216 DR. E. BROOM ON THE [Mar. 21, 



and is only a roofing scale of bone. It is probably the homologue 

 of the bone usually called " epiotic " in Stegocephalians, but it is 

 evidently a membrane-bone and not developed from the auditory 

 capsule, and hence not a true epiotic. A similar bone is found in 

 Pariotichus, but is lost in all the higher forms. If we omit from 

 consideration this so-called epiotic, we find the parietal, post- 

 orbital, squamosal, jugal, quadrato-jugal, and quadrate bones 

 having exactly similar relations to each other in Procolojjhon and 

 Sphenodon, the chief difierence being that there are two fenestra? 

 in the latter form. 



The condition of the teeth I do not look upon as a character of 

 much importance in the matter of broad classification, but the 

 teeth in Procolophon are by no means thecodont in the ordinary 

 sense. Owen, in 1876, rightly stated that " the base of the tooth 

 seems to be confluent with the osseous substance of the jaw " ; 

 and Lydekker, in 1890, stated that the marginal teeth are 

 " completely anchylosed to the bone." Most likely in the young 

 condition the teeth developed in sockets, but in the adult they 

 must be regarded as much more acrodont than thecodont. 



Boulenger states that " the thecodont dentition, the absence or 

 great reduction of the plastral bones, and especially the presence 

 of ossified precoracoids, are characters which are opposed to the 

 association of the Procolophonia with the Rhynchocephalia." 

 But, as has just been mentioned, the teeth cannot be regai-ded as 

 thecodont, being nearly as typically acrodont as in Sphenodon ; 

 the plastral bones are quite as well developed in Procoloplion as in 

 most Rhynchocephalians ; and the presence of ossified precoracoids 

 in Procolophon cannot be urged as a reason for removing it from 

 association with the ancestral Rhynchocejohalians, since the early 

 Rhynchocephalians must have had ossified precoracoids, if the Pely- 

 cosaurians are descended from them, as is believed by Boulenger. 

 If Procolophon is to be removed from a position near the ancestral 

 Rhynchocephalians and placed near Pareiasaurus, it must be for 

 other reasons than those advanced by Boulenger. 



Some additional evidence in favour of placing Procolophon in 

 the Diaptosauria, or at least among the ancestral Diapsidan 

 reptiles, is to be found in the striking resemblance which it bears 

 in many points of structure to Mesosaitrus. Unfoi^tunately, the 

 skull of Mesosaitrus is imperfectly known, but all the parts of the 

 skeleton that can be compared are fairly similar to those in 

 ProcolopTion. There is an ossified jorecoracoid, anchylosed, how- 

 ever, to the coracoid and scapula, and the pubis and ischium 

 closely resemble those of Procolo]ilion. Though the carj)us is 

 imperfectly ossified, there can be little doubt, when that of 

 Stereosternum is considered, that it has been modified from a 

 Procolophon-like type. The tarsus is almost identical in structure 

 with that oi Procolojjhon — the intermedium uniting similarly with 

 the tibiale, and a foramen being formed between the conjoined 

 bone and the fibulare. The plastron is closely similar in the two 

 forms. I have recently tried to show (Trans. S. Afr. Phil. Soc. 



