16 
Variation and Correlation in Brain-Weight 
2. Data. 
The statistical data for this study were obtained from several sources, as has 
been mentioned above. The work was begun on Marchand's statistics, then 
extended to include the old, but still valuable collection of brain-weights by Bischoff. 
With the results from these two series in hand, I was still not satisfied that I had 
analyzed sufiicient material to be certain of the conclusions. So I determined to 
include the following material in the discussion ; (1) the admirable series of brain- 
weights of Swedes collected by Retzius. (2) Matiegka's Bohemian series. (3) The 
Boyd-Marshall data observed at the St Marylebone Infirmary. As will appear 
later the last mentioned series was used only for a special purpose. The general 
results of this paper are based on the analysis of the Marchand, Bischoff, Retzius 
and Matiegka series. 
Marchand's* data were obtained from a series of brain- weighings made between 
the years 1885-1900 in the Pathological Institute at Marburg. The total number 
of brains weighed was 1234. Of this number 389 were from individuals under 
15 years of age, and hence were unavailable for adult brain-weight studies. The 
data recorded from each individual in Marchand's work included, in addition to 
brain-weight, sex, age, and stature. It was not possible, however, to determine all 
of these points in every case, so that omissions of one or more records for an 
individual occur not infrequently in these statistics. The same is of course true 
of all extensive series of brain-weight data ever collected. After all such incomplete 
records have been thrown out there remain in Marchand's tables 475 male and 
281 female complete records, or a total of 756 of both sexes. Marchand did not 
include body-weight in his determination, because he considered that : "das 
Kcirpergewicht bei dem Leichenmaterial der Krankenhauser ein so wechselnder 
Faktor ist, dass bestimmte Beziehungen zum Gehirngewichte sich daraus kaum 
ableiten lassen wllrden." 
The brains were weighed in the fresh condition, usually immediately after 
removal from the cranial cavity. The membranes were not removed before 
weighing. The weighings were made in the majority of cases to the nearest 
5 grams, greater accuracy not being possible on account of such unavoidable 
sources of error as varying amounts of blood in the vessels of the brain, amount of 
fluid in the ventricles and membranes, etc. These sources of error cause variations 
in the total weight outside the limits of 5 grams. Marchand's material comes 
from the " hessische Bevolkerung." 
Bischoff's-f data comprise the results of the weighing of something over 
800 brains of individuals between the ages of 17 and 85. The attempt was to 
determine for each individual the following characters in addition to brain-weight : 
sex, age, cause of death, stature and body-weight. Omissions in the records were 
* Marchand, F.: " Ueber das Hirngewicht des Menschen," Abhandl. d. math.-phy^. CI. d. Konigl. 
Sachs. GcscUsch. d. T(7ss. Bd. xxvii. No. iv. pp. 393—482, 1902. 
t Bisclioff, T. L. W. V. : Das Ilirncicwicht dcs Menschen, Bonn, 1880, Pp. vi. and 171, plus Tables. 
