Mimotricentes tedfordi, McKenna & Lofgren, 2003

McKENNA, MALCOLM C. & LOFGREN, DONALD L., 2003, Chapter 25 Mimotricentes tedfordi, a New Arctocyonid from the Late Paleocene of California, Bulletin of the American Museum of Natural History 279 (279), pp. 632-643 : 635-641

Genus Mimotricentes Simpson, 1937

Mimotricentes tedfordi , new species

? Chriacus . sp., Douglass Quarry, Fort Union Group, earliest Tiffanian, eastern Crazy Mountain Basin, Montana ( Krause and Gingerich (1983: 189).

HOLOTYPE: RAM 6908, heavily damaged right maxilla with slightly damaged P4–M3 ( fig. 25.2 ).

LOCALITY AND HORIZON: RAM Locality V 94133 (= UCMP Locality V 5870), near the base of member 4b ( Cox, 1982) or informal unit Tgsu ( Cox and Diggles, 1986) of the Goler Formation of Dibblee (1952), at the same stratigraphic level as RAM Locality V 94014 (= UCMP loc. V 5252); mammalian fossils from localities V 94133 and V 94014 are referred to as the Laudate local fauna ( McKenna, 1955, 1960; McKenna et al., 1987; Lofgren et al., 1999). These localities lie along the southern border of the USGS Inyokern SE 7 1/2 minute Quadrangle, Kern County, California, 1972 ( NAD 27).

AGE: Tiffanian.

DIAGNOSIS: Size significantly smaller than the sample referred to Mimotricentes subtrigonus from Swain Quarry in the Torrejonian of southern Wyoming (see discussion below), Mimotricentes elassus from the Torrejonian of the Dragon local fauna, Utah ( Gazin, 1941: 22–23), and all other described species of Mimotricentes and Lambertocyon . M. tedfordi apparently differs from the hypodigm of M. elassus in possessing a projecting and prominent parastyle on M1 and M2, and possessing strong mesostylar development on all the upper molars. Posterior and anterior cingula do not meet in a continuous cingulum around the base of the protocone on the molars.

ETYMOLOGY: tedfordi in honor of Richard H. Tedford, whose early prospecting efforts in the Goler Formation helped push back the history of land mammals in coastal western North America by about 15 m. y., and whose discovery of the first Goler mammal tooth sparked 50 years of subsequent exploration that eventually led to discovery of the species named here in his honor.

DESCRIPTION OF THE TYPE SPECIMEN

We emphasize that the following description is of a single specimen drawn from what, by analogy with other samples, was probably a rather variable population. The cluster of species to which M. tedfordi belongs is much in need of revision, but this will have to await the collection of much new material. Especially important will be the collection of statistically adequate samples from single sites, as well as acquisition of associated upper and lower teeth. Some of these taxa are presently known from upper teeth, others by lowers, and nearly all by fragmentary material only.

Not much can be said about the maxilla itself, except that the anterior end of the zygomatic arch arises opposite M2. Part of the passage of the infraorbital canal can be seen, but its anterior opening can only be stated to lie anterior to the level of P4.

P4 is damaged anterolabially, and also on the apices of the incompletely separate paracone and metacone, but otherwise the tooth is intact. There are three roots. The extent of the missing parastylar area cannot be determined. P4 has a steep anterior side and a sloping posterior side, resulting in a tilted appearance. The enamel surface is corrugated, particularly on the labial side of the high protocone and on the posterolingual wall of the metacone. The high paracone and somewhat smaller metacone are conjoined for much of their height, with the paracone bulging farther labiad than the metacone. A crest was present on the anterior side of the paracone. The metacone is continuous with a sharp ridge that connects to the metastylar area at the juncture of the posterior and labial cingula. There is no hypocone, and no metaconule is present. A tiny paraconule may have existed at the anterolingual base of the paracone. A broad, flat, corrugated cingulum runs as a shelf nearly around the tooth from the metastylar area around the protocone (with a short interruption at the lingual base of the protocone), continuing along the anterior base of the paracone, until encountering breakage in the parastylar area, where its nature is unknown. Midway along the anterior side of the tooth, the anterior cingulum rises in the occlusal direction. On the labial side of P4 the cingular shelf is narrow around the base of the paracone, but it broadens labial to the metacone, forming a labial lip of a small basin. There is no mesostyle.

