the Slull of Diadcmodon. 321 



described it, there is no post-articular projection, and tlic 

 only possible insertion for the digastric is onto the small, 

 almost vertically placed posterior face of the articular. 

 This implies that the pull of the muscle mnst have been 

 approximately horizontal. 'I here is no evidence of any 

 nni9cle insertion which would allow a direct pull in this 

 direction, but any muscle passing from within the temporal 

 fossa over the notch and down the groove which separates 

 the njcdian from the lateral regions of the occiput would be 

 in exactly the right position. 



A muscle in this po:^ition is attached at its npper cud to 

 the parietu-squamosal arch as the digastric should be, and 

 its peculiar insertion on to the anterior instead of the poste- 

 rior face of this arch is simjjly due to the great development 

 of the muscles of the neck inserted into the median division 

 of the occiput. Tliis development of neek-museles is asso- 

 ciated w ith increase in size and of power of jaw in a carnivo- 

 rous animal living on land. 



It will be remembered that in the type skull of Cytio- 

 ynatlius crateronotus the muscular insei'tions on that process 

 of the squamosal which extends forwards on the front face of 

 the parietal point downwards and backwards towards the 

 groove. I take it that these are the insertions of the 

 digastric; if they are not, I do not know what they can be. 



The recognition of the fenestra vestibnli settles the position 

 of the tymi)anic cavity. As in all reptiles with a primitive 

 palate this lies outside the fenestra, and is surrounded above 

 and behind by the paroccipital process and in front by the 

 posterior ramus of the pterygoid. 



The position of the membrana tympani is more difficult to 

 settle, but it must lie in one of two regions. 



In the majority of reptiles the tympanic membrane is 

 ])laccd almost vertically, attached to the posterior border of 

 the quadrate, and such Ave should expect to be its position 

 here. We find, however, that this position lies in the middle 

 of the great smooth groove in the squamosal which leads 

 round from the space between the end of the paroccipital 

 process and the posterior ramus of the pterygoid to the upper 

 and outer surfaces of the skull. The character of the surface 

 of this groove, w hich in the Diademodon skull is peifectly 

 preserved, shows that this position is an impossible one and 

 that the membrane must have been at one end or the other 

 of the groove. 



In Diademodon the border of the outer end of the groove 

 is continuous in front, and at first presents some resemblance 



