from below the Rim Rock of Presidio County, Texas. Amer. Journ. Sci., v. 246, pp. 78-95. VAN VALEN, L., 1966. The Deltatheridia a new order of mammals. Bull. Amer. Mus. Nat. Hist., v. 1 32, pp. 1-126. , 1967. New Paleocene insectivores and insectivore classification. Bull. Amer. Mus. Nat. Hist., v. 135, pp. 2 17-284. WEST, R. M., 1972. Upper deciduous dentition of the Oligocene insectivore Leptictis (= Ictops)acutidens. Ann. Carnegie Mus., v. 44, art. 3, pp. 25-32. WILSON, J. A., 1971a. Early Tertiary Vertebrate Faunas, Vieja Group, Trans-Pecos, Texas; Agrio choeridae and Merycoidodontidae. Tex. Mem. Mus. Bull. 18, 83 pp. , 1971b. Early Tertiary Vertebrate Faunas, Vieja Group, Trans-Pecos Texas: Entelo dontidae. Tex. Mem. Mus. Pearce-Sellards Series No. 17, 17 pp. WILSON, J. A., P. C. TWISS, R. K. DEEORD, S. E. CLABAUGH, 1968. Stratigraphic succession potassium-argon dates and vertebrate faunas, Vieja Group, Rim Rock Country, Trans-Pecos Texas Amer. Journ. Sci., v. 266, pp. 590-604. WOOD, A. E., 1974. Early Tertiary Vertebrate faunas, Vieja Group, Trans-Pecos Texas: Rodentia Tex. Mem. Bus. Bull. 2 1,112pp.