Structure and Affinities of the Endothiodont Reptiles. 269 



the ilium is supported by the four sacral vertebrae. There are no 

 caudal vertebrae preserved. 



The shoulder girdle is very imperfectly preserved, the only parts 

 remaining being the lower part of the right scapula with part of the 

 precoracoid, and a portion of the top of one of the scapula, probably 

 also the right. So far as preserved the scapula agrees pretty closely 

 with that of Dicynodon and Oudenodon. There is a well-developed 

 expansion in front which articulates with the precoracoid, and there 

 is evidence of there having been a distinct acromion, but whether 

 this was as well developed as in Dicynodon the evidence does not 

 satisfactorily show. On the inner side there is at the lower end, as 

 in the Dicynodont scapula, a deep groove leading down to the 

 precoracoid foramen. The glenoid surface of the scapula is nearly 

 at right angles to the axis of the bone. The precoracoid is well 

 developed but appears to have been relatively smaller than in 

 Oudenodon. The precoracoid foramen on the outer surface opens 

 about 10 mm. below the suture with the scapula. In this it differs 

 from the Dicynodont condition and agrees with the condition in the 

 Therocephalians {e.g., Ictidosuchus). It is interesting to note that 

 the anterior border of the precoracoid does not extend so far in the 

 direction of the precoraco-scapular suture as does the scapula. A 

 similar condition was long ago observed in the small Dicynodont 

 shoulder girdle figured by Owen in his Catalogue. 



The humerus, which is represented by the upper half of the right 

 bone, is strikingly Dicynodont. The head of the bone has the same 

 kind of broad articulation as is found in the Monotremes, Dicynodon, 

 and Oudenodon. When viewed from above the proximal end of the 

 bone has a somewhat sigmoid curve owing to the deltoid ' ridge 

 passing forward and .outward and the inner angle of the bone being 

 directed somewhat backwards from the transverse head. The deltoid 

 ridge is of large size and bears similar relations to the head as does 

 the ridge in Dicynodon and Oudenodon. At its lower angle the ridge 

 does not form a downward process as in some species of Oudenodon 

 {e.g., Oudenodon magnus [= Platypodosaurus robustus]). On the 

 posterior surface of the humerus a well-developed ridge passes down 

 the bone from the articular surface. On the inner edge of the bone, 

 near the level of the lower margin of the deltoid ridge, is a fairly well- 

 developed tricipital ridge. It is thus somewhat lower in position than 

 the corresponding ridge in " Platypodosaurus," and it is less pro- 

 minent. This ridge has been supposed to. be absent in Dicynodon, 

 but though less developed than in Oudenodon, it is quite distinct in 

 at least some species of Dicynodon. Close to the lower border of the 

 tricipital ridge is situated the beginning of the entepicondylar foramen. 



