1907.] OF MAMMAL-LIKE EBPTILES. 1059^ 



and slender and passes straight up. In Lahidosaurus (9) the 

 ilium is relatively small and is directed somewhat backwards.. 

 In Pareiasaurus (26) the pelvis has a large ischium and a smaller 

 pubis, and though much specialised, they can still be referred to 

 the plate-like type. The ilium, however, is quite unlike that of 

 any of the early types and resembles that of the Anomodonts 

 and Mammals in having a large crest which is directed markedly 

 forwards. 



In passing to the Therapsida the earliest type known is that 

 found in Gcdechirus. Here the ilium is directed vipwards and 

 the crest is short. The pubis and ischium are of the plate-like 

 type. The pubis is nearly square, with the anterior and outer 

 angles bent downwards, so tlaat, though the ischia make with 

 each other an angle of about 90°, the fronts of the pubis are nearly 

 in line. There is a round pubic foramen. The ischium is, as 

 usual in the plate-like type, longer than broad. 



The pelvis in the typical Therocephalians is unknown. In the 

 Dinocephalian Titanosuchus the ilium is dii-ected mainly upwards 

 and has a short crest. The pubis and ischium are unknown. 



In the Anomodontia the pelvis is well known in a number of 

 genera. The tyj^e seen in Oudenodon differs little from that 

 of the other known forms. The ilium is directed well forwards 

 and has a large crest very like that in some Mammals. The pubis 

 and ischium are somewhat removed from the plate-like type. 

 The foramen, though small, lies between the pubis and ischium, 

 and is thus a true obturator foramen. 



In the Cynodonts the pelvis is well known in Cynognathus (17) 

 and Diademodon (27), and less perfectly in some other genera. 

 In general, the structiu^e is strikingly mammal-like. The ilium is 

 directed mainly forwards and has a very long crest. The pubis 

 and ischium are almost typically mammalian, owing to the presence 

 of a large obturator foramen. 



Fi'om the consideration of the comparative anatomy of these 

 skeletal structures it will be seen that the mammal-like reptiles 

 form a well-defined group, whose earlier members show so much 

 affinity with the primitive Diaptosaurians and with the higher 

 Cotylosaurians as to render it highly probable that from some 

 Cotylosaurian ancestor all the later reptiles are descended. On 

 the other hand, the higher mammal-like reptiles approach so 

 closely to the mammals that it is not always possible to distinguish 

 between them. Tritylodon is held by many to be a reptile ; by 

 others it is believed to be a mammal. Dromatheriuni, Micro- 

 conodon, and Karoornys are generally believed to be mammals^ 

 but it is just possible they may be reptiles ; while Sesamodon and 

 Melinodon, which are believed to be Cynodont reptiles, may 

 possibly prove to be mammals. The difference between a 

 Cynodont reptile and a Monotreme is less than the difference 

 between a Monotreme and a Marsupial, and this again is not 

 much greater than that between a- Marsupial and an Insectivore. 



