196 ANIMAL LIFE PAST AND PRESENT. 



and-dried schemes. Thus we find that while some 

 palaeontologists have regarded this troublesome amphi- 

 cyon as a plantigrade dog, and consequently place it 

 among the canoid animals, other writers consider it 

 as a dog-toothed bear, and refer it to the arctoid group. 

 The truth is, however, that we must recognise in this 

 early canivore the common ancestor from which both 

 bears and dogs have taken their origin; both groups 

 having thence tended more and more to diverge from 

 one another with the advance of time, till they finally 

 have assumed their present extreme differences, which, 

 as we have already observed, are apparent to the most 

 unscientifically disposed mind. Accepting this con- 

 vergence of the two groups as a proved fact, we must 

 accordingly regard the separation of the dogs and bears 

 into two distinct families as merely a convenient 

 classification for their existing representatives, rather 

 than a real primary division. 



The interest of the ainphicyon does not, however, 

 cease with its relationship to the bears and dogs, since 

 it also shows evident signs of cousinship with other 

 lower Miocene and upper Eocene carnivorous animals, 

 which appear to connect the dogs with the civets, and 

 thus with the hyaenas and cats. One of these early 

 connecting types is known as the dog-civet (Cynodictis), 

 and appears to have been an animal closely allied botli 

 to the amphicyon and to the true civets ; so that we 

 have good evidence for regarding bears, dogs, and 

 civets as having originated from one common stock 

 more or less closely allied to our ainphicyon. "Since, 



