MAMMALIAN DAWN. 79 



food. The teeth are usually well developed, but incisors may 

 be entirely absent from the upper jaw, as in the ox, sheep, 

 and deer. 



At the ver\' lowest strata of the Tertiary is first found a group 

 of animals, the Condylarths, which is considered to represent 

 very nearly the central protungulate tj-pe from which all 

 hoofed animals are descended. The best known genus in this 

 ancient order is PJioiacodus, a skeleton of which {P. privKe- 

 vus) was found by Dr. Wortman in the Lower Eocene of the 

 Big Horn Basin, Wyoming, and described by Cope, who 

 regarded it as representing the form from which developed the 

 hog, hippopotamus, deer, camel, rhinoceros, and all other 

 hoofed animals. This very generalized mammal was about 

 the size of a sheep. It was digitigrade about as the tapir, 

 and had five fingers and five toes provided with nails which 

 were flat and could neither be called claws nor hoofs. It had 

 the full complement of teeth (44) which were bunodont, indi- 

 cating an omnivorous habit. The sides of the bod}- were 

 lank, back arched, tail long, and hind limbs longer and 

 stronger than the fore — cat-like characters linking it to the 

 primitive clawed stem from which the two great divisions of 

 mammals — the flesh eaters (Carnivora) and the herb eaters 

 (Ungulata) — sprung. 



While Phenacodus was much too large to be in the direct 

 line of the ancestors of the horse, its feet give us a ver}- good 

 idea of the character of those of the hypothetical five-toed 

 progenitors of the beginners of this series. The first undoubted 

 mammal of the equine evolutionary line appearing in North 

 America is the Eohippus, which is a little animal no larger 

 than a small fox. From this four-fingered and three-toed 

 ancestor we have an unbroken line, with intermediate steps in 

 development, through the stage with three fingers and three toes 

 and that with one functional f the third) and two rudimentary 

 (the second and fourth), until we reach the genus Eqiius with 

 theJateral digital columns entirely wanting and nothing left 

 of the lateral metacarpals and metatarsals but the two rudi- 



