liAItYTJIERIUM. 
175 
arises from the union of the two rounded ridges which form the outer and inner 
borders of the coronoid fossa ; it runs outwards and u})wards, forming a crest which 
projects strongly outwards and then a little backwards. There is also a ridge running 
up the front of the shaft from the lower end of the deltoid ridge. The head of the 
bone seems to have been large and rounded, but in no specimen is the upper end 
sufficiently w^ell preserved for descri[)tion. 
The groat expansion of the distal end of this humerus, arising from the large size 
of the condyles and the strong development of all the ridges and processes for the 
attachment of muscles, seems to indicate that the fore limb was employed for some 
other purpose than mere progression on land, possibly for digging or scraping up 
the ground. Ajiart from this great distal expansion the humerus is somewhat 
similar to that of Uintatherhim, resembling it in the small and deep olecranon 
fossa, the presence of the ridge running up the shaft from the lower end of the 
deltoid crest. At least, it may be said that this humerus is much more similar to 
that of the Dinocerata than it is to that of any Proboscidean. The differences from 
the humerus of Arsinoitherium are very considerable [cf. text-fig. 16) ; in that genus 
the form of the deltoid crest is different, there is no anterior ridge running up the 
shaft, while the form of the outer condyle and supinator ridge is dissimilar and 
approaches rather the Proboscidean type. 
A nearly complete specimen of the left radius (PI. XVII. figs. 8, 8 a, 8 b) was found 
in association with the upper and lower jaws above described. In general form it 
greatly resembles the radius of Uintatherium. The upper articulation for the humerus 
is very wide and consists of two parts, a much larger outer portion (o.), oval in outline 
with its long axis transverse and nearly twice the length ot the antero-posterior axis ; 
this surface is slightly concave in all directions. The other portion (7.) is very small 
and bent down at an angle with that just described ; it helped to form the articulation 
for the inner portion of the trochlear surface of the humerus. Beneath the expanded 
proximal portion, the shaft becomes trihedral, one angle being anterior while the flat 
posterior face appears to be roughened for union with the ulna. Towards the distal 
end the bone expands enormously : its posterior or postero-external face is deeply 
concave and must have fitted closely against the inner side of the ulna ; the antero- 
internal surface is convex. The very large distal articulation is incomplete on the 
outer side, and consists of two facets separated by a prominent ridge. Of these facets 
the smaller (sc.) for the scaphoid on the inner side is triangular in outline and 
concave from before backwards. The larger facet {lu.) for the lunar is concave in front 
and slightly convex posteriorly. It will be seen that this form of distal articulation 
is quite different from that found in the radius of Arsinoitherium (see text-fig. 17) 
or in Elephas, and that the structure of the carpus must also have been different, the 
scaphoid forming a relatively large share in the carpal joint compared to what it doe.s 
in the two genera referred to. In fact, both in its distal articulation and in its general 
form the radius here described resembles that of JJintatherium fairly closely. 
