no. 2086. SOME MAMMALS OF THE PLEISTOCENE— HAY. 541 



small loop. In the premolars this valley has its axis directed toward 

 the front of the median style of the next tooth in front. In the 

 molars it is directed still farther forward, to the anterior outer style 

 of the next tooth in front or even to the rear of the second tooth in 

 front. In all cases it lacks much of reaching to the center of the 

 tooth. The fold of enamel which in the most of the horses extends 

 forward between the postero-internal column (hypocone) and the 

 postero-median column (metaconule) is small in the premolars, ex- 

 tremely small in the front two molars, and missing in the last molar. 

 The small size of this fold constitutes an approach to Protohippus. 



The walls of enamel surrounding the lakes of each tooth are little 

 complicated; least so in the molars. In the premolars there is a 

 small reentering fold in both the front and the rear sides of both 

 the lakes, with perhaps some additional undulations of the enamel. 

 In the molars there is no infold in the front of the anterior lake 

 and only a very small one in the hinder wall of the posterior lake. 

 The hinder border of the anterior lake has a small infold opposite 

 the head of the anterior internal valley. The front border of the 

 posterior lake is slightly more complicated, there being an infold 

 and two or three undulations in the enamel. In the first upper 

 incisor the median longitudinal ridge is obsolete, so that the front 

 of the tooth from side to side is concave. In the third incisor the 

 ridges all appear to be obsolete. In the lower incisors the median 

 groove is broad and shallow. 



The writer proposes to describe with as much exactness as he can 

 and as briefly as is consistent with this exactness such of the bones 

 of the trunk as have been preserved. 



In making these measurements the writer has consulted especially 

 the paper of Nehring 1 and that of Tscherski 2 and has endeavored 

 to make the measurements conform as far as possible with those of 

 those authors; but the number of measurements is here reduced. It 

 is thought well to make the following explanations. In measuring 

 the bodies of the vertebrae the distance is taken between the point 

 on the anterior articular surface farthest in front and the point 

 farthest in front on their hinder articular surface, this being usually 

 nearly from center to center. The other measurements of the verte- 

 brae do not need explanation. The length of the scapula is taken 

 from the front of the glenoid cavity, along the spine, to the line 

 which separates the endosteal bone from that of the main part of 

 the scapula. 



In the case of the humerus the total length is the distance from the 

 upper end of the ridge between the two bicipital grooves to the distal 

 end. The length of the radius is measured on the outer border of 



1 Landwirthsch. .Tahrb., vol. 13, 1884, p. 81. 



3 Mem. St. Petersb. Acad. Sci., ser. 7, vol. 40, pp. 257-383. 



