TEMPOllAL ARCHES OF TUE REPTILIA. 21 



the fossa in the Mammal-like reptiles which really corresponds to 

 the lower fossa in Sjyhenodon. Though this eai'ly view of mine 

 has been opposed by Andrews, Willistoi], Watson, and others, 

 I have never seen reason to alter it, as apart from the superficial 

 resemblance of the upperside of the skull to that of a Therapsid, 

 the fundamental structure of the Plesiosaur is so diffei-ent from 

 that of the Mamma.l-like i-eptiles that I never could agree that 

 the two groups could be at all nearly related. It is with some 

 satisfaction, therefore, that I find Versluys supporting my view, 

 that the fossa in the Plesiosaur has had quite a different origin 

 from that in the Mammal-likc' reptiles. Yersluys makes it his 

 '' Synapsider Typus II." 



Elsewhere * I am dealing at some length with the question of 

 the affinities of the Plesiosaui's, and the conclusion to which I 

 come is that they are a specialised offshoot from land-reptiles on 

 the line of descent which is leading to the Diapsids, but which 

 have not yet developed a lower temporal fossa. 



In near association with the Plesiosaurs must be placed the 

 Placodonts. They are a different specialisation sprung from 

 very similar and closely allied ancestors. There is no essential 

 difference in the temporal regions of the two groups. 



The Cheionians are by far the most aberrant and puzzling 

 group among the reptiles. Unfortunately, we know nothing of 

 the ancestry. The earliest-known fossil forms are already typical 

 Cheionians, and help us very little. Among living forms we have 

 great variations in the temporal regions. Some have the region 

 well roofed. Some have practically no roof at all, like the 

 American Terrapene. Some have a small roof very deeply 

 encroached on from below, like the Australian Water-Tortoise 

 Elseya. 



For many years I have adhered to the view, which was also 

 held by Baur, that there is some close relationship between the 

 Chelonia and the Plesiosaurs. Unfortunately, each group is so 

 extremely specialised that hardly any apparent resemblance 

 i^emains. It is like comparing a Whale and a Bat. The 

 Ohelonian and the Plesiosaur has each a remarkable type of 

 shovilder-girdle, nowhere else to be found, and each is merely a 

 modification of the same common type. It is difficult to believe 

 that this remarkable type could have been twice independently 

 evolved. The pelvis, the tarsus, and many points in the skull- 

 structure confirm the affinity. 



If we assume that the Chelonian is a second remarkable 

 specialisation from a land-type, such as gave rise to the Plesiosaurs 

 by another line, we must conclude that the ancestor of the 

 Chelonian had, like the Plesiosaur, a single temporal fossa, and 

 that the condition now met with in Cheionians has resulted from 

 the loss of the posterior ai-ch. A skull like that of Trionyx with 

 its greatly elongated supraoccipital and the squamosal perched on 



* Williston Memorial Volume. 



