— 325 — 
Osborn the carpus is entirely devoid of ossification. Osborn, 1903 
Fig. 17a). The radiale is the largest element of the carpus and is 
in close union with the intermedium. 
M one case, indeed, they seem to be coossified, but this may be 
artefact. Between intermedium and ulnare is a small passage for 
transmission of a blood-vessel, probably the arteria perforans meso- 
podii, as described in Sphenodon by Howes. The ulnare gives 
attachment to the fifth metacarpal. Of the four distal carpals, the 
first is the largest. The first metacarpal is shorter and thicker than 
the other four, which are of about the samelength. In none of the 
specimens was the manus complete, but by comparison with Ste- 
reosternum material in the American Museum of Natural History, 
I think there is no doubt that the phalangeal formula for the manus 
is 2,3,4,4,8. Osborn’s statement that the fourth digit of the manus 
of Stereosternum has five phalanges is an error, probably only a 
typographical error, since his figure shows but four (Osborn, 1903, 
p. 488 and fig. 17a). The terminal phalanges are short and conical 
and probably boreonly vestigial claws, and there is reason to believe 
that the digits were more or less united by a web. 
Hind limb. 
The length of the femur is almost exactly that of the humerus. 
It is nearly straight with only the slightest indication of the 
S flexure seen in many reptilian types. The proximal end is slightly 
rounded but there is no very distinct head and in none of the six 
examples studied is any trochanter discernible. The distal extremity 
exhibits a convex articular surface of smooth triangular form. I may 
here again refer to the fact that both genera of the Mesosauridae, as 
fossilized, usually have the femur placed horizontally at approxima- 
tely right angles to the body axis, with the knee joint markedly 
flexed, recalling the swimming position of the hind leg in urodele 
amphibians, and I am inclined to believe that motion at the hip joint 
was relatively limited, being chiefly the fore and aft motion of 
swimming. 
The tibia and fibula are of practically equal length, about three- 
fifths that of the femur. In two examples of Stereosternum the ratio 
was about four to seven, so that, in so far as this limited 
series of measurements is of value, it appears that in Mesosaurus 
