RELATION OF PHEXACODUS TO AMELYPODA. 261 



US3 on all the feet, of which the first is the smrJlcst ; the 

 median is the largest aud is symmetrical within itself. The 

 feet are considerably shortened and were probably semiplanti- 

 grade ; in fact the feet of this animal constitute an approach 

 to the Amblypoda* The dental formula is : Incisors, 3—3, 



fined genera, which arc not positively known to have existed later than 

 the upper Eocene epoch. It may be recognized \,1) by the possession of 

 four toes on the anterior and three on the posterior limbs ; (2) by the 

 molar and premolar teeth being different ; (3) by the non-separation of the 

 anterior and posterior external cusps of the superior molars by an external, 

 rib-like pillar. The next family is the Chalicothenidce, to which ten genera 

 are referred. The digital formula is the same as in the Lophioilontidce. as 

 is also the relation of the molar and premolar teeth. The only distinction 

 is found in the separation of the anterior and posterior external lobes by a 

 vertical ridge. The remains of this family range from the lower Eocene to 

 the middle Miocene. The third family is the Pateotheriidce, having three 

 toes on each foot. The molars and premolars are alike, and the inferior 

 molars possess perfect double crescents. The fourth family is the Equidce, 

 in which the digital formula is reduced to one toe on each foot. The mo- 

 lars and premolars are alike and highly complex in structure. It is to this 

 family that all the existing horses belong, and it has been traced as far 

 back as the upper Miocene strata. The CondyloMJira, on the other hand, are 

 effectually separated from the Feri'SOdactyla by the non-alternating posi- 

 tions of the carpals and by the possession of an astragalus whose distal face 

 is convex in every direction, as in the carnivora, and unites with the navic- 

 ular alone. These families are the Phenacodontidce and Memscotheriidc^ 

 whose remains have been found so far only m the lower Eocene deposits 

 of this country. It is interesting to note that they are the most generalized 

 of any known Pcrissodactyla and supply a link long sought in the evolu- 

 tion of the later and more specialized forms of this order. 



- There has probably been no discovery among the ungulates since the 

 finding of the Amblypoda that has proved equal in interest and importance 

 to the discovery of this group (the Phenacodontidce). The descent of all 

 the ungulates from the A?nblypoda'has been hckl by Prof. Cope for some 

 time, but that it took place from any known genera of this order the com- 

 paratively specialized condition of the teeth of the latter distinctly forbids. 

 This moderate complexity of the teeth among Eocene mammals is a strik- 

 ing exception, especially when arsociated with such a low grade of organi- 

 zation of other parts as we find in these animals. The explanation of this 

 fact must, in my judgment, be sought for in their large size and in ths pos- 

 session of powerful canine teeth, which insure them greater immunity from 

 the attacks of fierce caniivorous contemporaries. With these means of 

 defense, they could take up their abode where food better adapted to their 

 wants was furnished. Hence we can with perfect consistency look for 

 a rapid modification of these organs, accompanied by slight change in 

 others. In order to make the connection complete between them and the 

 Phenacodonts, there should yet be found an Amblypod with bunodont 



