J. F. Tocher 
299 
under whose care they were. These persons were suffering from some congenital 
defect such as idiocy, or were rickety, syphilitic, or tuberculous, in such manner 
as directly to affect their anthropometric characters*. There were accordingly 
left 4381 males and 3925 females to represent the general lunatic population. 
Medical experts would no doubt agree that others might be excluded if a thorough 
knowledge of their history were available. It is therefore highly probable that 
an undetermined residue of exceptional cases remains. An elaborate investigation 
would, however, be required to reveal these cases, and as such an investigation 
was, under the circumstances, out of the question, and would affect the results 
of the present enquiry only in a very slight degree, the 4381 males and 3925 
females are taken to represent substantially what may be termed the ordinary, 
normal asylum, or general insane population — i.e. those mentally affected, exclusive 
of the specific cases just mentioned. In view of the results of recent investi- 
gations by Pearl -|- and Btakeman| establishing a direct connection between age 
and certain physical characters, an analysis of the data in age groups would 
have been useful, and would have furnished valuable additional information in 
the comparative study of the inmates of individual asylums. Since the age range 
in the asylums is a pretty wide one, there is little doubt that our information 
as to the physical characters of the inmates would have been more complete 
had an age analysis been made. In recording the measurements at the asylums, 
however, no note was made, at the time, of the age of the inmates, and it was 
only when the statistical analysis was being carried out that the importance of 
separation in age groups was fully recognised. It was then found that consider- 
able additional expense would have been incurred in furnishing an accurate 
statement of the ages of those observed, and any treatment of the data with 
respect to age groups was therefore abandoned. Since, however, none but adults 
are included in the analysis, any conclusions reached are those based on an 
adult population. 
As explained in the Supplement§ and in the Henderson Trust Report ||, 
observations were made on a selection of both measurable and non-measurable 
characters of inmates. The measurable characters observed and recorded were 
those of stature (S), head length (L), head breadth (B), and head height (H); 
the non-measurable characters were those of hair colour, eye colour and nose 
contour. Head length was measured from the most prominent point of the 
glabella to the occipital point. L is therefore maximum head length. The head 
breadth measured was the maximum breadth above the level of the ear. Head 
height was taken from the mid points of the auricular passages to the vertex ; in 
some respects, as will be seen from the analysis, this is a somewhat indefinite 
measurement. The hair categories were red (jB), fair (F), medium (M), and 
dark (D). Red included light, bright and dark red ; fair consisted of white, 
* All cases of idiocy when recognised ab initio were excluded, or if measured were afterwards 
excluded under this head. 
t Pearl : Biometrika, Vol. iv. pp. 13 — 104. J Blakeman : Biometrika, Vol. iv. pp. 124 — 160. 
§ Biometrika, Vol. v. Suppl., p. 3. il Henderson Trust Report, Vol. i. p. 14. 
