542 



PROCEEDINGS OF THE NATIONAL MUSEUM. 



vol. 48. 



the bone from the middle of the outer edge of the articular surface 

 for the humerus to the surface for union with the cuneiform bone, 

 really the lower end of the ulna. This is one of Xehring's measure- 

 ments, as understood by the present writer. 



The median metacarpal is measured on the outer border from one 

 articular surface to the other. 



The first phalange is measured on the outer border, beginning 

 above at the middle of the outer border of the upper articular con- 

 cavity. The length of the second phalanx is obtained in the same 

 way. In these cases we obtain the efficient length, but not the total 

 length. The middle of the length of the femur is midway between 

 the upper surface of the head of the bone and the internal condyle. 

 The lengths of the median metatarsal and of the phalanges are ob- 

 tained as in the fore foot. 



There are present 12 presacral vertebrae; but most of these are 

 more or less injured, and only a part of the desired measurements can 

 be obtained. For comparison, the corresponding vertebrae of E. 

 hemionus and of the Arabian horse are furnished. The atlas has lost a 

 part of the right wing. The following are some of the dimensions 

 of the bone, a part of them estimated, but they may be relied on. 



Measurements of the atlas in millimeters. 



E.fran- 



E. Tiemi- 



met. 



onus. 



38 



41 



110 



110 



110 



127 



73 



75 



67 



77 



Arabian. 



Length of the upper arch at the midline 



Width in front, across the anterior foramina 



Width in the rear, across the hinder foramina 



Distance across the articular surface for the occipital condyles 

 Width across the articular surfaces for the axis 



52 



136 



150 



90 



92 



The atlas of the fossil species differs from that of the Arabian 

 horse in having the same width in front as in the rear, instead of 

 widening backward ; but variations may here be looked for. In the 

 Arabian the lateral wings droop considerably, so that if a line be 

 drawn from the border of one to that of the other, passing just 

 behind the tuberosity on the lower face of the lower arch, this line 

 will fall much below the border of the articular surface, while in 

 the fossil species the line will touch the border of the articular sur- 

 faces at the midline. In E. hemionus the line falls a very little 

 below the articular surface. In the fossil the distance across the 

 anterior articular surfaces is greater, as compared with the length 

 along the upper arch, than in the Arabian horse. 



The axis is represented by only a fragment which clings in the 

 matrix in the rear of the atlas. The bone was undoubtedly present 

 in its natural position when the skeleton was found. 



