170 
MASARU NAGAO 
molar. This phenomenon is probably produced by the failure of the 
upper and lower bicuspids of the raccoon to occlude closely in the 
living specimen. 
4. The group of skulls in which the arrangement is best represented 
by a straight hne. This kind of curve is met with in the jaw of Sus 
barbatus (wild pig). Some individual variations were noted among 
the skulls belonging to this species. Among five skulls of Sus bar- 
batus examined by me, specimens 2210, 1569, and 6150, have shown 
a sHght curvature with its convexity downwards, while specimen 
2853 has the convexity upwards. 
III. SELECTION OF THE STANDARDS TO BE USED FOR THE PURPOSE OF 
COMPARING THE CURVE OF SPEE GIVEN BY VARIOUS MAMMALS 
In the preceding section it was shown that in some species of mam- 
mals the projection of the buccal cusps of the bicuspids and molars 
upon a plane forms the arc of a circle. These points lie in the same 
cylindrical surface, as pointed out by Spee; and the line connecting 
them, when projected, has been designated "the curve of Spee" by 
subsequent writers. Since the precise form of the curve of Spee is 
not always the same, it has been my purpose to find some convenient 
standard by the use of which the differences can be expressed quanti- 
tatively. For this purpose Spee himself (1890) selected the length 
of the radius of his curve and the results of his observation are given 
by him in a chapter entitled: "Specielle Befunde an verschiedenen 
Gebissen." This method of comparison is valuable when the degrees 
of the curvature of the various curves of Spee are compared with each 
other. I have therefore followed the principle of Spee and deter- 
mined the radius of the circle in order to determine the degree of 
curvature. 
Since the curve of Spee is obtained by connecting the summits of 
the cusps of the bicuspids and molars, the length of the arc cannot be 
represented by the radius alone. The full form of the curve of Spee 
in any instance is determined by both its radius and the length of the 
arc. It is thus clear that in order to make a comparison of the curve 
of Spee as given by various mammals, we need, besides the radius, 
which represents the curvature at any point, also the length of the 
