K, Pearson 
117 
Random sampling irregularities obscui-e the results, but there is a fundamental 
resemblance between the variations in boys and girls which does not allow of our 
attributing the results wholly to such irregularities. We see that both boys and 
girls start with greater ability in infancy ; their ability then falls between the ages 
8 to 9 — a period possibly when teetli ti'oubles are more marked ; it rises again 
from 10 to 11 in both cases, but only to make a more exaggerated dip from the 
ages 12 to 17 during oncoming puberty. After this the tendency is to steadily 
rise, probably more steeply in men than in women, although the influence of 
oncoming puberty seems more prolonged in boys than girls. Diagrammatically 
both sexes combined give a result of the following kind, where the deviations are 
measured from the mean of each sex (Fig. 3). 
0 12 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 
Fig. 3. Eough Diagrammatic Kepreseutation of Change of Intelligence with Age. 
• Now it must be noted that these variations in intelligence are very slight as 
compared with the total mental outfit of the average individual, perhaps 2 p.c. in 
boys and 4 p.c. in girls. Generally we must conclude that while there are sensible 
slight variations in intelligence with growth, these variations are such that they 
do not affect broad statements based on a consideration of the intelligence-classes 
of children at different ages, i.e. while the physical characters are rapidly altering 
and are so highly correlated with age that it is absolutely necessary to allow for 
this change, the mental chaiacters are far more stationary, the changes which take 
place in them are by no means always in one direction, and are associated rather 
with growth difficulties at various stages than with a uniform development with 
age. Of course in any such considerations as these, we must take, as I have 
endeavoured to do, a scale of intelligence which is not based on a test of 
knowledge or training applied to children of all ages without regard to the length 
of their school career. 
The points discussed in this section are illustrated graphically in Fig. 2. 
The vertical scale is one of intelligence, the horizontal one of age. The upper 
part of the diagram gives the results for girls, the lower for boys. The upper 
approximately vertical broken line shows that the duller girls are on the average 
