No. 446.] RECLASSIFICATION OF THE REPTILIA. 99 



Proposed Reclassification. 



At the Washington meeting, 1902, of the American Associa- 

 tion, I presented a joint paper with Dr. J. H. McGregor on the 

 diphyletic arrangement of the reptiles, based on comparison of a 

 large number of characters. I have since made a more searching 

 study of the same problem, designating these two groups as the 

 subclasses Synapsida, or primitively single-arched reptiles, and 

 DiAPSiDA, or primitively two-arched reptiles, and have grouped 

 all the most primitive forms of Diapsida in the superorder 

 DiAPTOSAURiA, a group equivalent in taxonomic rank to the 

 Squamata or the Dinosauria. I now propose to briefly describe 

 the reptiles which fall within these two groups. 



SUBCLASS SYNAPSIDA Osborn. 



The chief distinction of the single arched reptiles is that there 

 is either no opening at all in the temporal region (Cotylosauria), 

 or a single large supratemporal opening (Anomodontia, Plesio- 

 sauria, Testudinata) as in the upper view of the skull of a 

 plesiosaur a turtle or a mammal. This supratemporal fossa is 

 large, because the cranium or brain case is long while the facial 

 portion of the skull is relatively short, these proportions being 

 directly reversed in the Diapsida. The temporal arch consists 

 primitively of two arches combined. The squamosal is always a 

 large element. The quadrate is correspondingly more or less 

 reduced; it is never movable, and is functionally supported by 

 the squamosal. In the shoulder girdle the coracoid and procora- 

 coid are separate, or united by suture. The phalangeal formula 

 is primitively 2. 3. 3- 3- 3, H^e that of mammals. 



I. Order Cotvlosm kia Lope 

 Pareiasauria Sccle\-. 

 These are the most primitive of ..ptiles, retaining many 

 Stegocephalian (amphibian) characters, and a sohd cranial roo 

 with temporal openings rudimentary or not developed at all. 



