THE CANID^, OK DOGS. 285 



animals owing to the above-mentioned peculiari- 

 ties show a certain necessity for being classed 

 together, and it is self-evident that they must stand 

 nearer to one another, were it only on account of 

 the smaller period of time since their separation 

 from the primary stock, which must be assumed to 

 have been common to them all. However, in our 

 opinion, the characteristics specified above go no 

 further than to indicate a most general form of con- 

 nection. .When Cope maintains that the different 

 groups of Bunotheria are related to one another, 

 somewhat in the same way as are the orders of 

 Marsupials, it seems to us that this scarcely applies 

 to the case. To give an example : what has Tillo- 

 therium from the Eocene of Wyoming (Fig. 50) in 

 common with the flesh-eaters Arctocyon and 

 Oxytena? By the smallness of its brain most 

 certainly nothing ; nor by its five toes, and the cir- 

 cumstance that the whole sole of the foot is applied 

 to the ground. The resemblances in the molars are 

 not remarkable, while the decidedly rodent-like form 

 of the incisors only proves the peculiarity of the 

 animal. Any typical feature such as the marsupial 

 pouch or the marsupial bones, or the openings of 

 the urinary and genital organs peculiar to the Mar- 

 supials is not met with in any of these animals, 



