Eaymond Pearl 
43 
weights of the various viscera ; the weight and vohuiie of the spinal cord ; 
characters of the muscular system, etc. A study of the correlation between any 
or all of these characters and the weight of the brain would be most interesting. 
But, unfortunately, material for such study cannot be had. Brain- weight statistics 
usually provide data for but few other characters ; in the majority of cases only 
sex, age and stature. In two of the series used in this work, viz., the Swedish 
and Hessian, this was the case. For these two groups I have determined all 
possible correlations for both sexes, and for two age groupings ("young" and 
" total "). This gives as the pairs of characters, (a) brain-weight and stature, 
(b) brain- weight and age, (c) stature and age. Bischofif's Bavarian material 
furnished in addition to the above, data on the body-weight. The following 
correlations have been determined for this material : (a) brain-weight and age 
("total" and "young"), (h) brain-weight and stature ("total"), (c) brain-weight 
and body-weight ("total"), (d) body-weight and stature (" total"). In the case of 
the Bohemian material data were furnished from which the correlation between 
skull length and skull breadth could be determined so that for this group I have 
been able to calculate coefficients of correlation for the following pairs of characters : 
(a) brain-weight and stature ("young"), (b) brain-weight and age ("total"), 
(c) brain-weight and skull length ("young"), (d) brain- weight and skull breadth 
("young"), (e) skull length and skull breadth ("young"). In Table IX are 
exhibited the coefficients of correlation together with their probable errors. The 
correlation tables from which the coefficients were calculated are given in the 
appendix to this paper, Tables 1 to 44. References are given in Table IX 
telling for each coefficient the original table from which it was calculated. The 
coefficients of correlation were determined from the usual Bravais formula according 
to which the coefficient of correlation 
Ncr^ao ' 
where x and y are deviations from the means of the two correlated characters 
and cTi and o-, are the respective standard deviations; the usual* method of 
evaluating r from the above equation was used. 
The most striking fact of a general nature shown by this table is the generally 
low degree of correlation which exists between the weight of the brain and other 
characters. The coefficients run noticeably lower even than those of skull 
characters (cf. Macdonell's Table Yf) and very markedly below those between 
different characters of the long bones:|:. On the other hand our brain-weight 
correlations give values of the same general order of magnitude as those found by 
Greenwood§ for vai'ious abdominal and thoracic viscera. Attention may be called 
* Yule: Jour. llni/. Stnt. Hoc. Vol. r.x. Pint iv. )ip. 1 — 44 (Kepriiit). 
t BUiiiictrihi, Vol. III. p. 232. 
X liee and Pearson : Phil. Trans. Vol. 19G, A, pp. 228, 229. 
§ Loc. cit. p. 19 above. 
6—2 
