1888.] 85 [Cope. 
inferior molar follows a path much more nearly coinciding with the ante- 
rior 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 be- 
tween the glenoid cavities on which the mandibular rami move, as has 
been described by Ryder in various other ungulates.* This is the proba- 
ble cause of the development of this crest from its originally moderate 
proportions in Pantolambda (Fig. 1), and from the unknown ancestor of 
that genus, where its dimensions are presumably still less considerable. 
Third, the anterior external tubercle or V is reduced to a conical rudi- 
ment (Fig. 3a). This is evidently due to the disuse following the great 
development of the anterior cingulum which extends from the internal 
tubercle to the anterior external angle of the crown. A similar but less 
considerable development of this ridge is accompanied by a corresponding 
reduction of the anterior external lobe, in some genera of the Lophio- 
dontid Perissodactyla. The reason why this V has been extinguished and 
not merely pressed backwards, is the fact that the posterior external V of 
the superior molar has retained its place, and has not given way to allow 
room for the anterior one. This V has retained its place partly on account 
of its remoteness from the source of pressure in front, but principally 
because it fits the posterior transverse crest of the lower molar in front, 
and the anterior oblique crest of the next succeeding lower molar behind, 
so that its use has been only possible in its primitive position. : 
Fourth, the posterior limb of the posterior external V of the superior 
molar is wanting on the last molar in Coryphodon, and from the last two 
in Metalophodon (Fig. 3). The absence of this crest from the last supe- 
rior molar is due to the absence of a corresponding crest of the inferior 
molar (Fig. 4). This is the oblique crest at the anterior extremity of the 
inferior molar, and it shears against the posterior limb of the posterior 
external V of the superior molar, representing the sectorial blade of Car- 
nivora. It is little elevated in the Coryphodontide, owing to the fact that 
it is little used, since the crests of the inferior molars shear backwards and 
not forwards on those of the upper. The effect of this disuse tends, in the 
history of the Coryphodontidex, to become more and more evident. The 
non-existence of a fourth molar- behind the third in the lower jaw, 
accounts for the absence of the crest in question from the last superior 
molar, while the absence of the same crest from the second superior molar 
of Metalophodon, indicates the absence or rudimentary condition of the 
corresponding crest of the corresponding inferior molar. 
* Proceedings Philadelphia Academy, 1878, p. 56. 
+I have just detected an error in Plate xlvi, Tertiary Vertebrata, which has been 
copied in American Naturalist, 1884, p. 1198, by which the artist has drawn the left ramus 
mandibuli of Bathmodon radians in the place of the right one. The two rami are in the 
specimen separate from the symphysis, and the artist has simply drawn the ramus in 
connection with the wrong branch of the symphysis. I had not noticed this egregious 
blunder until the present writing, and no one else appears to have observed it. 
