42 
TERTlAJiY VERTEREATA OF THE FATOM. 
the up{)or cdg'o o(‘ the lunar. The distal end is occupied by a greatly elongated 
articular surface {tz.), slightly convex in all directions, which probably articulated with 
both the trape/iuin and trapezoid. Anteriorly the outer (postaxial) border of this 
surface is connected with two other facets, of which tlie anterior one (/.') is a flat 
elongated surtace looking outwards and downwards for union with the lower facet of 
the lunar, while the other {mag.) lying immediately behind the last is triangular in 
outline, and looks downwards and backwards, and apparently united with the postero- 
internal angle of the upper surface of the magnum, as is the case in Elephas. Both the 
outer and inner faces of the bone, where not occupied by articular facets, are greatly 
roughened and pitted. 
The form of scaphoid here described and figured is definitely known to be that 
Test-fig. 21. 
Eight lunar of Arshioitherium zittcli : A, from above; B, from front ; C, sea[)lioicl face. 
nuKj., facet for magnum ; r., surface for radius ; sc.', upper and lower surfaces for scaphoid ; 
0-.?, facet for trape/.oid ; u., surface for ulna. ^ uat. size. 
of Arsinoitheriuvi, but at the same time it is very like that of Eleplias, and its 
relations to the neighbouring bones, so firr as determinable, are similar. This scaphoid 
difiers from that of Elephas in the following respects : — (1) its proximal articular 
surface is concave, owing to the stronger convexity of the portion of the radial head 
with which it articulates ; (2) the lower surface for the lunar is shorter from before 
backwards and at the same time deeper ; (5) the distal articulation is much longer from 
before backwards, extendiug considerably behind the triangular surface for the magnum ; 
the whole distal end of the bone is, in fact, much more expanded than in Elephas. 
The lunar (text-tig. 21) here described is presumed to belong to Arshioitherium, 
