No. 2.] THE HARD PARTS OF THE MAMMALIA. 245 
ercle), which, when it is developed, opposes the metaconid of 
the inferior molar. The metacone of the superior series op- 
poses the protoconid of the inferior. Thus, although the teeth 
of the superior and inferior series are alternate, they do not 
interlock. On the contrary, they oppose each other, forming a 
mechanism adapted for the grinding instead of the cutting of 
food. 
The divergence of mammalian dentition into the two types, 
the tritubercular and quadritubercular, has been, as it appears 
to me, due to the adoption of different food-habits. The tritu- 
bercular is the primitive, and is adapted for softer food, as flesh, 
so that primitive placental Mammalia were carnivorous or nearly 
so. The mastication of hard food was impossible until the 
molars of the two series opposed each other, and this was not 
accomplished until the quadritubercular superior molar was pro- 
duced. This was accomplished, as I have pointed out, by the 
addition of a posterior internal tubercle, and I suspect that the 
mechanical cause of its origin was the attempt of the animal in 
mastication to crush substances harder than flesh against this 
posterior edge of the superior molar, by applying to it the ante- 
rior edge of the lower molar. In the devouring of flesh this 
movement is not necessary, or only necessary so far as to pro- 
duce a shearing movement to cut a resisting ligament or tendon. 
The different mechanical movements in the two cases were due 
to the manipulation of its lower jaw by the animals, just as we 
may see them to-day endeavoring to masticate substances in 
accordance with their hardness, form, etc. It would appear in 
the case of the tritubercular superior molar, that the impact 
during the effort to masticate hard and tough substances, as 
vegetable tissues and seeds, has had its usual effect to stimulate 
deposit of material. The shearing movement has had an oppo- 
site effect, viz.: that of wearing away the surface subjected to 
it, and the flattening of the sheared face. That the develop- 
ment of the grinding mastication should take place in ungulate 
Mammalia is entirely appropriate to the structure of their digits ; 
the hoofed structure unfitting them for the seizure of living prey. 
The completion of the quadritubercular dentition requires 
that the two anterior cusps and two posterior cusps of the infe- 
rior molar should reach the same horizontal plane. This in- 
volves the lowering of the anterior pair, and the elevation of the 
