302 Transactions of the Royal Society of South Africa. 
has further figured a large vomer in this position in some G-orgonopsia. I 
have elsewhere expressed my inability to see a vomer separating the 
pterygoids in Gorgonopsian skulls which I have examined, and the evidence 
afforded by this Cynodont skull seems to re-open the whole question of the 
identity of the bone which forms the median septum of the anterior part of 
the skull and the roof of the front part of the pharynx. 
Is it the anterior prolongation of the parasphenoid ? or is it the fused 
prevomers of the Therocephalia ? Its position in front of, and slightly 
flanked by, the pterygoids, and between the palatines, would seem to 
homologise it with the prevomers ; and until further evidence of its actual 
connection with the parasphenoid is obtained, it would seem best to consider 
it as the fused prevomers. 
Basicranium. — The basicranial region is similar to that of Cynognathus 
in general plan, but there are significant points of difference. The pterygoid 
is large, the vomer extends back at least to the front of the median bar ; 
and posteriorly the pterygoids stop some distance in advance of the triangular 
basisphenoid plate and are separated from each other by an anterior pro- 
longation of the basisphenoid along the median bar. Laterally the posterior 
portion of the pterygoid is overlapped by the epipterygoid, which has a long 
articulation with the basisphenoid. Posteriorly the epipterygoid is con- 
siderably shortened. It has the same relations with the openings in the 
brain-case as in other Cynodonts, but it does not extend outwards to 
articulate with the quadrate, being separated from that bone by a consider- 
able gap and lying as a thin lamina on the mass of the pro-otic. Watson 
describes a somewhat similar condition in Protacmon brachyrhinus, but there 
the epipterygoid terminates posteriorly in a thickened margin. Anteriorly 
in our form there is a foramen between the pterygoid and epipterygoid, 
which continues forward as a groove along the dorsal surface of the pterygoid. 
This is the opening seen by Broom in Gomphognathus, and considered by him 
to be the internal carotid foramen. 
The basioccipital forms most of the floor of the brain-case behind the 
sella turcica, extending from the condyles to the dorsum sellse, where a 
fracture shows it lying on the pro-otics as a thin plate of bone with triangular 
section. It forms the inner boundary of the internal auditory opening and 
is perforated on each side by two foramina for the exit of the branches of 
the Xllth nerve. 
Seen in dorsal view the basisphenoid forms the floor of the sella turcica 
into which the carotid foramina enter. The carotid arteries enter the 
ventral surface of the basisphenoid far back, and thus pass nearly horizon- 
tally forwards through the bone. The side walls of the sella turcica are 
formed by the anterior superior processes of the pro-otics, which bones lie 
here between the basisphenoid and basioccipital. Anteriorly to the sella 
