No. 2.] THE HARD PARTS (OF THE MAMMALIA. 239 
first the origin of the dental peculiarities of the Corypho- 
dontidee. 
First, no posterior inner tubercle was developed on the supe- 
rior molars. We may regard this as a consequence of the fact 
that a transverse (ectal) movement of the lower jaw was estab- 
lished before the appearance of this cusp, instead of after it, as 
was the case in other ungulate orders, and because the shearing 
has been always from before backwards, instead of overlapping 
from behind forwards, as in all other Ungulata. The stimulus 
already assigned as the cause of the development of the fourth 
tubercle is, under these circumstances, wanting (Fig. 64). 
Second, the anterior cingulum, which extends from the inter- 
nal cusp to the anterior external angle of the crown along its 
anterior base, is greatly developed. This may be reasonably 
ascribed to the stimulus produced by the friction of the poste- 
figure 64.— Mutual relation of inferior and superior molars of a Coryphodont 
Amblypod. Relations and letters as in Fig. 58. The superior series is that of the 
Ectacodon cinctus Cope; the inferior, that of Bathmodon radians (supposed); hence 
the anterior V of the inferior premolars is too far anterior. 
rior limb of the anterior V of the inferior molar in the trans- 
verse movement in mastication. The anterior crest of the 
superior molar is developed instead of the corresponding poste- 
rior crest of the superior molar in front of it, because the trans- 
verse movement of the inferior molar follows a path much more 
nearly coinciding with the anterior crest of the superior molar 
than with the posterior crest. That is, it follows a curved path 
of which the centre is posterior, and near or between the glen- 
oid cavities on which the mandibular rami move, as has been 
described by Ryder in various other Ungulates.! This is the 
probable cause of the development of this crest from its origi- 
1 Proceedings Philadelphia Academy, 1878, p. 56. 
