48 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 57. 



could have been derived. If we compare such reptiliian skulls as 

 those of Cynognaihus, Procolophon, and SpTiendon it will be seen that 

 they differ from each other considerably not only in the dis- 

 position of the component bones and the arrangement of the vacui- 

 ties, but in the general shape and form as well. Thus Ci/nognatlms 

 has an elongated, compressed, narrow type of skull with a small 

 orbital cavity and a large supratemporal vacuity, while Procolophon 

 exhibits a short, broad, flat type of skull without supratemporal 

 vacuity and with a large orbital cavity. The large temporal area in 

 Cijnognathus with its strong sagittal crest and a well-developed coro- 

 noid process of the lower jaw is in direct correlation, moreover, with 

 the size and strength of the temporal and pterygoid muscles, as well 

 as with the enlarged canines and the more or less molariform char- 

 acter of the teeth. These features are in marked contrast with the 

 small temporal area, the reduced coronoid, and the comparatively 

 weak development of the teeth in Procolophon. Then, again, in this 

 latter genus we observe how the parietal is extended laterally and 

 sends down a strong process behind the orbit to assist in forming the 

 postorbital bar. If this latter condition exists in any Cynodont I 

 have been unable to find any mention of it or refer to any figure 

 showing it. 



If now we turn to the skulls of the mammals we see that these same 

 different types are to be met with among them. Thus all the Carni- 

 vores and carnivorous Marsupials exhibit the com])ressed, narrow, 

 elongate skull, with large temporal area and prominent crests for the 

 attachment of the temporal muscles in directcoiTclation v.itn the en- 

 larged coronoid and the powerful laniary canines. The eye cavity is 

 relatively small and the postorbital process of the frontal has been 

 shoved far forwards in advance of the junction of the frontal and 

 parietal where these two bones meet above the orbit. On the other 

 hand, taking such types as Rhynchocyon, Tupaia,Procavia, Galeopterus, 

 and to a less extent the skulls of the Eodentia, Primates, and 

 Myrmecohius, it will be observed that the cranium is relatively short 

 and broad, the temporal area is reduced, the coronoid of the mandible 

 is small and the parietal sends a large })rocess do^\^l wards and for- 

 wards to form either a large part of the postorbital bar or to contrib- 

 ute to its make-up. This is especially true in the case of the first 

 four of these types in which the [)Ostorbital process springs either 

 from the point of junction between the frontal and parietal or largely 

 from the parietal alone, while in the others the origin of the postor- 

 bital bar above is slightly in advance of the parioto-frontal suture. 



If this latter type of skull arose from a reptilian condition such as 

 is seen in Procolophon — and this would seem to be reasonably de- 

 manded by the facts— then this region of the skull of Rhynchocyon 

 must be the most primitive of all this group, since it most resembles 



