54 1 Ml-. R. W. Hooley on the 



apparently from the central area of the shaft, where the 

 bone is smallest in Pterodactyls. The complete bone would 

 therefore be of enormous bulk, aud^ with the other bones of 

 the skeleton in proportion, we should have a reptile too 

 iieavy for flight. Moreover, it reveals no constriction as seen 

 in the Pterodactyl humerus. Whether we take the proximal 

 or distal ends^ the forty-nine specimens naturally form three 

 groups. In one group the entire bone is known for certain 

 by the perfect example J. a. 6, 30, and. by comparison with 

 Ornithodesmus latklens another is nearly as sure, and there 

 is reason to believe that the third group is comprised of 

 those proximal ends which are different from the other two 

 groups, although there can be no positive determination 

 until a perfect humerus with the same characters is dis- 

 covered. We give on pp. 542-543 the characters of the three 

 groups and their examples, and follow by a criticism of some 

 of the specimens inclaJed in the groups : — 



Proximal ends. 



Group A. 



Scelcy* gives J. a. 6, 25 as an example of the same kind 

 of proximal end as seen in the perfect humerus, J. a. 6, 30, 

 " having the pneumatic foramen radially situated on the 

 anterior aspect near the articular surface.'''' The pneumatic 

 foramen is, however, found on the postaxial side posterior 

 surface as a small circular hole in an oval-shaped depression. 

 Onits distal margin it is worn away, but the foramen can be 

 well determined. In J. a. 6, 30 the deltoid crest is produced 

 obliquely to the long axis of the condyle, while in J. a. 6, 25 

 it is nearly a right angle with little or no curve until its 

 termination contiguous to the preaxial border of the shaft. 

 In J. a. 6, 25 the outer surface of the deltoid crest is flat. 

 As the crescent-shaped condyle in this specimen has a 

 greater curve than that of the other examples in this or the 

 remaining groups, a greater convexity is found on the dorsal 

 and a greater concavity on the ventral margin. J. a. 0, 26 

 approaches closely in character to J. a. G, 25 ; but, although 

 the crest is carried as far down the shaft as in that specimen, 

 it is produced more obliquely to the long axis of the condyle, 

 recalling J. a. Q>, 30 (Group C) in that respect only. 



* ' Oruitliusauria,' Ib/O, p. o9. 



