256 COPE. [Vou. III. 
posterior cingulum of the superior molar suffers attrition from 
the anterior crest of the anterior exterior cusp of the inferior 
.molar. In Anchitherium this develops a tubercle which be- 
comes ultimately a short crest directed outwards and_back- 
wards. This forms the third or posterior intermediate tubercle. 
It is wanting from the rhinoceros line. The intermediate tu- 
bercles, now crests, have the following relations to the external 
crests, now V’s, of the inferior molar. The anterior is parallel 
to the posterior limb of the anterior V; the median is parallel 
to the posterior of the second V; the third is parallel to the 
anterior limb of the anterior V. The posterior forms, then, with 
the anterior of the tooth following it, an incomplete V, with its 
apex external, and therefore corresponding in direction and posi- 
tion to the anterior V of the inferior molar. The second in- 
ferior V is only met by a representative of its posterior branch 
in the upper molar. But a crest of the upper molar correspond- 
ing to its anterior branch develops later in the germs Anchip- 
pus. This branch first appears as a process from the median 
intermediate crest (Fig. 82, zap) extending anteriorly towards 
the anterior intermediate. In Anchippus the posterior and 
median intermediate tubercles unite and form a crescent oppo- 
site to the external, and connected with the posterior internal 
tubercle. In Hippotherium the anterior branch of the median 
joins the posterior interior extremity of the anterior median, 
thus forming another intermediate crescent opposite the ante- 
rior external. These intermediate crescents have exactly the 
same mechanical relation to the inferior crescents that the ex: 
ternal superiors have ; they alternate with them, and their horns 
cross obliquely in lateral mastication. We have here another 
illustration of the law that the horus of molar crescents are pro- 
longed in the direction of thrust by the corresponding parts of the 
opposing molar. The interior tubercles remain small in the 
horse line for the reason already given, —their little use. The 
only exception to this rule appears at the end of the series, 
where in the genus Equus the anterior internal cusp becomes 
enlarged like the others. 
In the rhinoceros line, where the posterior intermediate does 
not appear, the anterior intermediate falls into line with the 
transverse crest, and remains so. The median intermediate, on 
the other hand, develops an imperfect crescentic form in the 
