PELOBATES. 199 



not entirely furnished by it, as is sometimes the 

 case.* 



Praecoracoids slender, strongly curved, not entering 

 the glenoid cavity ; coracoids nearly straight, strongly 

 expanded distally ; a cartilaginous omosternum ; ster- 

 num with a bony style dilated proximally and a large 

 discoid cartilage. Humerus considerably longer than 

 radius-ulna. Eight bones in carpus, three of which are 

 in contact with radius-ulna ; a single bone in the pollex. 



Pelvis five-sevenths to five-sixths the length of the 

 vertebral column; ilia with an inner upper groove 

 into which the sacral processes slide; pubis absent 

 or reduced to a small bony nodule, not entering the 

 acetabulum. Femur and tibia with cartilaginous epi- 

 physes ; tibia shorter than the femur, which is shorter 

 than the pelvis. Calcaneum slightly shorter than 

 astragalus, half as long as tibia; three distal tarsal 

 bones ; pr^hallux with a single very large, compressed, 

 curved phalanx. Terminal phalanges pointed. 



* 1 have long ago pointed out tliat the vertebral column figured by- 

 Gene in 1839 as of Bombinator belongs to a Pelohates. I should not 

 allude to this again were it not for the fact that Gene's error has crept 

 into Bateson's ' Study of Variation,' p. 127, in a paragraph marked 

 with an asterisk as a sign of special importance to the example quoted, 

 and with the addition, by the compiler, of a further error, viz. that the 

 specimen came from Sardinia. The sacrum in the Gene specimen, as 

 well as in others that have since been described and figured by Adolphi 

 (' Morphol. Jahrb.,' xx, 1895, p. 449, pi. xix;), is formed entirely by the 

 pj-ocesses at the base of the urostyle, and there are thus nine instead of 

 eight prsesacral vertebrae. 



