GENETICS: W. E. CASTLE 
261 
Fi female hybrids between Cutleri and race B are practically as large 
in skull-length as race B females. Their means differ by less than half 
a millimeter. This harmonizes with the observations on general size as 
indicated by the weights given in the growth curves. The empirical 
range of the Fi hybrids is from class 12 to class 24, just within the limits 
of variation of race B . The F2 hybrids are considerably lower in mean 
skull-length than their Fi parents : indeed they are strictly intermediate 
between pure Cutleri and race B, in complete agreement with the growth 
curves (fig. 2). Their mean skull- length is 54.35 mm.; the middle point 
betwen the skull-length of Cutleri and race B is just 0.5 mm. greater. 
Their range (if we leave out of consideration one aberrant individual) 
extends from the middle of the range of Cutleri to the middle of the 
range of race B, and would show no increase of variability over Fi or 
race B. But the one aberrant individual makes F2 more variable than 
Fi; it is as small as the smallest Cutleri female in skull-length and also 
in femur length (Table 3). It was however not so small in skull- width 
(Table 2). Nevertheless it might pass for a very good size segregate 
closely resembling the Cutleri ancestor. There can be no doubt about 
its hybrid origin or that it was a genuine F2, not an accidental back- 
cross of Fi with Cutleri, for it was of a color variety, cinnamon, which 
could not be obtained from a back-cross or from either uncrossed par- 
ental race. We shall consider its significance further. 
In the second section of Table 1 are classified the skull-length measure- 
ments of Cutleri and race B males and of male hybrids between these 
races, both Fi and F2. These show that the skull-length of males regu- 
larly exceeds that of females of like ancestry but that the relations 
of race to race, as regards skull-length, are the same for males as for 
females. Cutleri and race B do not overlap in range of skull-length in 
either sex. Their Fi hybrids equal or exceed race B in mean skull- 
length, but show no greater variability. F2 is intermediate in skull- 
length between the parental races, lacking the vigor of Fi due to crossing 
but not a matter of heredity. There is however, among the male F2 
hybrids, one aberrant individual with a long skull comparable with that 
of the largest race B animals and Table 3 shows that one of the F2 males 
(the same one in fact) even surpassed race B males in femur length. But 
in Table 2 we look in vain for an aberrant F2 male of unusual skull- width. 
The animal in question accordingly had an unusually long but not an 
unusually wide skull. It also had an unusually long femur. Obviously 
some agency is at work which produces variation in length of skull and 
femur without greatly affecting width of skull. The aberrant F2 female 
as well as the aberrant F2 male indicate this. Whether the hypothetical 
