328 PROF. H. G. SEELEY ON THE [May I9OO, 



The external surface of each centrum is concave from front to 

 hack and rounded from above downward, with moderately elevated 

 rounded articular margins, which are rather wider in the early dorsal 

 than in the later dorsal vertebra?, but much narrower than the neural 

 arches. 



The Ribs. (PI. XXI, r.) 



As the ribs are preserved they extend in a semiovate contour, 

 which was 12 inches long and about 8 inches wide below the 

 middle of the back. As in many vertebrates, the ribs indicate 

 depth in the region of the lungs, with increasing width in the lower 

 part of the back. There is no indication of sternal or of abdominal 

 ribs. 



The anterior dorsal ribs are very imperfectly displayed at their 

 proximal ends ; but the distal ends of the first six or eight are so 

 curved as to cross some of the later ribs transversely resting upon 

 their undersides. The under surfaces of the ribs are rounded, with 

 the anterior margin compressed and widened and defined by an 

 inferior groove, like that seen in Pareiasaurus. There is no 

 evidence whether the compression is also seen on the posterior 

 margin. The ribs are compressed from above downward towards 

 the free end. Near the proximal articulation each is also compressed 

 from front to back. The rib terminates proximally in an expanded 

 disc, which articulates low down on the anterior half of the side of 

 the centrum. No tubercular attachment of the rib to the neural 

 arch is seen. There is no evidence of articulation between the bodies 

 of the vertebrae. In the genus Herpetoclwirus, there is no trace of 

 a tubercle in any of the ribs preserved. The articulation appears to 

 be not unlike that found in Microgomphodon, 1 but Eurycarpus shows 

 no trace of the uncinate process to the rib which Gomphodontia and 

 €ynodontia share with some Labyrinthodonts such as Eucheirosaurus. 



The Shoulder-girdle. 



The bone in Mr. Bain's slab which I had doubtfully regarded as 

 a scapula (or interclavicle) is proved by my new specimen to be 

 so. Its articular surface for the humerus is seen a little in 

 advance of the head of that bone. The scapula (PI. XXI, S) lies 

 laterally between the neural arches of the vertebrae and the humerus. 

 It appears to extend over the length of six vertebrae and to be rather 

 longer than the humerus. The preservation is bad, owing to the 

 dried condition of the animal at the time of f ossilization. The length 

 of the scapula exceeds 5 inches. At the humeral end it is 2 inches 

 wide, and towards the anterior border the bone carries a moderate 

 ridge of the usual type, such as might have supported a clavicle. 

 Both anterior and posterior borders of the scapula are concave, and 

 its least width did not exceed 1 inch. Its inner surface was concave, 

 adapted to the convexity of the ribs. 



There is no indication of precoracoid or coracoid bones A small 



1 Phil. Trans. Roy. Soc. vol. clxxxvi (1895) B, pi. i. 



