214 DE. R. BROOM ON THE [Mar. 21, 



developed independently in the two groups or that the groups had 

 a common ancestor in which most of the characters were present. 

 One or two may have originated independently, but it seems much 

 more reasonable to assume that the two phyla sprang from an 

 early true Reptile, than that they originated from a Batrachian 

 and that all the characters they have in common have been 

 developed independently. 



To what order the common reptilian ancestor belonged it is 

 difficult to say. The Cotylosauria of Cope is no doubt very 

 primitive, but unfortunately it is very imperfectly known. The 

 type genus on which it was founded is Diadectes, a form which is 

 not very well known, and the skull of which has so far not been 

 very satisfactorily figured. A considerable number of other genera 

 have been placed in the order by Cope and others, some at least of 

 which ai'e possibly not very nearly related to Diadectes. Pario- 

 tichifjS, which is placed in the same order, is much better known, 

 through the researches of Cope and, more recently, of Case. 

 It has a skull roofed as in the Labyrinthodonts, but a palate very 

 much like that of Sphenodon. But while Pariotichus might 

 perhaps have been the common Amniote ancestor, so far as the 

 skull is concerned, the condition of the shouldei'- girdle shows that 

 it is not primitive enough. It has lost the cleithrum which the 

 ancestor must have retained. Pareiasaur%is, though it retains the 

 cleithrum, is further removed from the ancestral type than Pario- 

 tichus^ but in another direction. It seems probable, however, that 

 there were forms somewhat resembling Pariotichus but sufficiently 

 primitive to have been the ancestor of Pareiasau7-us. If such a 

 form is discovered, then we may regard the Cotylosauria as con- 

 taining the commoK ancestor of all rej)tiles. 



The phylum which contains Pareiasaurus, Dicynodon, the 

 Theriodonts, and which culminates in the Mammals, probably 

 originated through the ancestral Cotylosaurian living in marshy 

 regions and having to walk with the body well supported off" the 

 ground. This habit caused the ilium to become directed mainly 

 upwards and forwards from the acetabulum, and necessitated the 

 retention of the precoracoid. From the start once given in this 

 direction, the evolution went on steadily till the mammal was 

 formed. 



The other phylum, which gave rise to the Lizards and Birds, was 

 probably started by some of the early Cotylosaurians having to 

 stalk insect prey on dry sandy or rocky places. The body rested 

 for the most part on the ground and the legs became relatively 

 feeble. As this was probably no very great change from the 

 newt-like gait of the ancestor, it was long before there was much 

 change in the structure. And some reptiles which are well 

 advanced along the Diapsidan line, such as Mesosaurus or Palceo- 

 hatteria, still retain the early types of shoulder-girdle and pelvis 

 with very little modification. In Palceohatteria the now useless 

 precoracoid ceases to be ossified, and in all the later Diapsidan 

 types there is no trace of a precoracoid bone. The plate-like 



