CERVICAL VERTEBRAE OF CHELONIANS 217 



of both Testudo eiihippium (fig. 1) and of Eretmochelys imbricata 

 there are eight bones; but whether this eighth one is followed by 

 a true dorsal vertebra I am unable to say, for the reason that I 

 know nothing of the prepai-ation of this material, nor have I 

 seen, in either case, the rest of the vertebral column, 



Still omitting the axis vertebra, the seventh of either of these 

 series is peculiarly formed, its neural spine, situated on the fore- 

 part of the bone, being enormously developed and elevated far 

 above the centrum, being broad and spreading, with its superior 

 aspect converted into a great, cup-like concavity, the postero- 

 lateral projections of which are the postzygapophyses. This 

 peculiar formation is most highly developed in the great tortoise 

 of the Galapagos, though it is present, too, in the hawksbill, 

 where it is so extended as to include most of the modified neural 

 spine of the vertebra next beyond it. In the latter, all the 

 cervicals are considerably shorter than in the former species, 

 where they are much elongated and present entirely different 

 characters. 



In neither of these gigantic species do w^e find the infero- 

 median part of the atlas vertebra coossified with the rest of that 

 bone, and especially is this evident in the hawksbill. 



Both species present a foraminal perforation at the middle of 

 the base of the 'cup,' which, in many vertebrates, admits the 

 anterior end of the odontoid process of the second or axis vertebra. 



Throughout the literature of the osteology of the chelonians, 

 this second vertebra has been simply nominated the 'odontoid 

 bone,' and no writer heretofore has considered it in the light of 

 one of the cervical vertebrae. As a matter of fact, it is the 

 centrum of the axis, with the neural arch and the apophysis 

 usually found on this bone, aborted. The true odontoid process 

 is still to be found well developed at its usual site in some existing 

 turtles, and particularly in those of the genus Amyda (figs. 3, 

 4, and 5). Indeed, in A. cartilaginea, the axis or second cervical 

 vertebra not only has a well-pronounced odontoid process, but 

 the anterior face of the centrum presents three facets for articu- 

 lation with the atlas — an inferomedian one and an upper one 

 upon either side above it. Inferiorly, the centrum is produced 



