Mereiam.] 



Pliocene and Quaternary Canidae. 



2S7 



noticeable characters separating it from this group is the absence 

 of the most distinctive feature of the upper dentition of 

 Aelurodon, viz.: of the protostyle of P-. 



Some of the closest general similarities to any other group 

 that this form shows are its resemblances to certain of the species 

 of Palaeocyon ( Speothos) described by Lund* from the Brasilian 

 cave fauna. Palaeocyon was a short-nosed, broad-palated dog 

 with simple, crowded premolars and a very greatly reduced M- 2 -. 

 M- was also small and the superior sectorial had no deuterocone. 

 This genus is distinguished from the Californian form by its 

 relatively small I- ; minute M-; narrower and structurally 

 different crushing lobe of M 1 ; more slender facial region; and the 

 absence of the peculiar characters of the nasal region, jugal, and 

 posterior palatine foraminia found in H.f ?) diibius. The two 

 types are quite distinct and they may simply represent evolution 

 in the same direction along two different lines. Palaeocyon is 

 evidently a member of the Icticyon group in which extreme 

 reduction has taken place at the posterior end of the upper and 

 lower molar series before great crowding or elimination occurred 

 in the premolar series. In the Californian form we have prob- 

 ably an older species. It shows less of the special kind of molar 

 reduction than we see in the Icticyons, but is in many ways more 

 highly specialized. If we consider the Hyaenognathus mandible 

 as representing the same group as H.f?) dubius, we discover a 

 farther resemblance to the Icticyons in the absence of the meta- 

 conid from Mr. On the other hand the presence of three molars, 

 the peculiar reduction of the premolars and the presence of two 

 low tubercles on the heel of the sectorial show it to be distantly 

 removed from Palaeocyon. 



For the present we can not do more than consider this spec- 

 imen as representing a very aberrant type of dog, considerably 

 removed from any known group. As it belongs to a compar- 

 atively late epoch, we may hope to establish its relationship to 

 one of the older and better known groups when we learn more 

 of the Canidae of the West American Pliocene. 



Should this form prove to be identical with Hyaenorjnathus, 

 as has been suggested, we shall have made but little advance in 



* P. W. Lund. Blik paa Brasiliens Dyreverden, 1841-45. 



