Specime7is of South African Fossil Beptiles. 
23 
are also quite Therocephalian. So indeed are all the parts of the skull 
preserved. Owen is in error in fancying that the depressions seen at the 
front of the specimen are anterior nares. All the snout-like portion pre- 
served is merely the cast of the upper olfactory region, and probably 
about 3 inches are lost in front. 
The skull does not agree with any known Therocephalian, but I do 
not think there is any doubt that it must be placed in this group. A 
number of large Therocephalians are known from the same horizon, and 
this skull is probably that of an immature specimen of one of the large 
undescribed forms. 
^LUEOSAUEUS. 
This genus is founded on a very fine little skull which lacks the 
postorbital portion. In all respects it is a typical Therocephalian, and 
the missing portion is doubtless similar to that in typical members of the 
order. The published figures, and especially Seeley's restoration, give a 
most misleading idea of the skull. Owing to a thin film of lime having 
been mistaken for bone, the articulation for the jaw has been considered 
to be below the posterior margin of the orbit. In reality the jugal passes 
straight back as in other Therocephalians. Lying inside of the supposed 
deflected jugal can be distinctly seen the surangular and angular bones. 
On the left side the mandible is nearly complete, and shows the typical 
Therocephalian structure. The dental formula seems to me to be 
i 5, cl, m 5. 
Lycosaueus. 
Under this generic name are grouped a number of specimens in the 
British Museum, some of which pretty certainly belong to different 
genera. 
The type species Lycosaurus pardalis is founded on a very badly 
preserved skull. The right side is much weathered, and only the anterior 
part of the lower jaw is preserved. On the left side the bones are better 
preserved, but the whole skull is greatly crushed, and the published figure 
gives a very erroneous idea of the structure. The skull is typically 
Therocephalian, and the dental formula appears to be i 5, c 2, 7n 4, the 1st 
canine being small and the 2nd large. The molar teeth do not appear to 
be serrated posteriorly. 
The specimen which has been called Lycosaurus curvimola is a 
much better preserved skull, but I think clearly belonging to a different 
genus from the former. The parietal region is broader than in most 
Therocephalians. There appears to be no secondary palate and no 
teeth on the pterygoids. The molar teeth are simple and strongly 
serrated posteriorly. The dental formula is i 4:, c 1, m 6. The curving 
of the molar series is probably in part due to distortion. 
