1880.] DENTAL CHARACTERS OF THE CANID^. 261 



differs from tliat of other Canidae and approaches the characters of the 

 same tooth iu the Viverridse and, especially, in the Procyonid£e. 



It is obvious that, in all these features, Otocyon represents a lower 

 term iu the series of the Alopecoids than C. cinereo-arfjenteus and 

 C. velox ; and although the interval between these and Otocyon is 

 as wide as that between C. velox and C. vxdpes, there would be no 

 reason for separating Otocyon from the rest of the group were it 

 not for the extraordinary excess in the number of molar teeth. Of 

 these there are four on each side above in one specimen, three in the 

 others, four on each side below in all. So far as any conclusion can 

 be drawn from this limited number of examples, therefore, it would 

 appear that the presence of the fourth upper molar is exceptional ; 

 and that the dentition is tending towards a higher type by its sup- 

 pression. The crowns of the hindermost upper molars, in the one 

 specimen in which they are shown, are of a triangular shape, the 

 base of the triangle corresponding with the two distinct external 

 cusps. On the inner side only one cusp remains. The crown of 

 the small fourth lower molar exhibits two well-developed cusps, of 

 which the outer is rather the smaller. These apparently represent 

 the anterior cusps of the other molars, as the posterior pair of cusps 

 are proportionally smaller than the anterior pair in the third molar. 

 10. The facts now adduced appear to me to permit of the arrange- 

 ment of the Canidse hitherto considered in the following manner: — 



Canid^. 

 I. Molars H 



3.3 



A. Alopecoids. B. Thooids. 



a. Macrodonts. 



Ex. C. argentatus. Ex. C. lupus. 



C- vul^jes. c. aureus. 



b. Microdouts. 



a. Mandible nonlobate. 

 Ex. C. caama. Ex. C. azarce. 



C. zerda. C. vetulus. 



/3. Mandible lobate. 

 Ex. C. littoralis. Ex. C. cancrivorus. 



II. Molars ^^. 



4.4 



C. Otocyonoids. 

 Otocyon. 



It will, however, be readily understood from what has already been 

 said, that It IS not intended to suggest the possibility of sharnlv 

 separating the macrodont from the microdont forms, or those with 

 k.bed mandibles from those with mandibles of the ordinary character 

 Un the contrary, they pass into one another ; while the lower Thooids,' 



