136 Annals of the South African Museum. 



to that of the Labyrinthodonts, in its elements. The Labyrinthodont 

 girdle consists of the three large dermal elements — the interclavicle 

 iind the two clavicles — with, above the clavicles, a pair of delicate 

 splint-like bones. These are sometimes called the scapulae, but are 

 doubtless the cleithra. Then there is on each side usually a single 

 ■cartilage bone. This most probably represents the scapula, but may 

 be the eoracoid. It is probable that in the Labyrinthodonts there 

 was a cartilaginous precoracoid, such as is met with in the lower 

 Batrachians of to-day. In Eryops, where the girdle is well ossified, 

 we apparently have an ossified precoracoid. In most of the 

 primitive reptiles in which the girdle is well ossified, there is 

 found an ossified precoracoid, and we have a girdle essentially 

 similar to that found in Pareiasaurus. In Procoloplion there is 

 apparently no cleithrum, but there is a distinct precoracoid. The 

 scapula, however, differs greatly from that in Pareiasaurus, and 

 resembles somewhat that of Palceohatteria. The clavicle lies along 

 the whole of its anterior border. In Palceohatteria the precoracoid 

 is unossified. In the scapula of Procolophon there is no acromion, 

 therein differing greatly from that of Pareiasaurus. In Oudenodon 

 and Lystrosaurus the shoulder girdle resembles that of Pareiasaurus 

 very considerably, though in Oudenodon, but probably not in 

 Lystrosaurus, the precoracoid forms part of the glenoid cavity. In 

 Ictidosuchus there is no acromion. In the Pelycosaurian shoulder 

 girdle the condition is essentially similar to that of Pareiasaurus, 

 though there is probably no cleithrum. 



The humerus of Pareiasaurus belongs to a type found with 

 modifications in most of the early reptilian groups. The greatest 

 resemblance is found in the Anomodont humerus. In Procolophon 

 the humerus differs very greatly, though it agrees in having an 

 entepicondylar foramen. In the Theriodonts as the humerus is 

 lengthened out in all the known examples, any great resemblances 

 in type are obscured. In the Pelycosaurians the humerus is a 

 modification of the same type that is found in Pareiasaurus. In 

 the Labyrinthodont genera, Euchirosaurus and Stereorhachis, we 

 have other modification of the same type. 



It will be seen that those characters of the shoulder girdle and 

 humerus which have placed Pareiasaurus in the same order or 

 division as the Theriodonts and their close allies, the Anomodonts, 

 are characters which are possessed, more or less, in common by 

 almost all primitive reptiles, and by forms which, as judged by the 

 cranial characters, are not very nearly related. 



The Pareiasaurian pelvis, though bearing some superficial resem- 



