XEW FOSSIL REPTILES. 861 



sutures. The snout is missing in front of the nostril, but from 

 the shape of the lower jaw its length can be ii\iv]y well assumed. 

 The nostril is large and the distance between it and the orbit very 

 short. It is probable that the septomaxillary meets the lachrymal. 

 The suborbital and postorbital arches are slender. The frontal 

 and parietal regions are both fairly wide and there is a large oval 

 parietal foramen. The squamosal is somewhat like that of the 

 Dinocephalia, but is slender and the descending process very long. 

 There appears to be a distinct quailrato-jugal. The palate, so far 

 as preserved, agrees with that of the Therocephalia. There is a 

 pair of long slender prevomers, and the palatines and back of the 

 pterygoids are not unlike those of Scylacosaums, but whether in 

 other respects the palate is Therocephalian or Dinocephalian the 

 evidence does not show. Certainly the palate is not the least 

 like that of the Anomodont. The lower jaw is short and tooth- 

 less. It agrees fairly closely with that of the Anomodont, but 

 there is a rudimentary coronoid process. 



The shoulder-girdle has a large distinct precoracoid and coracoid 

 loosely articulated to each other. The scapula is somewhat like 

 that of the Dinocephalia and also a little like that of the Thero- 

 cephalian Ictidosuchus, but unlike that of the Anomodont. The 

 limb bones are long and slender. 



Galeops is an entirely new type of Dromasaurian and represents 

 a new family, the Galeopidse, characterised by the absence of 

 teeth and the presence of a small coronoid process. The other 

 family, which may be called the Galechiridfe, includes GcdecJiirus 

 and Galepus, both with teeth and without a coronoid process. 



Suborder Therocephalia. / 



SCYMNOGNATHUS WHAITSI, gen. et sp. u. (PI. XC. figs. 4, 5.) 



This interesting type was found by Mr. Whaits near Beaufort 

 West. Within quite a small area remains of four or five 

 animals were obtained. Unfortunately, most of the bones are 

 extremely weathered, and so infiltrated with lime as to be 

 practically limestone nodules so hard that any development i.s 

 almost impossible. Two skulls are suificiently well preserved to 

 show the general characters of the genus, though it is impossible 

 to be sure of most of the sutures. 



The new type is a very near ally of Oorgonops tortus Owen, 

 and it i,s only after considerable deliberation that I have decided 

 to place it in a, new genus. With perfect certainty it can be 

 placed in the Family Gorgonopidse. When Ow^en first described 

 Gorgono'ps in 1876 he unfortunately came to the conclusion that 

 the temporal fossa, was roofed over as in Labyrinthodonts or 

 PareAasaurus, and this mistake was also made by Lydekker in 

 1890, and by Seeley in 1895. On examining the type and, at 

 that time, only knoAvn specimen when in London three j'ears ago, 

 I discovered that the temporal fossa is not roofed over, though 

 the parietal region is broad, and noted the observation in a paper 



