278 Annals of the Carnegie Museum. 



have mc. V present and does not agree with the statement on page 

 159 of the same work; viz., "There are only three digits in the manus, 

 hence the name tridactylum, there being no trace of the fifth digit. . . ." 

 I infer, however, from his statement on pages 130 and 131 that mc. V 

 might be present in a rudimentary form in A. tridactylum. This 

 would seem perfectly natural in the light of present facts, since we 

 should otherwise be forced to accept a rather involved variation of 

 considerable morphologic importance. That is we should expect 

 the various species in the Oligocene, leading up to the Diceratheres 

 to be differentiated in a very marked degree so far as the absence 

 and presence of mc. V is concerned, and also that these differences 

 held good until the close of the lower Miocene. 



In all the material of Diceratherium from the Agate Spring Fossil 

 Quarries which I have seen, mc. V is present in a rudimentary form; 

 nor is this surprising considering the fact that this digit is yet of con- 

 siderable size in some of the earlier forms {Aceratheriiim tetradactylum 

 Lartet and Trigonias osborni Lucas). The late Eocene and early 

 Oligocene ancestors of Diceratherium had undoubtedly the manus more 

 or less tetradactyl, the fifth digit, however, was rapidly reduced in 

 later forms but was still present in a rudimentary condition, possibly 

 in all the species of that genus, at the close of the lower Miocene. 



The carpus is of equal height and breadth, the three functional meta- 

 carpals are of medium length with their shafts placed rather close 

 together and displaying but slight divergency at the distal ends. 

 The second phalanx of the median digit is raised very slightly from the 

 ground while the lateral toes appear to be more elevated. 



The pelvis is short and broad, the ilium being broadly expanded 

 with a large area for the gluteal muscle, while the supra-iliac border 

 is strongly emarginated as in older forms {A. tridactylum), but the 

 ischium and pubis are apparently shorter. The femur is quite long, 

 but comparatively slender, the different trochanters are, however, 

 well developed, indicating a heavy thigh, and the antero-posterior 

 diameter of the distal end is great in porportion, due chiefly to the very 

 large internal border of the rotular trochlea. One of the more charac- 

 teristic features of the tibia and fibula is the tendency of the proximal 

 and distal ends to become coossified in fully adult and old individuals. 

 The pes is strictly tridactyl. It is high and narrow with a broad, rather 

 low astragalus, a very heavy tuber of the calcaneum, and quite elon- 

 gated metatarsals. The ungual phalanges, especially the median pair, 



