OF DIDELPHYS VIRGINIANA. 



71 



maximum of development in height and width, though wanting the antero-posterior expansion 

 of that of the axis. It forms a massive, upright, four-sided piUar, slightly expanded at the 

 top, and slightly grooved from top to bottom on its anterior and posterior sides, for the more 

 complete interlocking of the contiguous vertebrae. The spine is perfectly vertical, and just 

 as thick transversely as it is antero-posteriorly. The pre- and post-zygapophyses are alilie 

 in shape ; their articular faces are in parallel planes, looking in exactly opposite directions. 

 The' transverse process here first shows an anterior downward and forward prolongation, as 

 well as a posterior one, beginning the series of lateral overlapping and interlocking that has 

 been already alluded to. The /oi«'^A vertebra is like the third in all essential particulars, 

 but may be distinguished by its somewhat shorter spine, with a slight backward inclination 

 and gi-eater breadth in front than behind. The fftli again repeats the characters of the pre- 

 ceding, but the spine is much reduced, with a decided backward inclination, and narrower 

 across than in the longitudinal direction ; but it is still a pillar. The sixth is decidedly 

 modified in two respects. The spine is a simple, short, pointed process, scarcely half the 

 height of the preceding. Both zygapophyses are very prominent. The transverse process is 

 unique in size and shape ; neither that of the vertebra which precedes, nor that which follows, 

 showing any approach to its peculiarity. This consists in a remarkable expansion downward 

 and outward, producing a broad, thin, oblique plate of bone, lying within and below the line 

 of the transverse processes of the other vertebrae. This plate is overlapped by the back- 

 ward extension of the process of the preceding bone, but does not in turn overlap, or other- 

 wise restrict the motion of the succeeding vertebra. The part of the process that corres- 

 ponds to the overlapping portion of the other vertebrae, here stands out, with but slight 

 backward inclination, appearing as a sharp spine springing from the back part of the root 

 of the expanded plate, with the vertebrarterial foramen at its base. The seventh would 

 scarcely be taken for one of the cervical vertebrce, were it not betrayed by the foramen. 

 The spine is rudimentary, merely a little sharp point of bone, recognized by its position. 

 The pre-zygapophyses project far forwards. The transverse processes have no overlapping 

 by longitudinal expansion ; they are slender spicula, projecting straight out sideways, per- 

 forated at the base. The neural arch is broad, both crossways and longitudinally, and pro- 

 jects back of the centrum, partly overlapping the arch of the first dorsal ; its edge is nicked 

 in the middle for the reception of the first dorsal spine. Such are the modifications of the 

 several cervical vertebrae, resulting in the production, in this part of the spinal column, of 

 the conditions noticed in speaking of the neck as a whole, and for the purposes there stated. 



Dorsal Vertebrce. — (Fig- 

 14.) As in most marsupials, 

 the dorsals are thirteen in 

 number. The transition from 

 the last cervical to the first 

 dorsal is abrupt and well 

 marked. At the opposite ex- 

 tremity the characters of the 

 dorsal merge insensibly into 

 those of the lumbar series. 

 The general tendency of the conjoined dorsals is to assume the shape of an open S, or 

 rather italic/, the lower front part of which is completed by the upturned cervical series. 



Fig. 11. — Dorsal Vertebrse. 



