376 Pigmentation, Selection and Anthropometric Characters 
with more suitable means, some not very marked connection between pigmentation 
and height may ultimately be found. 
Birmingham Data. Pigmentation and Weight. The Birmingham data also 
give the weight of the children, and therefore afford an opportunity of investigating 
the relation between pigmentation and weight. As has been explained, this is a 
problem separate from that of the relation between pigmentation and height, and 
there might possibly be a correlation between pigmentation and weight although 
there were none between pigmentation and height. The bair and eye colour 
method of classification has been described ; the weight was measured in kilogs. 
TABLE XIV. (See Tables XXI and XXII, pp. 383—384.) 
Hair Colour and Weight. Birmingham. 
Mean Weight in Kilogs. 
Hair Colour 
Boys. Age 13 
Girls. Age 13 
Red 
Fair 
Light Brown 
Medium Brown 
Dark Brown 
Black 
33- 69+ -38 
34- 28+ -21 
34-40+ -14 
34- 15+-12 
35- 08+ -15 
34-95 ± -57 
36-22+ -43 
35-71+ -26 
35- 59+ -19 
36- 28+ -15 
36-62+ -15 
35 -92 ±-49 
General Mean ... 
34 -39 ±-04 
36-22 + -04 
The data have been treated in exactly the same way as in the previous section 
of the paper. Table XIV gives the mean weight for the total number of both boys 
and girls and their probable errors ; the means and probable errors are also given 
for each hair colour group. Again we notice no striking deviation from the mean ; 
but certain groups differ from the remainder enough to require special attention. 
Figure III is constructed in the same manner as before in the case of the 
pigmentation and height data. With regard to the boys, the only group which 
differs in any remarkable way is the dark brown group. The deviation is not very 
large, and would be covered by four times the probable error. It is interesting to 
find that among the girls the same deviation is apparent, though to a much smaller 
extent. In the latter case the mean of the group only falls just above the line 
representing twice the probable error. These deviations are similar to, but not 
quite so marked as, those which we have observed in the pigmentation and height 
table for the dark brown group. The coincidence is certainly interesting and 
suggestive, though it cannot fairly be called more. The coincidence is, however, 
further increased by the fact shown in Figure III B that the light brown group have 
a lower mean weight than the mean of the whole. It is impossible to estimate 
the true value, if there be any, of these results ; further research alone can show 
whether there is anything in these possible connections between pigmentation and 
