158 SHUFELDT. 



Returning to the atlas (figure 10) we find, in addition to characters 

 ,alread_y ascribed to it, that it possesses upon its anterior aspect the usual 

 two, extensive, cup-shaped facets intended for the condyles at the back 

 of the cranium. They face fonvard and inward in about equal propor- 

 tions, and are completely out of view when the vertebra is regarded from 

 directly above. The neural canal is short and cylindrical, being covered 

 by the broad neural arch above, but very slightly so protected ventrally. 

 Each apophysial, transverse, lateral expansion is twice pierced by for- 

 amina; the anterior ones are for the vertebral arteries and suboccipital 

 nerves, the other pair, entering the neural canal at its sides, are for the 

 vertebral arteries. On the posterior floor of the neural canal, the articular 

 surface for the odontoid process of the axis and the entire fore part of the 

 latter bone, save its neural spine, are continuous, thus affording very 

 considerable play betn^een the two bones. The under sides of the nearly 

 horizontal, lateral, apophysial plates of the atlas are decidedly concave, 

 with the mesial, almost entirely aborted, hypophysial tubercle standing 

 between them behind. 



This ventral surface of the atlas is bounded in front by a deep concave 

 articular border with the concavity directed backward. A similar border 

 with thickened edge forms the posterior boundary; its concavity, which 

 is not so profound, is directed forward. Between the nearest points of 

 these two concave borders in the middle line, the separating isthmus of 

 bone measures only a few millimeters. The margins of the lateral edges 

 of this vertebra are sharp and convex outward. 



Passing next to the axis, or second cervical vertebra, it is to be noted 

 that its odontoid process is but fairly well produced, being bluntly trian- 

 gular in form, considerably compressed from above, downward, and to- 

 gether with the rest of the articular surface on that aspect of the bone, 

 projecting entirely beyond the neural spine above it. This is by no means 

 always the case in mammals, for among certain Felidfe and Canidse it 

 may be observed that the anterior projection of the neural spine in this 

 vertebra overhangs the odontoid process. There are no prezygapophyses, 

 while the postz3'ga.pophyses are much aborted, the elliptical articular 

 facets, which they support in other vertebrs being represented, one on 

 either side, by similar surfaces situated beneath the neural lamina. They 

 project beyond the small and vertically much compressed centrum, which 

 presents, posteriorly, a rather large facet for the third vertebra. (Plate 

 111, figure 11.) The transverse processes are moderate in size, triangular 

 in outline^ and much compressed in the vertical direction. N"o foramen 

 for the passage of the vertebral artery, on either side, is to be found in 

 the axis, while the cylindrical neural canal is much smaller than it is in 

 the atlas. In fact it has about the same caliber and form throughout 

 the cervical series as it has in the axis now being considered. 



The remaining five cervicals are all very much alike, and we find each 



