144 Biometric Constants of English Brain-weights 
If we pass from intra-racial to inter-racial comparisons, Pearl's memoir offers 
valuable tnatei ial. Thus he has calculated the correlation of w with L and B for 
Bohemians, based, it is true, on Matiegka's rather doubtful skull measurements. 
Again, he provides us with the brain-weight and stature and brain-weight and age 
correlations for Swedes and Hessians. These are reproduced in the accompanying 
table. Considering how widely correlation results differ from race to race the 
general agreement is not without significance. It increases our confidence in 
the possibility of good results being obtained from careful post-mortem room 
measurements. 
TABLE XVIII. 
Comparison of English and other European Brain-weight Coi'relations. 
Correlation 
Middlesex Hospital Data 
Other European Data 
6 
? 
Group 
6 
? 
w and L 
10 and B 
VJ and <S^ 
11 11 
w and ^1 
)) )) 
•598 ± -040 
•489+ •048 
•289 ±-057 
- -256 ± •058 
11 11 
•545 + -049 
•607 + ^044 
•367 ± -060 
-•372±^060 
11 11 
Bohemian 
Bohemian 
Swedes 
Hessians 
Swedes 
Hessians 
•548 + -027 
•466 + ^031 
•183 + ^032 
•182 + -030 
- ^249 + ^031 
-•167 + -030 
•360+^047 
•504 +^040 
•349 + ^039 
•183 + •0.39 
-■234+^042 
-•360 + ^035 
The general agreement is, we have said, not bad considering the differences in 
method of measuring and probably in heterogeneity of material. The Swedish are 
quite fairly concordant with the English. So are the Bohemians (considering the 
poor method used by Matiegka for determining the breadth and length of the 
skull), except for the correlation of weight and length for women, where there is a 
bad and inexplicable fall, altliough a slight fall might have been expected from 
the English results. The Hessian values appear to be irregular, and one is inclined 
to doubt the accuracy with which the corpse length has been determined, or the 
relative homogeneity of the series ; individuals under 20 ought not to have been 
included in the age and stature correlations. 
(7) On the Regression Equations of Brain-iveight on other Characters. 
(a) The following table will allow the reader to see at a glance the average 
changes in brain-weight m and product P involved in unit change from the mean 
in each of the observed characters. 
As illustration, take the following case, which is closely that of Jeremy 
Bentham : 
Stature: Mean (?), f/= 560, X = 192, 5= 153, i7* = 127, and Age = 85 yrs. 
* Measured from top of auricular orifice H is 121 (see Biometrika, Vol. ni. p. 394), we must add, 
say, 6 mm. to compare with Gladstone's data from centre of orifice. 