M1 is a smaller tooth than M2, is three­rooted, and has a strongly projecting parastyle. The paracone is a high cone, well in front of the metacone, bearing two crests: an anterior crest connects anterolabially to the posterolingual base of the parastyle as on M2, and a posterior crest takes a slightly sinuous course rearward, connecting to the metacone apex. The metacone itself is now a peculiar crestlike cusp whose present shape may be caused by wear. The remaining labial crest of the metacone lies anterolingual to the slightly damaged metastylar area and turns linguad at its junction with the crest coming from the paracone. From the junction, the lingual continuation of the crestlike, worn metacone continues until it meets the metaconule. The enamel of the posterior slopes of both the metacone and metaconule is heavily worn, but the shapes of the metacone and metaconule are nonetheless distinctive. If any corrugation of enamel surfaces was originally present, it is worn away. Both the anterior and posterior cingula are strong, but not so shelflike as on P4 and M2. The posterior cingulum is heavily worn. The anterior and posterior cingula do not connect across the broad, lingual base of the protocone. The protocone (when unworn) was evidently quite high, but before the animal’s death it was substantially lowered by wear, which has resulted in a transverse wear trough that seems to divide the worn protocone into anterior and posterior parts. The paraconule is small, with a faint posterior connection to the base of the paracone and a stronger anterolabial ridge that runs along the anterior base of the paracone. On the posterolingual slope of the protocone a hypoconal shelf is present. The shelf is worn, but evidently was not so prominent as that of M2. A small, well­worn hypocone is present as a swelling of the posterior cingulum. It does not project lingual to the level of the protocone base. Posterior to the large, crested parastyle is a small stylar cusp, but there is no labial cingulum around the labial base of the paracone. However, posterolabial to the base of the paracone, a massive stylar ridge begins at a high stylar cusp (mesostyle?) and continues in a curve, terminating in a small, slightly damaged metastyle.

M2 is decidedly less worn than M1, and, although larger and a bit more transverse, is similar in most respects. There are three roots. The nearly unworn labial cingulum has the same curved crest as M1, running posteriad from a large mesostyle­like cuspule, in front of which is another, smaller, mesostyle­like cuspule lying at the posterolabial corner of the base of the broken paracone. The metacone of M2 is normal in structure, suggesting that the condition observed on M1 is due to wear. Crests lead posterolabially and anterolabially from its apex. The crest from paracone to metacone is interrupted, with a third mesostyle­like cuspule present at the base of the crest leading anterolabiad from the metacone. At the anterolabial base of the paracone, yet another stylar cuspule lies just posterior to the curving posterior crest of the parastyle. From the middle of the parastyle, a posterolingually directed crest runs to the paracone base as well, creating a small stylar basin on its posterolabial side. The anterolingual end of the parastyle hooks around to the rear lingually, before dying out in a series of three steplike corrugations. The hypoconal shelf is strong and worn. Lingually, after swelling to a small hypoconal apex, the hypoconal shelf curves forward, enclosing a tiny basin, before joining the posterior base of the protocone. There is no cingulum across the lingual base of the protocone, and the hypoconal shelf does not project linguad of the level of the base of the protocone. The paraconule and metaconule of M2 are both rather worn, but each exhibits two labial wings. The anterior wing of the metaconule is weak, but the anterior wing of the paraconule forms a strong crest that runs along the anterior base of the paracone until it meets the stepped corrugations of the lingual crest from the parastyle. Lingually, the anterior cingulum is slightly broadened, suggesting an incipient pericone (protostyle) just anterior to the lingual base of the protocone. The enamel of M2 is worn, but not completely enough to obscure traces of fluting on the labial slope of the protocone.

M3 is three­rooted and similar to M2, but is smaller and has a much smaller metacone. A faint hypoconal shelf is present, and a series of tiny cuspules decorate the anterior cingulum, terminating lingually in a tiny pericone­like cuspule like that adumbrated on M2. The anterior and posterior cingula do not connect as a continuous cingulum across the lingual base of the protocone. The stylar shelf possesses a row of three small mesostyle­like cuspules. Another pair of mesostyle­like cuspules lies at the juncture of the ridges joining the paracone and metacone to each other. The metaconule lacks an anterolabial wing, but the anterolabial wing of the paraconule is strong and similar to that of M2, and corrugated all the way to the parastyle.

Tooth measurements for RAM 6908, the type specimen of Mimotricentes tedfordi , new species, are given in table 25.1.

COMPARISONS

Tricentes , Mimotricentes , and Lambertocyon are closely related Paleocene arctocyonids occurring in the Torrejonian through Clarkforkian NALMAS. Tricentes Cope, 1884 , was based on the type species, T. crassicolidens Cope, 1884 , from New Mexico. Simpson (1935) named a related taxon Metachriacus , based upon M. punitor Simpson, 1935 , a taxon from Montana said by Van Valen (1978) to be a junior synonym of Chriacus orthogonius Russell, 1929 , from Alberta. Van Valen and Sloan (1965: 745) opined that T. crassicolidens is a synonym of ‘‘ Chriacus ’’ truncatus Cope, 1884, and that Tricentes is also a ‘‘senior synonym of Epichriacus [ Scott, 1892], Metachriacus , and possibly Prothryptacodon [ Simpson, 1935].’’ ‘‘Other species’’ of Tricentes were placed by Van Valen and Sloan in Mimotricentes Simpson, 1937 , whose type species is Mimotricentes latidens Gidley in Simpson, 1935, and to which genus M. angustidens Simpson, 1937 , was later assigned by Simpson. By ‘‘other species’’ of Tricentes, Van Valen and Sloan presumably meant Tricentes subtrigonus Cope, 1881 , T. elassus Gazin, 1941 , and T. fremontensis Gazin, 1956 . Williamson (1996: 65) synonymized T. ellassus with M. subtrigonus , but gave no reasons. McKenna and Bell (1997) included Tricentes crassicolidens in Chriacus , whose type species is Lipodectes pelvidens Cope, 1881 .

