48 
Variation and Coi^relation in Brain-Weight 
males and females in respect to variability in the weight of the brain. In view of 
the general reliability, from a statistical standpoint, which has been previously 
demonstrated to exist in the material on which this work is based, it seems not 
unreasonable to predict that it will be found that equal variability of the sexes 
in brain-weight holds generally. 
Another problem which presents itself in this connection is as to how much 
of the observed difference between males and females in brain-weight is to be 
accounted for by the fact that the male body is in general larger than the female. 
In other words, would a group of females of the same stature and age, say, as a 
given group of males have the same brain-weight ? It is possible to reach a 
general solution of this problem by the use of characteristic equations based on 
the correlations and regressions. I propose now to discuss this question by this 
method. Of course, the prediction may be made by characteristic equations in 
either direction ; i.e., we may predict the probable brain-weight of a group of 
males having other characters the same as iu a given group of females, or we can 
predict the probable brain-weight of a group of females having the characters the 
same as iu a given group of males. In the former case characteristic equations 
based on the male data would be used, in the latter the equations would be those 
deduced from female data. We may consider first the characters, ago, and stature. 
In the following tables are given {a) the probable brain-weights of a group of 
individuals of the sex indicated, as calculated from the characteristic equation 
referred to in the last column of the table ; (6) the observed brain-weights of a 
group of the oppo.site sex having for the other characters the mean values 
used in the characteristic equation in calculating (a) ; (c) the difference between 
(a) and (h); (d) a reference to the equation from which (a) is calculated. 
TABLE X. 
Probable Female Brain- iveir/Jit with Stature and Age equal to Observed Male. 
Eace 
Predicted female 
brain-weight 
Observed male 
brain-weight 
Difference 
Equation 
Hessian (Total) 
„ (Young) 
Swede (Total) 
(Young) 
1287-963 
1308-941 
1307-340 
1319-783 
1391-737 
1405-756 
1400-481 
1415-267 
103-774 
96-815 
. 93-141 
95-484 
25 
26 
21 
22 
The mean difference here is 97'304. The mean observed difference between 
male and female without any allowance for stature and age is 137-8 for the same 
races. So then the differences in stature and age between the males and females in 
our Swedish and Hessian samples account for only 40-5 gr. or about 29 per cent, 
of the observed difference in mean braiu-weight between the sexes. 
Let us now proceed to the reversed prediction, and find the probable brain- 
weights of a group of males having the same stature and age as the means for 
