274 Transactions of the South African Philosophical Society. 



lost in later life. Owing to the very imperfect condition of the 

 type specimen it is impossible to settle the point beyond doubt. In 

 the meantime it will perhaps be safest to regard it as a distinct 

 form. 



A restoration of the palate is given in fig. 19, pi. xiii. 



Prodicynodon pearstonensis, Broom. 



This interesting type, which is unfortunately known only by the 

 front half of the skull, and that in a rather imperfect condition, was 

 recently described by me (9) in Eecords of the Albany Museum. It 

 was found at Pearston, Cape Colony, in the Lower Karroo beds. In 

 . general appearance the skull resembles considerably that of Dicy- 

 nodon, but it differs, among other things, in having a few molar teeth 

 arranged in two rows. 



The premaxillary portion of the beak is much broader than is 

 usual in Dicynodon, and also flatter. On the upper surface of the 

 snout only the impressions of the bones remain, and it is a little 

 difficult to be certain about the sutures. The premaxillary certainly 

 passes backwards on the upper side of the snout a short distance, 

 but how far cannot be decided with absolute certainty. In the figure 

 I previously gave of the upper side of the specimen I figured the 

 premaxillaries as passing back between the nasals as far as the front 

 of the frontals. I am now inclined to think that the posterior half 

 of the portion of the snout in front of the frontals is really made up 

 of the nasal bones, and that what I believed to be a suture between 

 the nasal and premaxilla is the impression of a groove on the under 

 surface of the nasal produced by a blood-vessel or nerve. If this 

 latter be the correct view the nasal, instead of being a small element, 

 would be a fair-sized bone. The inter-orbital portion of the skull is 

 much narrower than is usual in Anomodonts. The pterygoids and 

 palatines, so far as preserved, are very similar to those bones in 

 .Dicynodon and Oudenodon. The lower jaw is not unlike that of 

 some of the smaller species of Oudenodon, especially 0. trigoniceps. 

 The dentary has a well-marked longitudinal ridge above the fenestra. 

 The angular fits in between the dentary and the splenial in the same 

 way as in Endothiodon. The canine tooth is round and is relatively 

 . smaller than in most Dicynodons, and it is directed well forwards, as 

 in those specimens which are believed to be females. The molar 

 teeth are few in number, arranged in two or possibly three rows, 

 and are not serrated. 



In fig. 22, pi. xiii., is shown a restoration of the side view, and in 

 fig. 21, pi. xiii., a restoration of the palate. 



