Kaymond Pearl 
61 
Diagram XVI. Probable Brain-weight for given Stature. 
Hessian s Young. 
) 
J 
C 
) 
) ( 
) 
( 
} <■ 
) 
( 
) 
( 
) 
c 
) 
1 
5 
( 
> 
139-5 142-5 145 5 148-5 151 5 154 5 1575 1G0 5 163 5 166 5 169-5 172-5 175-5 178-5 181 5 184-5 
Stature { — s) in centimetres. 
Diagram XVII. Probable Brain-weight for given Stature. 
Hessian $ Young. 
127 5 130 5 133 5 136 
139-5 142 5 145 5 148 5 151 5 154-5 157 5 160 ! 
Stature ( — s) in centimetres. 
103 5 166-5 169 5 172 5 175 5 178 5 1B15 
It is evident from these diagrams that a straight line represents the regression 
relation between brain-weight and stature better than any simple curve would. 
So that we are justified in concluding, as in the case of age, that until we have 
very much larger collections of data than are at present available we can do no 
better than proceed on the assumption that the regression of brain-weight on 
stature is linear. In fact, in these samples there is no evidence that there 
is any tendency towards anything but a linear relation between the two 
variables. The deviations of the means of the arrays from the regression line 
are only such as would reasonably be expected to arise when we deal with 
relatively small samples. In connection with all of the regression diagrams in 
this paper it should be kept in mind that the outlying points are based on single 
individuals, or at most only on comparatively few individuals. Hence deviations 
of these outlying pouits from the regression line have very little or no significance 
as regards the general trend of the results. In fact it would really be better if 
