An almost Perfect Skeleton of Pareiasaurus serridens. 131 



The structure of the glenoid cavity is remarkable. It consists of 

 an irregular concavity formed by the scapula and coracoid, and 

 whereas in most reptiles both recent and extinct the articulation is 

 a fairly regular hollow, here really two cavities are formed by a 

 vertical dividing ridge. The larger or main concavity includes the 

 whole of the coracoid part of the articulation, and the posterior 

 two-thirds of the scapular, while the anterior one-third of the 

 scapular articulation forms the lesser concavity. The head of the 

 humerus has its articular surface divided by a shallow groove, into 

 which the ridge of the glenoid cavity fits. As a result of the 

 complexity of this joint, the movements of the humerus must have 

 been limited to those in a vertical direction and to rotatory move- 

 ments. There can have been little or no antero-posterior movement. 



The sutures between the precoracoid, the coracoid, and the scapula 

 are almost obliterated, though here and there they can be distinctly 

 made out. The precoracoid almost certainly forms no part of the 

 glenoid cavity. 



The clavicles, interclavicle, and the cleithra are very similar to 

 those in P. baini. The clavicle fits into an irregular hollow groove 

 in the front of the lateral process of the interclavicle. 



The Humerus. 



The right humerus is in almost perfect condition, only a very 

 small fragment of the outer edge of the lower part being lost. It 

 agrees pretty closely with the humerus of P. baini, though at first 

 sight this would not appear to be the case on comparing the specimen 

 with either Seeley's figure, or with the original specimen in the 

 British Museum. Seeley, in a note (II. p. 360), says that the 

 continuity between the upper and lower ends of the humerus was 

 lost while the specimen was being cleared of matrix, and that the 

 two ends were placed together, to the best of his belief, as they were 

 in the rock, but that it is impossible to be certain that the original 

 contact was as it is drawn. When the humerus of P. baini, as fixed 

 together, is compared with this perfect humerus, it at once becomes 

 evident that the top end has been rotated about 70° out of its true 

 position ; and that whereas in the P. bai)ii skeleton, as set up and 

 figured, the two ends of the humerus are nearly in the same plane, 

 in reality they ought to be nearly at right angles. The figures in 

 Plate XV. show the humerus in various positions. It will be seen that 

 the bone has a distinct entepicondylar foramen. The lower end of 