Krishtalka et al. (1975) reported Mimotricentes sp. from the late Tiffanian of the Badwater Creek area, Wind River Basin, Wyoming, but the animal in question is too large to belong to M. tedfordi : M1 length 6.3 mm (N = 2), width 6.5–6.7; M2 length 6.9 mm, width 8.1 mm. Possibly this sample belongs to Mimotricentes fremontensis , or Lambertocyon eximius as Gingerich (1979) postulat­ ed, but additional specimens are necessary before a firm decision can be reached.

Van Valen (1978: 57) described Mimotricentes mirielae , n. sp., based upon AMNH 58219, a fragmentary lower jaw with onehalf m2 and m3 from the Puercan of New Mexico. Although smaller than M. subtrigonus , the species cannot be compared adequately with M. tedfordi , the samples being inadequate and there being no comparable teeth.

Rose (1981a) reported cf. Tricentes sp. from the Clarkforkian of Wyoming, but the material on which the report was based may represent Mimotricentes or Lambertocyon instead. The Clarkforkian animal is larger (M1 or M2 length 5.65 mm, width 6.50 mm) than M. tedfordi and is otherwise similar to some specimens of the Swain Quarry Torrejonian sample referred to Mimotricentes subtrigonus . Rose (1981b) reported both Mimotricentes sp. and Tricentes cf. T. punitor from Rock Bench Quarry in the late Torrejonian and? Mimotricentes sp. in the mid­Tiffanian Cedar Point Quarry from the Fort Union Group of northwestern Wyoming.

Gingerich (1978) named Mimotricentes ischyrus from the Clarkforkian of Plateau Valley, Colorado. The type specimen of the species was a lower jaw with p4 and m1, which is not directly comparable with the type maxilla of M. tedfordi but evidently was a larger animal, based on measurements of its lower teeth compared with lower teeth of animals also known from upper teeth. In 1979, Gingerich created the genus Lambertocyon for what appear to be advanced (read ‘‘late’’) species formerly placed in Mimotricentes . The type species of Lambertocyon is L. eximius Gingerich, 1979 , from the late Tiffanian of Northwestern Wyoming and western Texas ( Schiebout, 1974; Gingerich, 1979). Gingerich transferred Mimotricentes ischyrus to Lambertocyon . M. tedfordi is significantly smaller than either species of Lambertocyon .

Fox (1990) reported? Mimotricentes sp. from Cochrane II, a Tiffanian locality in southwestern Alberta. Cochrane II occurs in the Porcupine Hills Formation, assigned in age to the early Tiffanian. The specimen upon which the report was based has not been described and cannot be assessed here.

Comparisons of the type specimen of Mimotricentes tedfordi with closely similar taxa are made difficult by the fact that only one statistically useful sample of a quarry population of Mimotricentes exists. This is the material referred to M. subtrigonus that occurs at Swain Quarry in the late Torrejonian of Carbon County, southern Wyoming ( Rigby, 1976, 1980). Other related species are represented at present by few specimens at any particular site, whose variability cannot be assessed except by analogy with the Swain Quarry sample.

Prior to Rigby’s (1976, 1980) description of specimens from Swain Quarry, which were referred by him to Mimotricentes subtrigonus , and the description of Lambertocyon eximius ( Gingerich, 1979) , upper teeth of Mimotricentes ­ like mammals were virtually unknown except for Mimotricentes elassus from the Dragon local fauna of Utah ( Gazin, 1941) and two unnumbered upper teeth from the Fort Union Formation of Montana in the AMNH collection (now numbered AMNH 35433). These latter are from Locality 81 of Simpson (1937), which lies 91 m above the base of ‘‘Fort Union No. 2’’ (upper Lebo Formation of Fort Union Group) in the Crazy Mountain Basin. Simpson (1937: 205) did not figure these teeth, left M2 and M3, but he did discuss them: ‘‘Like the lower molars, they closely resemble those of Tricentes , the only clear difference, and this of doubtful value, being that the internal cingulum does not circle the protocone and that on M2 the external cingulum does not cross the paracone.’’

