342 Mv. C. W. Andrews on the Development of 



The coracoids, which must have been separated by a thick 

 pad of cartilage, had commenced to grow forward in the 

 middle line towards the scapula, and their anterior prolonga- 

 tion already extended considerably in advance of the surfaces 

 for the scapulse. 



The next specimen (Leeds Coll. 37, E,. 2416) is a shoulder- 

 girdle wanting the clavicles and part of the right scapula, 

 but found in association with the greater portion of the 

 vertebral column, the paddles, and some of the pelvic bones, 

 all of which, allowing for their immature condition, show the 

 generic characters given above. At this age (fig. 3 B) the 

 anterior ramus of the scapulae is very incompletely developed, 

 so that the two bones were widely separated in the middle 

 line. On the upper surface of the thickened anterior edge is 

 a slight depression, which comparison with older stages shows 

 to be that occupied by the extreme outer angle of the clavicle, 

 which in this stage therefore is entirely anterior to the scapula. 

 The coracoids, wliich were of the form shown in fig. 3 B, had 

 as yet not commenced to grow forward in the middle line, and 

 their outer posterior angles were not prolonged into pro- 

 jecting processes. 



The last pectoral girdle (Leeds Coll. 36, R. 2417) (fig. 4) is 

 that of a young individual of the smaller (? female) type. The 

 scapula3 show that it is in about the same stage of develop- 

 ment as that last described, though considerably smaller. 

 This specimen is here described and figured on account of its 

 completeness and because it forms part of the nearly entire 

 t^keleton now mounted in the Gallery of Fossil Reptiles at 

 the Natural History Museum *. The cervical vertebrae, 

 abdominal ribs, &c. all agree essentially in structure with the 

 type specimen of CryptocUdus platymerus. The clavicles 

 are of the same form as the adult specimen described above, 

 but smaller and very much thinner. The structure of their 

 inner border shows that they already met in median sym- 

 physis, at least in front. The extreme end of the external 

 angle is broken away, but the ventral surface at the fracture 

 shows the beginning of the rough surface of contact with the 

 anterior border of the scapula. This latter is in much the 

 same condition as in the stage last described, but is perhaps 

 a little older. The coracoids also are very similar to those 

 last described. 



* The clavicles of this specimen have been figured by Professor Seeley 

 under the name Plesiosavrus dwohrivevsis in the * Proceedings of the 

 Pioyal Society,' vol. h. (1892) p. 133, fig. 5, where also other parts of the 

 skeleton are described and measurements given. 



