BETWEEN TYPICAL REPTILES AND OTHER ANIMALS. 193 



The vertebral column of Serpents resembles that of Iguana in 

 the form of articulation of the neural arch by addition of a zygo- 

 sphene ; but the Iguana has the neural spine inclined backward 

 and thickened posteriorly, which is not the case in Serpents ; also 

 in transverse section the part of the arch at the base of the neural 

 spine which is convex in Serpents, in Lizards is concave. The 

 articulation for the rib is more elongated vertically in Serpents 

 than is usual in Lizards. 



§ 6. The Urodelan Characters of Lizards. 



As in Ehynchocephalia and Ophidia the palatine abuts against 

 the maxillary and carries a second row of teeth, the pterygoid 

 and palatine are more expanded than in Lizards (in this rather 

 recalling Chamceleon) , and, with the parasphenoid between the 

 pterygoids, in the Hell-bender, make a closed palate. 



The nasal sacs are double, and in the Hell-bender appear to 

 be surrounded by a similar set of bones to those which mar- 

 gin the anterior nares in Monitor. 



As in Monitor, the Hell-bender does not prolong the maxillary 

 arch backward, and the orbit has no margin of bones behind ; the 

 animal is unlike 3Ionitor in having all the median roof-bones of 

 the skull double. 



Supraoccipital and basioccipital in the Hell-bender would seem 

 not to exist, though the posterior part of the basitemporal looks 

 as though it might well become a basioccipital bone like that 

 of mammals. 



The atlas of the Hell-bender has a strong resemblance to the 

 axis of mammals and Lizards, what would be called the odontoid 

 process fitting into the vacuity where the basioccipital is usually 

 found, while the flattened lateral facets of the centrum fit on to 

 the exoccipital bones. And this would raise the question whether 

 if a vertebra with the characters of an ordinary atlas came to be 

 developed between this vertebra and the skull, its centrum would 

 not go to form a basioccipital bone. The outline of a vertebra in 

 Hell-bender is very similar to that in Skink, difi'ering in more 

 perfect suppression of the neural spine, and in the development 

 of transverse processes from the centrum, which in many Sala- 

 manders are double-headed. These processes are long in the 

 Hell-bender ; in Triton they are short, and give attachment to 

 double-headed ribs, which have in the middle of their hinder 

 margin an epipleural element, also seen in the eai-lier ribs of the 



