108 



University of California Publications in Geology 



[Vol. 10 



of Oregon, and there can be little doubt that, as suggested by Freuden- 

 berg, the form represents the general group of Hyaenarctos. The 

 closer relationships can hardly be determined with certainty, but it 

 is of great interest to note that this group is represented in beds of 

 approximately lower Pliocene or late Miocene age in Mexico, adjacent 

 to the South American region, in which the arctotheres were so largely 

 developed in the Pleistocene. 



Figs. 23a and 23b. Hyaenarctoft or Indarctos. Mj, natural size. Fig. 23a, 

 outer view; fig. 23b, occlusal view. From late Cenozoic deposits, Tehuichila, 

 Mexico. (Drawn from cast furnished by Dr. W. Freudenberg.) 



Two forms from South America described by Ameghino as Par- 

 arctotherium and Proarctotherium have been assumed to be ancestral 

 types of the South American arctotheres. Pararctotheriuni is 

 known from the Pampean formation, which corresponds to a eonsider- 

 able portion of the Pleistocene of North America. Pararctotheriuni 

 is clearly a member of the Arctotherium group in which M 5 is con- 

 siderably specialized. The anterior premolars of Pararctotheriuni are 

 crowded, and the facial region was evidently shortened. There is no 

 good reason for considering that this type is ancestral to Arctotherium; 

 in fact, it may lie one of the more specialized members of the Arcto- 

 tln rium group. 



Proarctotherium is known from a very fragmentary specimen 

 consisting of Mj and Mg from Parana, The Parana deposits have 

 been assumed to be Pliocene, and may contain a small number of North 

 American Pliocene types. Ameghino 's s figure of the type specimen 



8 Ameghino, F., Contribucion al conocimiento de los mamiferos fosiles de la 

 Eepublica Argentina, p. 319, pi. 21, fig. 1, 1889. 



23a 



