400 Bulletin American Museum of Natural History. ['N'ol. XXVII, 



Eocene Perissodactyl as from the Meniscotheroid type. The third lobe of 

 nig is lost, in correlation with the atrophy of the posterior part of m'', in both 

 groups. The manus of Meuiscotherium has no suggestion of the Chali- 

 cothere type, the carpus being of the incipiently serial type. The astragalus 

 of Meniscotherium resembles that of the Chalicotheres in the absence of a 

 cuboid contact, but this was equally true of Phenacodus and probably of the 

 remote ancestors of the Perissodactyls (p. 396). 



Conclusions. 



The numerous resemblances of the Chalicotheres to different Perisso- 

 dactyls in the skull, dentition and skeleton can hardly all be set down as 

 convergent, though many of them may be homoplastic as that term is used 

 by Osborn, i. e., independently evolved from a common ancestor. Osborn 

 and Peterson seem justified in concluding that the Chalicotheres are an 

 aberrant family of the Perissodactyls. It seems probable that the affinity 

 is nearer to the Pahieothere-Horse-Titanothere division than to the Tapir- 

 Lophiodont-Rhinoceros group and the Chalicotheres may be an offshoot 

 from some Lower Eocene forms resembling Paloplotherium. The deeply 

 cleft ungues are of course secondary — an intensification of the slightly 

 cleft ungues of Lower Eocene Perissodactyls. The resemblances to Hovialo- 

 dotherium are entirely convergent. The resemblances to Meniscotherium 

 are chiefly primitive characters inherited from Lower Eocene Perissodactyls 

 which still retained many Condylarth characters (p. 396). 



X. The Artiodactyl.a.. 



The Artiodactyla are often grouped with the Perissodactyla under the 

 term "Ungulata Vera" in contrast to the Hyracoidea, Proboscidea, Ambly- 

 poda, Toxodontia, etc., which are collectively called **Subungulata." Cope 

 applied the term "Diplarthra" to the "Ungulata Vera," apparently because 

 both Artiodactyla and Perissodactyla are distinguished first by the contact 

 of the astragalus with the cuboid as well as with the navicular, and secondly 

 by the interlocking character {i. e., the double set of facets) in the carpus. 



Aside from the characters of the feet, the Artiodactyla and Perissodactyla 

 possess very few characters in common and these are of a very inconclusive 

 nature {e. g., placenta non-deciduate, uterus bicornuate, testes descending 

 into a scrotum, os penis absent. Flower and Lydekker, 1891, p. 275). 



The points of resemblance between the two orders in the limbs are 

 sometimes very striking. Li the most generalized Artiodactyl manus 

 known, that of the Oligocene Ancodus, as described by Scott (1894), the 



