1012 
Journal of Agricultural Research 
Vol. XXVIII, No. 10 
others—7 males and 11 females. Needless to say, the allocation of a number 
of examples in regard to the presence or absence of sueh a trace is purely arbi¬ 
trary. As this tendency is present to a greater or lesser extent in at least 30 
per cent of the skulls, it is of considerable importance. The pattern which may 
be considered as normal is illustrated in figure 23, 6. 
Third molar. —The normal pattern of this tooth consists of an anterior cres¬ 
centic loop, two external, and one internal closed triangles, and posteriorly, two 
open, internal loops, highly variable in this species as in most other microtines. 
In 33 individuals the penultimate loop is rather constricted, and the posterior 
one is subtended by a distinct, intero-posterior notch (fig. 24, e). In two or 
three cases this indentation is sufficiently pronounced so that, in reality, one 
might consider that there are five internal salient angles instead of the usual 
four. As hardly any two individuals can be found in which this portion of the 
third molar is precisely duplicated, the allocation of many specimens is arbi¬ 
trary; but this general type may be considered normal, as it is the one more often 
found than any other. It insensibly grades into another subtype of pattern, 
represented by 26 individuals, in which the posterior enamel space has no marked 
indentation (fig. 24, d). In a third type, represented by 5 specimens, the in¬ 
ternal salient angle of the posterior enamel space is definitely hooked forward 
(fig. 24, c). In two skulls the posterior enamel space is exceedingly small, with- 
a b c d e 
Fig. 24.-—Individual variation in the pattern of the third upper molar , M3 : a, No. 3817; b, No. 3873; 
c, No. 3864; d, No. 3837; e, No. 3835 
out indentations (fig. 24, a), and in one (fig. 24, b) the penultimate loop upon the 
right side, but not upon the left, is entirely closed, a condition that may be con¬ 
sidered as aberrant. These five types of enamel pattern may be still further 
subdivided at the pleasure of the investigator, but nothing is to be gained there¬ 
by. There is also a tendency exhibited by a few individuals for other triangles 
to remain open, as illustrated by figure 24, d , and the posterior internal triangle 
is found in this condition not infrequently. 
It is entirely impossible to correlate these variational trends with age or any 
other character, and they can be considered only as evidence that the third upper 
molar is at present much more unstable in an evolutional sense than either of the 
two other maxillary cheek teeth. This is the last of the upper teeth to emerge, 
and hence it is relatively less worn and developed in small juveniles than in older 
specimens. 
Mandibular Molars 
First molar. —As is the case with the majority of forms in this subfamily, 
the first lower molar is the most variable as well as the most complex of all the 
cheek teeth. It is in the anterior, tripartite enamel space that this diversity is 
encountered, and the variation is well nigh infinite. Truly no two are alike, nor 
is there bilateral symmetry between the two rami in this respect, for as often as 
not each ramus must be assigned to a different group. If it be attempted to es¬ 
tablish too many groups for the different types of variants, however, a condition 
bordering on chaos results, with weakly-defined criteria for designation, and it is 
better to limit the number of main groups to three. 
