J. Blakeman 
125 
(c) the exclusion of all males under 24 years of age and all females under 
20 years of age. The reason for adopting this restriction was that from 
considerable experience of anthropometric measurements we have found that 
these ages roughly mark in man and woman a turning-point in the nature of the 
growth curves. Up to these ages there is a rather rapid increase, after these ages 
in most characters a slight but continuous decrease. 
With the above exclusions we had 117 male and 94 female cases. These 
numbers are small, but the material is fairl}^ homogeneous and the probable errors 
have in each case been given. The publication of the actual data (pp. 118 — 123) 
has been undertaken largely with the hope that other medical schools may collect 
similar material so that ultimately proper numbei's will be available, large enough 
for due allowance to be made for the cause of death and the condition at death. 
No such separate classifications are possible at present. 
(2) On the Special Character of tlie Present Material. 
The present material is what has been described in this journal* as a "General 
Hospital Population," — with perhaps in this case a rather large proportion of cancer 
cases. It cannot be too often insisted upon that such a population is not a fair 
sample of the "general population" of a given district. There is a larger amount 
of what it would be convenient to call "shrinkage" due to illness and defective 
nourishment. The well-nourished, physically well-developed middle classes are 
largely absent, and taking the brain-weights of these hospital returns as typical 
of the " general population " — especially applying them to problems in the 
relationship of intelligence to physical measurements — is liable to lead to very 
erroneous conclusions. An illustration of this may be given from Gladstone's 
classification into acute and chronic cases — the A and C of the second column of 
the tables. We find for average brain-weights : 
TABLE I. 
Ages 
Cause of Death 
No. 
Male 
No. 
Female 
20 to JtG ... 
Acute 
16 
1430 
5 
1303 
Chronic 
27 
1331 
32 
1227 
Over JfG ... 
Acute 
22 
1318 
3 
1248 
Chronic 
34 
1354 
.40 
1178 
It will thus be seen that, with the exception of the males over 46, the cause 
of death makes a very substantial difference in the average brain-weight. The 
numbers are however far too few to allow us to make any separate classification, 
but the point is a vital one and must undoubtedly be properly dealt with when 
greater numbers are forthcoming. The reader will find as he proceeds further that 
we have other grounds for suspecting that the general hospital population diff'ers 
in some essential features from the general population. 
* Greenwood : Biometrika, Vol. iii. p. 65. 
