THE VERTEBRAL COLUMN. 



53 



I 



sacrals. The transverse processes of the eighth and ninth sacrals are in contact, though not 

 coossified at their extremities, and inclose a large and elongate foramen. The transverse 

 process of the seventh sacral is expanded at its extremity, but it is widely separated from the 

 transverse processes of the preceding 3 

 and succeeding sacrals. 



Seen from the side the sacrum is 

 strongly arched upward with the neural 

 spines of the true sacrals coossified and 

 forming a strong bony plate. Begin- 

 ning with the second sacral and con- 

 tinuing to the ninth the extremities of 

 the diapophyses are expanded and pre- 

 sent rugose surfaces for contact with 

 the ilium, as shown in figs. 54 and 55. 



Viewed from above the neural 

 spines appear much compressed. The 

 anterior diapophyses are shorter than 

 the parapophyses and broader and 

 thinner than the diapophyses of the 

 posterior sacrals. As shown in fig. 55, 

 all the diapophyses are in contact with 

 and give support to the ilium, save 

 those of the first and last sacrals. 



There is an enlargement of the 

 spinal cord in the anterior portion of 

 the sacrum, as shown in fig. 56, though 

 it is not so pronounced as in Stegosau- 

 rus. Between each pair of sacral ver- 

 tebrae a pair of nerve branches are 

 given off from the spinal cord, and there 

 was a marked constriction in the spinal 

 cord just at each of the points where 

 these branches have their origin. 



THE CAUDAL VERTEBRAE. 



Unfortunately only a few scat- 

 tered caudal vertebra? of Triceratops 

 are known, and it is quite impossible 

 to determine the exact number of ver- 

 tebrae in the tail. 



The first caudal, as shown in fig. 

 57, has the centrum broader than long. 

 The neural spine is strong, expanded 

 at the apex, and is directed upward and 

 backward at an angle of about 45°. 

 The transverse processes are rather 

 short and are directed outward at right angles to the centrum. The centrum is flat beneath 

 and is broader than deep. The neural arch is low and is not so deeply incised for the zyga- 

 pophyses as in the presacrals, more especially the dorsals. 



&2 



o- H . 



