38 THE CAT. [chap. hi. 



fuhercular, because ttey articulate Avith the tubercles of tbe ribs. 

 The tubercular surface {d} is a smooth, oval surface, slightly pro- 

 longed from before back-wards, and placed one beneath the distal 

 end of each transverse process and giving attachment to the tubercle 

 of the fifth rib. 



The capitular surfaces (p, p) are two in number on each side. 

 One is placed at the junction of the pedicle with the centrum in 

 front ; it is smooth, and looks forwards and outwards. _ The other 

 is placed just beneath the posterior notch of the pedicle ; it is 

 smooth, and looks so almost directly backwards and so very Httle 

 outwards as to form part (the outer and upper angle) of the posterior 

 surface of the centrum. The anterior capitular surface concurs with 

 the posterior capitular surface of the fourth vertebra to form with it 

 an articular cavity for the head of the fifth rib. Similarly its 

 posterior capitular surface concurs with the anterior capitular surface 

 of the sixth vertebra to form an articular cavity for the head of the 

 sixth rib. 



The ring formed by the neural arch and centrum is oval, trans- 

 versely extended, and somewhat flattened below. 



The eleventh, twelfth and thirteenth dorsal vertebra) have each but 

 a single capitular surface on each side — namely, an anterior one. 

 The first dorsal has an anterior capitular surface large enough to 

 receive the whole head of the first rib. The eleventh, twelfth and 

 thirteenth vertebra) have no tubercular surface. 



The first two dorsal vertebra) have the front surface of the centrum 

 strongly convex and looking much downwards, and its hinder surface 

 concave and looking much upwards. The tubercular surface also is 

 strongly concave. 



As we proceed from before backwards, through the series of dorsal 

 vertebra) to the tenth, the transverse processes come to extend less 

 outwards and to be more expanded from before backwards at their 

 distal ends ; the postzygapophyses become situated further backwards, 

 and the neural spines (counting, at least, from the seventh,) also 

 become shorter. 



The tenth dorsal vertebra has its transverse process very much 

 extended from before backwards (Fig, 17,^°). Its posterior end projects 

 backwards more than in any preceding vertebra, reaching to, or 

 even beyond, the anterior margin of the pedicle of the eleventh 

 vertebra. The postzygapophyses look as much outwards as down- 

 wards, or even mainly outwards. 



In the eleventh dorsal vertebra the neural spine projects more or 

 less forwards (Fig. 17, ^^ *'), instead of backwards, abutting against 

 that of the tenth vertebra, which it may, or may not, slightly 

 exceed in length. It has no transverse process, but there arc 

 two conspicuous processes on each side, which evidently answer 

 to the processes (one at each end) which terminate the transverse 

 process of the tenth vertebra, but which, in the eleventh vertebra, 

 are quite separated the one from the other. 



The anterior process (Fig. 17, ^^ '"), which projects forwards. 