Rigby (1980: 103) did not illustrate upper teeth from the large Swain Quarry sample, but his descriptions and measurements of them are worth repeating here:

‘‘All dental elements from Swain Quarry are inseparable from M. subtrigonus from the Nacimiento Formation of the San Juan Basin. Because of the lack of associated material and because of their general similarity to a number of other taxa, no isolated P4’s have been assigned to the taxon. M1–2 show variably developed cingula around the base of the protocone. M2’s have a complete cingulum although a few show a continuous wrinkle. The cingulum of M1 has a suggestion of continuity but is generally incomplete. M2’s also show a mesostyle in various forms of development on the external cingulum or in the centroloph if the external cingulum is reduced. Some M1’s and M3’s also show the phenomenon but substantial development is rare. The conules are variably developed although the metaconule is generally larger and all crests are variably inflated. The parastyle can be either strong, separate and inflated, or may be present only as a small cuspule at the confluence of anterior cingulum and preparacrista. M1 and M2, when represented only by isolated teeth, may be easily separated because of the more quadrate nature of the M1. M3’s are exceptionably variable, particularly in width, which affects the metaconule, posterior cingulum, and confluence of the labial and posterior cingula.’’

Rigby (1976, 1980) provided statistical calculations based on measurements of upper molars referred to Mimotricentes subtrigonus as part of his table 38. These are given in part in table 25.2.

We have omitted calculations concerning M3 width from Swain Quarry because of an error: the mean (6.27) given by Rigby (1976, 1980) falls outside the OR (3.45–5.95), an impossibility. M3 width is quite variable in any case, and indeed, the only measurement of M. tedfordi to fall within any observed range of a character of referred M. subtrigonus from Swain Quarry is M3 width.

Measurements provided by Rigby (1976, 1980) show that the type specimen of Mimotricentes tedfordi differs by the following amounts from the referred Swain Quarry M. subtrigonus specimens, measured in multiples of standard deviations, s, from the mean, M: length M1: X3.75; width M1: X3.97; length M2: X3.14; width M2: X3.99; length M3: X2.24. These results affirm that M. tedfordi is statistically highly unlikely to be a small specimen of the same taxon as that recovered from Swain Quarry.

The type specimen of M. tedfordi is about 84% the size of the type specimen of Mimotricentes elassus , from the Torrejonian Dragon local fauna, Utah, but the length and width of M1 of that specimen, 5.1 and 5.6 mm, respectively, fall within the observed ranges exhibited by the Swain Quarry sample referred to M. subtrigonus by Rigby (1976, 1980). If the Swain Quarry sample is truly a single species, possibly M. elassus is synonymous with M. subtrigonus . We leave the matter open.

In his description of ‘‘ Tricentes .’’ fremontensis from the Tiffanian Saddle and Ledge localities, Bison Basin, Wyoming, Gazin (1956) was able to assign lower teeth only. These are clearly too large to belong to an animal the size of M. tedfordi : m1 length 6.2 mm; m2 length 6.4 mm; m3 length 6.6 mm.

From the previous discussion it is evident that, based on size and on morphology (when described), Mimotricentes tedfordi is distinct from all named taxa of arctocyonids investigated. However, Krause and Gingerich (1983) described and illustrated, but did not name, a species of small arctocyonid referred to by them as Chriacus sp. , from Douglass Quarry in the early Tiffanian of Montana. Although about 25% larger than the type specimen of M. tedfordi , these teeth are quite similar to and may be members of a species of the genus Mimotricentes rather than Chriacus . They seem to be closely related to M. tedfordi based on morphology. A Tiffanian age of the Laudate local fauna is thus additionally supported.

Gradually, enough specimens have been recovered from the Goler Formation to establish that the age of the part that bears the Laudate local fauna is Tiffanian (late Paleocene) in age. This age determination is compatible with data from overlying early Eocene marine fossils found near the exposed top of the Goler Formation that are correlative with the marine Penutian Stage, planktic foraminiferal zone P8 (and P9?) and Nannofossil zones NP12 and/or NP13 (see McDougall, 1987, and references therein). The age determination for beds stratigraphically beneath the early Eocene marine fossils is based on first appearances of such taxa as cf. Phenacodus , cf. Neoliotomus , and Plesiadapis . However, this puts the Conacodon ­ like taxon from the Laudate local fauna ( McKenna, 1955, 1960; McKenna et al., 1987) in a new light. It must be a Tiffanian relict taxon with Puercan relatives elsewhere.

Figures

Fig. 25.1. Map of the El Paso Mountains, Kern County, California, showing (A) position of Tedford Discovery site for Paleocene vertebrates and (B) site of discovery of Mimotricentes tedfordi. Goler Formation shaded; triangle marking Black Mountain points to geographic north. Modified from Cox (1987: fig. 1).