56 MAEINE EEPTILES OF THE OXFOED CLAT. 



The radiale (rod.) is a rounded, somewhat pentangular bone, thinning towards its 

 anterior edge, which, however, is not sharp as in the radius, but was fringed with 

 cartilage throughout life ; it articulates with the radius, intermedium, and the first 

 and second of the distal carpals. The ulnare (uln.) is a roughly quadrate element 

 articulating with the ulna, intermedium, and the third and fourth of the distal carpals ; 

 it may also have been in contact with the pisiform. No centrale can be recognized. 

 The four distal carpals are more or less oval or rounded in outline ; the two middle 

 elements articulate with the intermedium. The remaining bones of the paddle are 

 arranged in four rows, at least in most specimens, the more proximal beingoval in outline, 

 with the long axis transverse, while the more distal elements may be merely small 

 rounded masses of bone. No specimen has been collected in which these bones have 

 been retained in the matrix in their natural positions, so that there is some doubt 

 as to their precise arrangement. In one of the best specimens carefully mounted 

 by Mr. Leeds as found in the rock, the first digit consists of three elements, the second, 

 third, and fourth of four each; but, judging from the other examples, the phalanges 

 may have been more numerous, the terminals being represented by mere nodules of 

 bone easily overlooked in removal. 



Another disadvantage of not having the paddles actually embedded in the matrix 

 is that it is impossible to say whether any of the digits ever bifurcated, as sometimes 

 happens in the Ichthyosaurs, or whether there were accessory ossicles along the 

 margins of the paddles : in one example it seems possible that there was a small radial 

 sesamoid which may have been in contact witli the humerus and perhaps formed the 

 proximal end of a rudimentary additional digit on the preaxial side of the fin. From 

 examination of the best-preseiTed paddles, it seems that there is no reason for 

 regarding the digits present as other than those of the primitive pentadactyl limb, and 

 the type of structvire might easily be derived from such a paddle as that of the Triassic 

 Mixosaurus cornalianus as figured by Merriam * by way of some latipinnate Liassic 

 Ichthyosaurus, as, in fact, has been suggested by Merriam f. 



Pelvic Girdle and Hind Limb. — The pelvis has undergone great reduction, con- 

 siderably greater than in the Liassic Ichthyosaurs. It is improbable that the ilium 

 articulated above with tlie sacral ribs, and the pubes and ischium are fused with one 

 another at their proximal and distal ends, their original separation being usually 

 marked only by a narrow slit-like foramen and occasionally a notch between their 

 ventral ends. 



The ilium (text-fig. 38) is a curved sickle-like bone laterally compressed into a thin 

 blade above, but thickening towards the acetabular end, where it terminates in 

 a cartilage-covered surface, which in some specimens is distinctly divided into two 

 portions, which make an angle with one another, one being the surface for union with 



* "Triassic Ichthj-osauria," Araer. Journ. Sci [4] vol. xix. (1905) p. 25, fig. 1. 

 t Loc. cit. p. 27. 



