TnEKIODONTb AND IHEIK AlI.IES. 289 



PKI.VCOSALKIA. 



The researches of Baur and Case have conclusively shown that 

 the Pelycosaurians are distinct in organisation from the Theriodonis 

 of the Galesaurus tvpe. and are more nearly allied to the Rhyncho- 

 cephalians. Though the Pelvcosaurians have two temporal arches 

 thev show in manv points athnities with Procolophon especially in 

 the structure of the limh bones. Whether Pelycosauria and Procohj- 

 phonia should be regarded as distinct orders or as sub-orders of thf 

 large order Rhynchocephalia is a matter of little importance so long 

 as it is recognised that the affinities are with the primitive Rhyncho- 

 cephalians rather than with the Anomodonts and Theriodonis. 



PRIMITIVE THERIODOXTS. 



I"or some vears it has appeared to me that in the order Therio- 

 dontia as generallv accepted are included a number of forms not \eT\ 

 nearl) related to the typical genus. The palate of yElurosaurus, 

 though only imperfectlv known, is manifestly so very different from 

 that of Galesaunis. that if no other evidence were forthcoming it 

 would f)e necessarv to remove ^^^kirosatirus from the tvpica! 

 Theriodonis. 



Having recently, at the request of Mr. W. L. Sclater. made an 

 examination of the reptilian fossils in the South African Museum. I 

 came across one or two interesting small Theriodont skulls that had 

 been for many years in the Museum. The most perfect of the 

 specimens, which I propose to name Scylaco.saurus Sclateri, shows, 

 after having been developed, almost every detail of the structure of 

 the palate. And though the palate probably agrees fairlv closelv 

 with that of yElurosaurus. it is entirely different in type from that of 

 Galesaurus. Cynognathus. or Gomphognathus. Externally the .skull 

 differs but little from that of yElurosaurus. though the snout is longer 

 and more slender. On each side there are six incisors, followed 

 by two canines, of which the first is very small and the second large. 

 There appears lo be a third canine about as large as the second, but 

 which is not yet functional. Fcjllowing the canines are seven, or 

 possibly eighl. small i)ointed molars. 



The palate in type is essentialK Rhvnchocephalian. The 

 internal nares are situated in front, as in the Sphenf)don, and 

 separated from each other by the long, narrow, paireil " vomers " (or 

 prevomers. as 1 believe they ought to l)e regarded). In from these 

 paired " vomers "" meet a pair of short premaxillary palatine i)rocesses. 

 while posteriorly they form a consideral)le i)art of the hard palate, 

 separating the palatines and articulating with the anterior ends of the 

 pterygoids. The maxillaries form no part of the palate — the pala- 

 tines bearing the same relations to the maxillaries as is .seen in 

 Sphenodon. The pterygoids have well developed lateral processes, 

 which are supportefl by a pair of transpalalines. 



