74 
Variation and Correlation in Brain- Weight 
Now using equations (35) and (36) for males and females respectively we have 
for the probable mean brain-weiglits of arrays having the types of length and 
breadth of the English and Altbayerisch series respectively, the following values : 
6 
? 
Eno-lish B.W. =1501-580 
Altbayerisch B.W. = U94-135 
English B.W. =1289-901 
Altbayerisch B.W. = 1329-676 
In the males the dolichocephalic and brachycephalic arrays have practically 
equal brain-weights, while in the females the brachycephalic array gives a brain- 
weight about 40 grs. higher than the dolichocephalic. But, as has been pointed 
out above, the male equations probably come much nearer representing the normal 
relations than do the female. On the whole, then, there seems to be little or no 
evidence that either dolichocephaly or brachycephaly is associated with large 
brain- weight. 
It should perhaps be noted that no significance is to be attached to the absolute 
values of the brain-weights calculated from the English and Altbayerisch skull 
types, as indicating the probable brain-weight of those races. The differences only 
are of significance. The measurements of these skulls were taken only for the 
sake of convenience ; hypothetical dolichocephalic and brachycephalic skull means 
would have served the purpose as well. The brain-weights obtained are of course 
the probable mean brain-weight of arrays of Bohemians having the specified skull 
characters. Intra-racial regression equations can of course only be used for inter- 
racial 23rediction when the regression of all characters on one another is the same 
in both races, which is a condition never likely to be exactly fulfilled in practice. 
Of the races and characters discussed in this paper the Hessians and Swedes in 
brain-weight, stature, and age come nearest to fulfilling this condition. 
12. The Weight of the Cerebrum. 
Up to this point the discussion has had to do entirely with the variation and 
correlation of the whole brain or encephalon. In the manuscript material placed 
at my disposal by Prof Donaldson (vide supra, p. 15) data were given from 
which it was possible to determine the variation in the weight of the cerebral 
hemispheres, and the correlation of this chai'acter with stature and age for English 
males. The results obtained from this material may now be examined. It should 
be stated that the records of weight were in ounces, and of stature in inches in 
the original material. The units of grouping and the distribution of the frequencies 
may be seen in Tables 45 and 46 of the Appendix. After the constants had been 
calculated in ounces and inches they were transferred to the metric system in 
order to facilitate comparison with the other parts of the paper. The values of 
the constants are exhibited in Table XVIII. 
