On the Classification of the Reptilia. 



263 



groups which he called the Synapsida (with one lateral temporal fossa and 

 one bar) and the Diapsida (with two lateral temporal fossa? and two bars). 

 This paper marks an epoch in classification. In the Synapsida were placed 

 the Cotylosauria, Anomodontia, Testudinata, and Sauropterygia ; in the 

 Diapsida were placed the Protorosauria, Pelycosauria, Bhynchocephalia, 

 Procolophonia, Proganosauria, Choristodera (Simcedosauria), Ehynohosauria, 

 Phytosauria, Ichthyosauria, Crocodilia, Dinosauria, Squamata, and Ptero- 

 sauria. But, owing to a great extent to the incomplete state of knowledge 

 at the time, this classification has many weak points and requires modifica- 

 tion. For instance, it is now generally recognised that the Cotylosauria 

 ^and allied Pareiasauria, Procolophonia, and Microsauria) belong to a low 

 grade of reptiles preserving the original complete roof of the skull, and other 

 Stegocephalian characters. Among these primitive forms will perhaps some 

 day be distinguished the ancestors of the more advanced Synapsida and 

 Diapsida ; but, for the present, they may be included in a provisional 

 assemblage of early reptiles forming a sub-gfade rather than a true sub- 

 order. In the second place, various orders are placed in Osborn's Diapsida 

 which on further evidence seem to belong to the Synapsida — such as the 

 Pelycosauria, Proganosauria, and possibly the Ichthyosauria. 



Useful and important as the roofing of the skull is in classification, it is 

 often difficult if not impossible to ascertain for certain its structure in fossils. 

 Moreover, as is always the case when we endeavour to classify by a single 

 character, we are liable to confuse forms in which foramina have begun 

 to appear with others in which they have been or are being secondarily 

 obliterated, and to misinterpret aberrant modifications. Consequently 

 many reptiles have been repeatedly shifted backwards and forwards 

 from the Diapsida to the Synapsida. Any corroborative evidence derived 

 from other parts would therefore be of value as a clue to affinity, and 

 it is the object of this paper to show that such evidence may be found 

 in the structure of the heart and of the skeleton of the hind foot. The 

 former unfortunately only applies to living forms ; but the latter is often 

 available even in fragmentary fossils. 



For the purpose of facilitating description, and of mak our results 

 clear, the grade or class Peptilia may be provisionally subdivme*cl into three 

 groups (fig. 1) : — In the first, which may be called the Protosaurian group, 

 we place the primitive forms connected with the Amphibia and leading on 

 from them to the central point of divergence of the Synapsida and Diapsida, 

 together with other side branches. The Protosauria, then, include the 

 Microsauria, Cotylosauria, Pareiasauria, and Procolophonia (see p. 269). The 

 second group, in which becomes developed one lateral temporal foramen 



