174 Transactions of the South African Philosophical Society. 
In Cistecephalus the anterior part of the palate is unknown, but 
as the pterygoids, the vomer, and at least the posterior part of the 
palatines are constructed almost exactly as in Ptychosiagum, and as 
the rest of the skull, so far as known, is thoroughly Dicynodont in 
type, it is very unlikely that the anterior part of the palate differs 
appreciably from that in Ptychosiagum. 
In Endothiodon, which forms the type of the other family of the 
Anomodontia, the palate, though closely agreeing with the 
Dicynodont type, presents a number of peculiarities. The main 
differences are due to the greater development of the maxillary, 
which bears a row of teeth, and the corresponding less development 
of the premaxillary. The premaxillary is, as in Dicynodon, eden- 
tulous. Its anterior palatal portion is moderately developed in 
Cryptocynodon sumus, but is relatively smaller in Hndothiodon 
(Hsoterodon) wniseries. The median ridge of the premaxillary is 
much more distinct than in Dicynodon, and though in none of the 
known specimens is there conclusive evidence that it is distinct from 
the premaxillary, there is certainly a strong suggestion of its being a 
prevomer. The vomer is very distinctly seen in the three known 
specimens immediately behind the median ridge—exactly as in 
Dicynodont type. In Seeley’s figure of Cryptocynodon* the 
division of the vomer from the anterior element is not very dis- 
tinctly brought out, and from his description it is evident that he 
did not recognise the two elements as he speaks of the whole ridge, 
vomerine and premaxillary or prevomerine, as ‘‘ presumably formed 
by the vomer.’”’ In the original specimen the anterior end of the 
true vomer is sharply defined from the premaxillary ridge, almost 
exactly as is the case in Hndothiodon bathystoma and EH. wniseries. 
The palatines are formed on a very similar plan to that of 
Dicynodon. They articulate with the vomerine wings or lateral 
ridges, and form the posterior and upper walls of the nasal passages 
behind. Anteriorly they form an imperfect secondary palate, 
though a larger one than in Dicynodon. The hinder part of the 
palate is unknown, but presumably similar to that in Dicynodon. 
The Anomodont palate, it will be observed, does not show a 
marked affinity with that in my Reptilian order, except the 
Theriodontia and the Chelonia; while as regards the vomer and 
palatines there is a distinct affinity with the Mammalian type. In 
almost all the known Reptilian orders the palate is formed essentially 
on the Rhynchocephalian type. In the Chelonia and the Crocodilia 
there are specialisations which greatly mask the ancestral type, and 
this is likewise the case with the Theriodontia and the sub-order 
Anomodontia. 
* H. G. Seeley, ‘‘ On the Therosuchia,” Phil. Trans., vol. 185, B., 1894. 
