Dr. C. W. Andrews — The Fore-limb of Metriorhynchus. 445 



vol. xlix, 1902, p. 56, pi. viii, fig. 3), but in his specimens the bones 

 of the proximal pair are much larger than the distal. If this 

 interpretation is correct, then the distal carpals must have disappeared 

 or, as seems unlikely, have fused with the proximals. The first 

 metacarpal is an expanded plate, the other three rod-like ; the fifth is 

 wanting in Ammon's specimen, but is figured by Fraas, who also 

 shows the number of phalanges as 2, 3, 4, 3, 2, but this is very 

 uncertain; in Geosaurus gracilis Ammon figures a single phalange in 

 each of the first three digits. 



Recently Mr. A. N. Leeds has collected from the Oxford Clay of 

 Peterborough a nearly complete left fore-limb of Metriorhynchus, 

 together with portions of the right. 



Fortunately the clay in which the left carpals and proximal ends 

 of the phalanges lay has been preserved, and from the impressions of 

 the bones their position can be determined. On the right side the 

 second, third, and fourth metacarpals are in situ in the matrix, and 

 there is an impression of the proximal end of the fifth. The 

 humerus, radius, and ulna on both sides are unfortunately completely 

 freed from matrix, and their exact relative position remains 

 doubtful. Figure 2 shows what is regarded as the most probable 

 arrangement of the bones of the left paddle (p. 446). 



The humerus (h.) is much compressed from before backwards, and 

 its preaxial border is produced into a strong deltoid crest, forming 

 a triangular expansion and terminating in a small tuberosity. The 

 head is oval and is strongly convex. Beneath the deltoid crest the 

 shaft is a flattened oval in section, and the distal articulation is 

 a slightly convex facet occupying the whole of the end of the bone 

 and running up a little on to the preaxial border. 



The radius (a) is a flattened plate the edge of which is thickened, 

 and bears an articular facet on its upper inner border and another at 

 its distal end. In front of these facets the body of the bone forms 

 a large flattened expansion, projecting considerably preaxially. This 

 bone is regarded as the radius (1) on account of its larger size, there 

 being a tendency for the preaxial bones of a paddle to become enlarged, 

 a condition here seen also in the radiale and first metacarpal ; 

 (2) because its distal articulation agrees very well with the upper 

 articular surface of the radiale. 



The bone here regarded as the ulna (b) is smaller than the radius. 

 At its upper end it is much compressed and bears an oblique facet 

 for the humerus ; its posterior border is produced into a thin, sharp 

 process, marked with striae directed towards its apex — possibly this 

 served for the attachment of a strong tendon. Distally the bone 

 narrows and thickens, terminating in a rounded facet for the ulnare. 



As already remarked, the exact relations of the radius and ulna to 

 one another and to the neighbouring bones are uncertain, they being 

 freed from the matrix, but in the case of the carpals and metacarpals 

 the case is different, the bed of matrix in which they lay being 

 preserved. The radiale (c) is the larger of the two carpals ; its proximal 

 border is thickened and bears a straight articular surface for the 

 radius ; its anterior border is convex, while the posterior is concave, 

 and bears at its lower end a facet for the ulnare. Distallv it 



