PEDIGREE OF MAMMALS. 273 



of the placenta might similarly arise in different ways, 

 just as it has been variously developed in the well-sub- 

 stantiated division of the Ungulata. To corroborate our 

 view by example, we are certainly unable to make any 

 positive statements as to the derivation of the Probos- 

 cidae. It is, however, none the less certain that nothing 

 positive is implied by the customary classification by 

 reason of their zonary placenta. But we shall more 

 nearly approach the truth if we place this branch of 

 unknown origin typically nearer to the Ungulata than 

 to the beasts of prey. If, moreover, as non-deciduate 

 mammals, the Cetacea are held to be more closely 

 allied to the Ungulata than to the Carnivora, which 

 are deciduate, in our eyes, this circumstance is not 

 decisive, as more important reasons argue that the 

 Cetacea were first developed from carnivorous genera. 



In our exposition of the geographical distribution of 

 animals, we derived instruction from Riitimeyer with 



Horses. 



Hipparion. 



Macrauchenidae. 



Anchitherium. 



\ I 



Palaeotheridae. 



reference to the relationships of the Ungulata in 

 particular. In no other division do we possess such 



