APRACTOCLEIDUS TERETTPES. 625 



marked with deep rugosities, which continue down the posterior margin to the inner 

 surface of the shaft of the bone, where they merge with the deep pits and tubercles 

 occurring; there for muscle insertion. The shaft of the femur is oval in section, 

 smooth and somewhat flattened in front. The distal expansion is not so wide as 

 in the humerus, but it is 4 '5 cm. thick in the centre, which is greater than the 

 corresponding measurement of the humerus. The surface of the expansion is rugose 

 and plicated, the plications spreading out fan-wise as in the humerus. Since nothing 

 distal to the femur has been preserved, it is uncertain with how many elements it 

 articulated. Going on the similarity to Tricleidus presented by the fore limb, it 

 might be expected that, as in that genus, the femur articulated with two elements 

 only, tibia and fibula. However, the facets on the distal end of the femur of 

 Apractocleidus do not lend support to this supposition, but point rather to articula- 

 tion with four elements as in the humerus. The facets on the femur are smaller, less 

 well differentiated and slightly different in inclination in the pOst-axial portion, but 

 otherwise they are essentially similar to those in the humerus. Thus it seems 

 probable that the femur articulated with tibia, fibula, and other two elements similar 

 to the pisiform and accessory ossicle of the fore limb. These two additional elements 

 would probably be smaller and less definite in shape than in the fore limb, for the 

 facet on the post-axial portion of the femur contains a rounded depression indicating 

 that the fourth element may have been no more than a rounded ossicle. 



From the foregoing description of Apractocleidus it will be seen that, of pre- 

 viously described Oxford Clay Plesiosaurs, it resembles most the genera Crypto- 

 cleidus and Tricleidus. Though in some respects intermediate between those two 

 genera, in others it gives indications of a higher organisation than either, and 

 seems to combine the strong points of both. The fore limb (text-fig. 8 ; Plate, 

 fig. 4) is the chief illustration of this feature. The humerus (B, hum.) compares 

 favourably with the humerus (C, hum.) of Cryptocleidus in size, shape, ossification, 

 and width of distal expansion, and in those same criteria of efficiency it is immeasur- 

 ably superior to the humerus (A, hum.) of Tricleidus. The last genus compensates 

 to some extent for lack of width of distal expansion by the presence of a well- 

 formed ossicle (a.) articulating with the post-axial portion of the humerus. This is 

 lacking in Cryptocleidus but present in Apractocleidus. As in Ophthalmosaurus, 

 the presence of a large pisiform bone articulating with the humerus provides 

 "a broad base for the expanded terminal portion of the paddle" (Andrews, 1915). 

 In Cryptocleidus the humerus articulates with only two elements (r. and u.), although 

 in some specimens an indefinite ossicle attached to the ulnare comes up into the 

 position of the pisiform (Andrews, 1910, p. 184, text-fig. 91, A, a.o.). 



The development of the shoulder-girdle has proceeded exactly on the lines 

 indicated by Dr Andrews (1910, p. 107), and has been carried a stage further than 

 in Cryptocleidus : — " and from the Nothosauridse through the Plesiosauridse to the 



TRANS. ROY. SOC. EDIN., VOL. LI, PART III (NO. 19). 91 



