1907.] OF MAMMAL-LIKE REPTILES. 1053 



the jaw is not well known. In Dimetrodon the jaw bears con- 

 siderable resemblance to that of Frocolojj/ion. There is a large 

 splenial on the inner side and a large angular on both the inner 

 and outer sides of the posterior part. There is also a large sur- 

 angular and a small articular. There is some evidence of a 

 coronoid element. In front of the articular on the inner side a 

 distinct element called the prearticular is said to occur. 



In Gcdechirus only the outer side of the jaw is known, and it 

 differs from that of the early Diaptosaurian types mainly in the 

 absence of a distinct coronoid bone. The angular forms more 

 of the outer surface and probably less of the inner. 



In the Anomodont such as Oudenodon (13), the jaw is fairly 

 similar to that of Galechirus except that there are no teeth. On 

 the outer side the angular is a large bone somewhat oval in shape 

 which forms the greater part of the posterior half of the jaw. 

 The surangular lies above it, but is for the most part hidden by 

 it. There is no coronoid bone. The splenial is well developed. 

 The articular is only in part hidden by the angular. 



The mandible of the Therocephalians is pretty fully known, 

 almost all details being known in both Lycosuchus (13) and 

 Hycenasuchus (14). The only important difference between this 

 type of jaw and that of the Anomodont is that the dentary has 

 its upper and posterior end developed into a large coronoid 

 process. 



In the Cynodonts the mandible differs greatly from that of the 

 Therocephalians, owing to the dentary becoming greatly developed 

 and the other elements greatly reduced. The dentary forms not 

 only a large coronoid process but nearly the whole of the back 

 part of the jaw, and hides the whole of the surangular and much 

 of the angular and articular. 



In the mammal the dentary forms the whole jaw, the rudi- 

 mentary elements having disappeared. 



The next important structure whose evolution may be con- 

 sidered is the shoulder-girdle. Fortunately this is well known in 

 most groups. 



In the Labyrinthodonts the girdle is made up of a large flat 

 interclavicle, with two large flattened clavicles and a pair of 

 slender cleithra — these membrane-bone elements supporting the 

 cartilage-girdle proper, which no doubt was made up of a per- 

 manently cartilaginous precoracoid and coracoid with an ossified 

 scapula. Of course in most specimens of the girdle proper only 

 the scapula remains as a fossil. In some forms, e. g. Eryops, the 

 whole girdle is ossified and we find well-developed coracoids and 

 precoi"acoids. 



In most of the Cotylosaurs the same eleven elements are met 

 with. In the Diadectidse there is a large precoracoid and coracoid, 

 with a large scapula which is supported in front by a well- 

 developed cleithrum. In Pao^eiasccanis there is also a well- 

 developed cleithrum, and the shoulder-girdle for the first time 

 has a distinct acromion formed by a twisting of the anterior 



70* 



