On the Osteology of the Hyopotamidse. 167 



Paridigitata and Imparidigitata was a great gain to science ; it radi- 

 cally separated two groups that previously were always hopelessly 

 mixed together ; but now the same principle must be carried further. 

 The separation of the two groups of Paridigitata and Imparidigitata 

 took place in very ancient time, not nearer than the Cretaceous 

 period ; and the striking diversity exhibited by both groups from 

 the lowest Eocene is a proof of their ancient separation. But one 

 of the bi'anches, the Paridigitata, in its turn, split very anciently 

 again into two distinct groups, one with tubercular, the other 

 with crescentic teeth. This occurred at nearest in the Lower Eocene, 

 perhaps even in the Cretaceous period. These groups, once sepa- 

 rated, kept entirely apart and followed different lines of descent, 

 although the modifications which both undergo along the descend- 

 ing lines are parallel and analogous even to the greatest details. 

 Following these two divergent lines of descent, both groups culmi- 

 nate in the recent fauna in such forms as the PhacocJicerm and 

 Dicotyles for one group, and the Bovidre for the other. Links 



Lower Cretaceous Ungulata. 



Paridigitata. ImparidiKitata. 



O O 



Parid. with Parid. with 



tuberc. teeth. crescentic teeth. 



between them we discover none ; and to discover their parentage, 

 we must pass along the ascending lines to the point at which 

 they diverge, as the linking genera, which doubtless existed at 

 the time of separation, are long ago extinct, and both groups are 

 now widely separated. I suggested this view, whilst studying 

 in the British Museum the remains of Hyopotamido', to Professor 

 Owen ; and he finds no objection to it. He aided me in finding for 

 the two groups convenient names ; and by his suggestion I should 

 call the Paridigitata with crescentic teeth Par. selenodonta, and 

 those with tubercular teeth Par. bunodonta. 



To the Jirst gr^up would belong all the ancient and living Pari- 

 digitata having crescentic teeth, as the Anoplotherium, Xiphodon, 

 Dichobune, Anthracotherium, Bothriodon, IL/opotamus, Rhagatherium , 

 and the living Ruminantia. 



The second would embrace all the Suina, ITippopotamina, and 

 Entelodon. 



Each one of these two groups may be again si-.bdivided on the 

 principles adopted in this paper. 



