62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I44 



teeth, there is a maxilla with M^ and M^ (Y.P.M. No. 14077 ; see 

 pi. 9, fig. I ) from Bitter Creek that can be closely matched in Gray 

 Bull materials from the Bighorn Basin. Lower jaws (U.S.N.M. Nos. 

 22429 and 22427; see pi. 9, figs. 2, 4) from the Knight exposures at 

 the base of the escarpment west of Elk Mountain in the Fossil Basin 

 have anterior molars comparable to those in larger Gray Bull materials, 

 but Ms in both is actually and relatively smaller than in all but one 

 of the U.S.N.M. jaws from the Gray Bull with this tooth preserved. 

 A lower jaw with P4 and Mi (U.S.N.M. No. 22428; see pi. 9, fig. 3) 

 from the Red Desert locality east of Steamboat Mountain can also be 

 matched in Gray Bull specimens, but only among those having the 

 smallest teeth. 



There is considerable variation in tooth size in various Gray Bull 

 materials referred to Haplomylus speirianus. Simpson (1937, p. 22), 

 however, has shown that the materials from the Sand Coulee or low- 

 est Wasatchian levels average distinctly smaller than those from the 

 Gray Bull proper, although there is an overlapping in range. Teeth 

 decidedly smaller than any of those in the National Museum Gray 

 Bull sample were found very low in the Knight at a locality southeast 

 of the Rock Springs uplift, approximately 20 miles southwest of 

 Bitter Creek Station, by Henry W. Roehler of the Mountain Fuel 

 and Supply Co. This must surely be the Sand Coulee level. 



HYOPSODUS LOOMISI McKenna 

 (Plate 9, figures 5-7) 



The species Hyopsodus loomisi as proposed by McKenna would 

 appear to be a substitution for Hyopsodus simplex to include much 

 that had been attributed to that species but which was not determin- 

 able in a lower jaw, or not actually distinctive of the type specimen. 

 Most of the characters attributed to Hyopsodus simplex are highly 

 variable throughout the lower Eocene but appear more consistently 

 evident in the earliest Gray Bull specimens. The characters selected 

 by McKenna as diagnostic for H. loomisi include the weakness of 

 the hypocone and near absence of an external cingulum of the upper 

 molars, which when coupled with small size would seem indicative 

 of a distinct species, or at least of a primitive mutant of the Gray- 

 bullian form referred to H. miticulus. 



Several very small jaws in the Bitter Creek collection are believed 

 to be this species, but only one upper molar conforms to the descrip- 

 tion. Other Bitter Creek maxillae show better development of the 

 hypocone and external cingulum. Among the very small lower jaws 



