;8 Bulletin American Museum of Natural History. [Vol. XIV, 



(In either case, this may indicate that neither Clmiodon nor 

 Anacodon is precisely in the line of Bear descent, but does not 

 seriously weaken the arguments given above as to the derivation 

 of the Bears from the family Arctocyonidae.) 



3. The real objection to the theory lies in the remarkable 

 approximation of certain of the Dog and Bear families in Upper 

 Miocene time. Although, as Schlosser has recently shown, 

 Hycenarctos and Amphicyon cannot stand in the direct line of de- 

 scent, yet they exhibit a striking gradation between the two 

 families and might well be considered as survivals of primitive 

 links connecting the two. Cephalogale also, among the more 

 ancient Canidse, shows several approximations to the Ursid», and 

 is placed by Dr. Schlosser as ancestral to the family ('99, 146). 



There are some characters, however, that have not been 

 bridged, and these are perhaps more important than they appear 

 at first sight. 



1. All Canidae have triangular upper molars. All Ursidae have 

 quadrate upper molars. 



2. All Canidge have the inner cusp of p^ anterior. All Ursidse 

 have it medial. 



The trigonal molar and the anteriorly placed trittocone are in- 

 dications of a formerly tuberculosectorial dentition, and appar- 

 ently very difficult to get rid of. Procyon, however, seems to 

 show us an earlier stage of their disappearance in a line descended 

 from the Canidae ; so that we may yet discover the intermediate 

 stages in the Ursid phylum. 



To sum up — the Arctocyons were progressing towards the Bear 

 line in all the most distinctive characters of both teeth and feet. 

 But the wide gap between Lower Eocene and Middle Miocene 

 makes any connection between the two somewhat uncertain. In 

 the Canid line, on the other hand, we have a number of appar- 

 ently intermediate stages known. But these intermediate stages 

 cannot have been actually in the line of descent, and even if 

 they are unaltered descendants of more ancient types we still 

 have a gap of some importance unbridged. The connection, 

 moreover, is based on teeth alone. Professor Osborn, in discuss- 

 ing the evolution of the Mammalia, remarks : " The teeth and 

 feet, owing to the frequent parallels of adaptation, may wholly 

 mislead us if taken alone ; while if considered together they give 



