OSTEOLOGY OF AMERICAN PERMIAN VERTEBRATES 393 



squamate reptiles, their presence in Araeoscelis is only what would 

 be expected in a primitive lizard. Nor is the presence of teeth 

 on the transverse bone (if it be distinct) a character to which any 

 weight can be attached. Though no lizards have such teeth we 

 may be sure that their ancestors had them. 



In the structure of the skeleton of Araeoscelis there is very 

 little that would not confidently be expected in the primitive lizard. 

 The ribs of the neck and lumbar region are single-headed in the 

 strictest sense, not holocephalous, as in the cotylosaurs, and they 

 articulate exclusively with the centrum. In the dorsal region they 

 are dichocephalous, but the head and tubercle are close together, 

 separated only by a short emargination. The head is articulated, 

 unlike the condition in contemporary reptiles, high up on the side 

 of the centrum near the front end, the tubercle to a short 

 diapophysis on a level with and just back of the anterior zyga- 

 pophyses. The fusion of these two articulations, or, what is more 

 probable, a loss of the diapophysis, which actually occurs in the 

 lumbar region, would make the rib attachment typically lacertilian. 

 I am quite sure that the primitive lizards had dichocephalous or 

 holocephalous ribs, that is, capitular and tubercular articulations. 

 Araeoscelis offers a rational explanation of the way the very 

 peculiar rib attachment of the Squamata arose. 



The pectoral girdle shows certain peculiarities not known 

 among the Squamata, especially the separate articular facets in 

 front of the glenoid facet, and the possible absence of the supra- 

 coracoid foramen. The distinction of the coracoid is not quite 

 certain, though the dotted line of the figure represents what seems 

 to be a real sutural division. I have urged that the posterior 

 coracoid bone has been lost in all modern reptiles; that such a 

 bone was present in the ancestral lizard is quite certain. 



In the anterior extremity the presence of a distinct ectepi- 

 condylar foramen, quite as in the lizards, is suggestive, at least, 

 of genetic relationships; and there is also a suggestion that the 

 entepicondylar foramen was obsolescent. The structure of the 

 feet is unHke that of the contemporary reptiles in the elongated 

 calcaneum and reduced astragalus. The pelvis is plate-Hke, with- 

 out a decided pubo-ischiadic vacuity; but no other kind of a pelvis 

 could be expected in the primitive lizards. 



