No. 434-] 



THE AM ERIC. 



.YCOSAURIA. 



99 



ance in the quadrate region, and in all these forms it is notably 

 weak. The development of a chewing habit, either grinding or 

 sectorial, would bring a greatly increased strain upon this region. 

 We know little that is definite concerning the origin of the 

 temporal and quadrate region in the reptiles next above the 

 Cotylosauria, but as Baur has shown, this may well have 

 developed by a sort of natural trephining of the solid cranial 

 roof, perhaps in response to the demand for a lighter skull con- 

 nected with changing feeding habits, and resulting in the Pro- 

 ganosauria. 1 



In the Proganosaurian skull the quadrate was well developed 

 and had much the same position and proportions as in the 

 modern Sphenodon, but was of slightly less vertical extent. As 

 indicated by the character of the teeth these forms used the 

 jaws very little if at all for chewing. The beginning of a chew- 

 ing habit, either sectorial or the simpler kinds of crushing or 

 grinding, would demand a simple vertical motion of the jaws, 

 with a consequent demand on the quadrate to resist a stronger 

 thriLst directed vertically upward. The shape and character of 



of the American forms, as in Dimetrodon, make any other motion 

 than a vertical one impossible. 



The strengthening of the temporal and quadrate regions to 



ways: First bv an enlargement of the quadrate, especially in 

 the vertical direction, and its closer union with the skull wall; 

 this was accompanied by a decrease in the number of bones in 

 the temporal arches and an enlargement of those remaining 

 (compare the condition in the Squamata and Testudinata). 

 Second by a reduction in the size of the quadrate till it finally 

 disappears (in the mammals or Promammalia) and at the same 

 time a union of the bones of the temporal arches till the articu- 



