New South African Fossil Amphibians and Ecptiles. 275 



distinct canal for itself, only joining the larger foramen at its 

 exit. 



The occipital condyle is in an extremely interesting condition, 

 being intermediate in type between that of the Anomodonts and that 

 of the typical Cynodonts. In Dicynodon and allied forms the 

 condyle is single and made up of a median basioccipital portion and 

 two lateral exoccipital parts. Here the basioccipital part has become 

 very much reduced, and the exoccipital parts are approximated. 

 Though there is thus produced a double condyle by the reduction 

 and retreat of the basioccipital, the two condyles are so near that 

 they probably act as a single condyle. In the Cynodonts proper the 

 exoccipital portions become enlarged and separated by a very deep 

 cleft. 



The lower jaw is in a very interesting condition. Hitherto one of 

 the most striking distinguishing features of the Cynodonts has been 

 that the dentary extended backwards nearly to the articulation, 

 whereas in the Therocephalians the posterior third of the jaw was 

 formed entirely by the articular, the angular, and the surangular. 

 In Bauria we have a condition exactly intermediate between the two 

 types. The dentary is long and not very powerful, and has most 

 probably, like the maxilla, 10 molars. Those which are displayed are 

 like the upper teeth simple and unserrated. The canine is fairly 

 large and perfectly round, and there are four incisors slightly 

 flattened transversely. The complete dental formula of Bauria 

 would thus be : i ' f- c }- m -^} X 2 = 60. The coronoid process is 

 thicker and shorter than in most Therocephalians, and less expanded 

 than in typical Cynodonts. The Cynodonts hitherto known have all 

 differed from the Therocephalians in having the dentary nearly 

 reaching to the articulation. In Bauria the dentary only forms the 

 anterior three-quarters of the jaw. The posterior part of the jaw is 

 much more Therocephalian in type than Cynodont. The surangular 

 forms a rounded convex upper border to the hind part, and the 

 angular is of fair size. The articular is also Therocephalian in type. 



The atlas vertebra, so far as can be seen, is fairly similar to that 

 in the ordinary Cynodonts and has a large inferior piece. It is 

 apparently the arch mainly that gives articulation to the occipital 

 condyles. The broad inferior piece, though not anchylosed with the 

 arch, must be firmly attached to it by ligaments. There is a proatlas. 



The axis has a flat broad odontoid process which lies above the 

 inferior piece of the atlas, and owing to its shape it must have been 

 almost impossible for any rotatory movement to have been made 

 round the odontoid. In this feature Bauria is distinctly more 



