260 TOWARDS THE HUMAN FORM 



resemblances that are sometimes more remarkable still, 

 we find certain differences that are no less striking. Tritylodon 

 had long incisors deeply imbedded in the jaws and of perennial 

 growth, no canines, but in their place a gap, as in Rodents. 

 The molars of Trirachiodon were multituberculate. There were 

 likewise no canines in the continuous tooth-series of Stereorachis 

 of the Permian of France, nor in Empcdias molaris of Texas, 

 which had palatal and vomerine teeth. Clepsy drops, Dime- 

 trodon, and N anosaurus of Texas, on the other hand, some- 

 times had more than two canines in each jaw, which never 

 happens in Mammals. Thus for these reptiles we have no 

 typical dental formula fixed by heredity, as for the placental 

 Mammals, from which it would be possible to deduce all 

 later formulas by simple reductions. The dental matrix was 

 altered in shape under the pressure of the teeth according to 

 the use to which the teeth were put. Naturally, the muscles 

 of the lower jaw were also modified according to the use the 

 animal made of it, and this entailed a corresponding modifica- 

 tion of the facial region to which the muscles were attached. 

 A zygomatic arch was constituted, recalling that of Mammals, 

 but somewhat differently composed. The chewing habit 

 adopted by the Theromorpha suffices to explain the resem- 

 blances the skull presents to that of the Mammals. A certain 

 identity in gait likewise explains the form of their scapula and 

 the construction of their pelvis. There is nothing to prove that 

 we are here in the presence of the ancestors of Mammals. 



In the Triassic, indeed, there were already small Mammals, 

 Marsupials like Dromatherium of the Carolinas, and Micro- 

 conodon. 



Certain Theromorpha, particularly Cynognathas, had all 

 the ways of our Carnivora, but were much larger. Their skull 

 was sixty centimetres long. They captured their prey and 

 carried it away in their jaws like our Tiger. The head in conse- 

 quence had to be firmly fixed. We also see the single condyle 

 of Pareiasaurns becoming broadened and kidney-shaped, 1 

 and finally assuming the two-fold shape seen in Mammals. 2 

 At the same time the cervical region was shortened. It had 

 only six vertebrae, whereas the dorsal region had twenty-nine. 

 The existence of such carnivores also supposes the contem- 



1 Cynognatus platyceps. 



2 Cynognatus berryi. 



