314 Mr. D. M. S. Watson on 



second.iry plates of tlie palatines aud maxilUe, with which it 

 is not in contact, until on the anterior end of tlie specimen 

 it is seen to have acquired a U-shaped section, being 

 channelled on the upper surface. 



This bone is very difficult of interpretation. It is a quite 

 constant feature of all Diademodon skulls, in fact of the 

 skull of all higher Cynodonts. 



It appears to be pretty ceilainly the bone which in the 

 type skull of Gomphognuthus pohjpliayus projects between 

 the palatal processes of the premaxiilai on to the palate. 



Broom has suggested that this bone is here paired, a 

 suggestion which receives some support from its apj)earance, 

 but in the light of the other specimens described appears to 

 be fallacious. If it is not jiaired it cannot be a prevomer as 

 Broom has suggested. It is also certainly not tlie anterior 

 end of the vomer, because in the specimen under consideration 

 it is undisturbed, whilst the vomer is cruslied out of place. 

 It can apparently only be an ossification developed in the 

 narial septum. 



In these sections the teeth are beautifully shown : they 

 are inserted in a completely mammalian fashion by deep 

 V-shaped roots, inclosing a small pulp-cavity which is closed 

 below. There appeal s to be absolutely no trace of sueces- 

 sioiial teeth. 



The teeth in the lower jaw are always narrower than those 

 in the upper jaw. Anteriorly one of them is shown which, 

 like the anterior teeth described by Dr. Broom in Diade- 

 modon mastacns, has a high, bluntly pointed, and enamelled 

 crown. 



Cynognathus sp. 



K. 2572 of the Biitish ^Museum collection is the undis- 

 torted back of the skull of a Cynognathus which has already 

 been described by Seeley. Secley suggested that it might 

 belong to Cynognathus berryi ; but there appears to be no 

 reason why it should not really belong to C. crateronotus, 

 all differences from the type skull of the latter being 

 exjjlicable as due to crushing. 



The specimen is broken off anteriorly just behind the 

 columellae cranii, but shows very well all the bones of the 

 brain-case behind that plane. The structure agrees exactlv 

 in general features with that which 1 have already described 

 in Diademodon. 



The basi-, ex-, and supraoccipitals arc all fused and be- 

 tween them surround the small, almost circular, foramen 

 magnum. A well-marked suture starts in the middle of the 



I 



