MUSEUM OF COMPARATIVE ZOOLOGY. 



69 



rend die beiden iibrigen JElurodon- Arten sich hochst wahrscheinlich als Vor- 

 laufer der Hyanen erweisen werden." (Beitr. z. Palaont. Oesterr. Ungarns, 

 Bd. VIII. p. 252.) 



In these statements Schlosser has been misled by the fact that the specimen 

 of JElurodon scevus which was figured by Cope is very old, and the teeth so 

 much worn down that the anterior lobe of the upper sectorial is hardly distin- 

 guishable. The specimens before us demonstrate clearly that Cope's reference 

 of the species is correct, and that JE. Wheelerianus and hycenoides cannot be 

 generically distinguished from it. The only characters of JElurodon which in 

 any way resemble those of the hyaenas are (1) the massive premolars, (2) the 

 presence of an anterior basal cusp on the upper sectorial, and (3) the reduction 

 (in some species) of the postero-internal cusp of the lower sectorial. These 

 resemblances are obviously merely analogical, and are of far less importance 

 than the characters of the skull and limbs, which are distinctively cynoid. 

 These animals are genuine dogs, if somewhat peculiarly modified, and to regard 

 them as ancestors of the hyaenas is to ignore the close connection between the 

 latter and the viverrines, besides being improbable on geographical grounds. 



CANIS. 



?Canis vafer, Leidy. 



This small alopecoid is represented in the collection by a mandible with 

 broken teeth and some other fragments. It agrees almost exactly with Leidy's 

 type {op. cit.j Plate I. fig. 11), except that the diastema between the canine and 

 plri. 1 is shorter. The small size of the sectorial places the species in the nii- 

 crodont division of the alopecoid series. M. 2 is very elongate an tero- posteriorly, 

 and m. 3 is implanted by two fangs. The mandible is very slender, much 



Figure 3. — ? Canis vafer X f. A. First superior molar; B. Mandible. 



curved, and non-lobate. The first upper molar is nearly quadrate in shape, the 

 metaconule being almost as large as the protocone, and placed upon nearly the 

 same antero-posterior line. The cingulum is internally very greatly enlarged 

 and thickened, and is disposed symmetrically around the inner side of the 

 crown, instead of being confined to the postero-internal angle, as is usual 

 among the recent Canidce. 



