180 TEN years' progress in vertebrate paleontology 



the other three superfamilies are represented today by a single family 

 each. The living species in their present natural habitat, with the excep- 

 tion of the American Tapirs, are all confined to the Old World, a fact the 

 more interesting in that North America seems to have been the birth- 

 place or at least the stage for the development, not only of the early rep- 

 resentatives of all the living Perissodactyls, but of most of the extinct 

 groups of the order as well. More than half the total number of Perisso- 

 dactyl species described have been founded on specimens from the Ter- 

 tiary and Quartenary formations of this country. 



The earlier theories held by Cope, Osborn, Sclilosser, and others respect- 

 ing the ancestry of the Perissodactyla and the relationship of this order, 

 as well as that of the Artiodactyls to the Condylarths, have been for the 

 most part abandoned, and on very good ground. Matthew,^ in his re- 

 vision of the Puerco-Terrejon fauna in 1897, pointed out in Euprotogonia, 

 a formerly proposed ancestor of the Perissodactyla, certain characters of 

 foot structure, which, while denoting a position for this genus ances- 

 tral to Phenacodus of the Wasatch, shoAved conclusively that it could not 

 be regarded in any way ancestral to the Wasatch Perissodactyls. No later 

 discoveries have thrown any real light on the early development of this 

 great order ; hence our present knowledge extends no farther back in time 

 than to the Wasatch Eocene in America and equivalent horizons in 

 Europe. Here we find its representatives not only sharply and distinctly 

 separated from the Artiodactyls, the order to which the Perissodactyls 

 are supposed to be most closely related, but some of the natural primary 

 groups of the order are already definitely indicated. The origin of the 

 order and even its true position relative to the Artiodactyla, therefore, are 

 still unknown and are important problems, depending for solution on 

 future exploration and discoveries in the fossil-l)earing horizons of the 

 basal Eocene or perhaps even an earlier period. 



The acceptance in recent years of the very important Poh^hyletic 

 theory, or law, has led to the abandonment also of the monophyletic 

 origin theory and the attempts to trace out direct ancestral lines, such as 

 was so elaborately done by Marsh for the development of the horse. The 

 polyphyletic theory has been verified by Osborn in his study of the Titano- 

 theres and Ehinoceroses, and by Osborn and Gidley for the horses and 

 their relatives of the Tertiary formations. 



Thus the Titanotheres of the Oligocene, instead of being a diversified 

 group derived from a single Eocene ancestor, are now known to belong to 

 at least four contemporaneous phyla, represented by as many distinct 

 types in the Eocene. Likewise the rhinoceroses show a poh^hyletic origin, 



' Bull. Amer. Mus. Nat. Hist., vol. ix, pp. 30S-310. 



