different bands and then treated collectively. Rather than be guided — or 
misguided — in this selection by my own personal opinion, which would 
naturally be prejudiced against any who did not possess typical features, 
I have preferred to rely upon the decision of the different interpreters, 
even though at times it was not in accord with my own opinion on the 
matter. 
The interpreters were able from their personal acquaintance with the 
Indians and from the knowledge of their family histories to indicate 44 
adult male Chipewyans whom they felt convinced were absolutely pure. 
(From F-d-1, 18; Ch., 11 ; F. and F. S., 12; McM., 3; making a total of 44.) 
Again, 22 were extracted because on their own or on their interpreters’ 
evidence they were Chipewyan-white breeds. (From F-d-1, 3; Ch., 2; 
F. and F.S., 7; McM., 10; making a total of 22.) The 22 may, I think, be 
unhesitatingly accepted as Chipewyan-white breeds; concerning the 44, 
assumed to be pure Chipewyan, it seems probable, now that the report is 
worked up, that the judgments of the interpreters have proved fallible, 
for it seems doubtful if the 44 men they selected as pure, are any more pure 
than the 33 men of the Fond-du-lac band who were induced tio be measured, 
and who were taken at random. 
Abbreviations 
F-d-1 refers to Fond-du-Lac. 
Ch. refers to Chipewyan (Chipewyan reserve). 
F. and F.S. refer to Fitzgerald and Fort Smith. 
McM. refers to McMurray. 
The numbers in brackets in the text are the serial numbers of the 
individuals referred to, and if the appropriate appendix on page 47 et seq. 
be consulted, full details of these individuals will be found. 
In the Frequency Distribution Tables: 
Mean refers to average, 
tr refers to standard deviation. 
E ra refers to probable error of the mean. 
V refers to coefficient of variation. 
No. refers to number of cases examined. 
Diff. refers to the numerical difference between two measure- 
ments. 
P.E.diff. refers to probable error of the difference. 
