In Table 1 are the original data (2 sheets) provided by John Howe. One specimen was measured twice ; two seem to be the left and right of the same animal. In Table 2 are the data I used (diaphysis and proximal depths were not used because they are difficult to measure and bring little information) ; later on I discarded also the diameter of caput femoris because of its too large variation.

Scatter diagrams (Fig.1-2) show three very large and robust specimens ; a large variation exists even among the other femora. Most Natural Trap femora are bigger than in extant E. hemionus onager.

Robust Femora (Fig.3)
– The largest one is similar to the average Amerhippus leoni.
– Two smaller may belong to a Caballine.

Gracile and very gracile Femora (Fig.4)
I am unable to propose any attribution for these specimens.