THE CANIDJE, OR DOGS. 269 



tribe is to be found in their relation to the spoon- 

 dog. What Huxley states regarding the simi- 

 larity between its dentition and that of the lower 

 bear-like genera is certainly well worth conside- 

 ration, but is of less importance than the con- 

 clusion he draws from a discovery of his own. In 

 several hundred different species of Dogs he 

 found fibrous formations which are said to corre- 

 spond with the marsupial bones (ossa epipubica), the 

 distinguishing feature of the Marsupial group. If 

 this observation becomes an established fact, the 

 direct descent of dogs from Marsupials would seem 

 in the highest degree probable. However, as one 

 of our first comparative anatomists has maintained, 

 we still require further proofs for Huxley's observa- 

 tion. In imagining the Dogs connected with the 

 Marsupials we should not, in the first instance, 

 have to consider our present carnivorous Marsu- 

 pials (Thylacmus, Dasyurus), whose row of molars 

 consist of one tooth less than that of Otocyon, 



and are generally characterised as p -, m -. The 



' 4 



Marsupial Eats would more likely have to be taken 

 into consideration. They are the only known 

 animals from the Eocene with four molars. More- 

 over, by using the flat part of their hands and feet, 



