The data collected by Boas is biased toward employees of Indian agenci
es and school populations. The data show that intermarried Amerindians
in the last century tended to marry back into the Amerindian populati
on rather than into the general population, as in this century. Distri
butions of blood quantum within tribes show the intensity and duration
of contact with the dominant society and other tribes. Amerindians wh
o married into other tribes tended to choose spouses from contiguous t
ribes, but over one-fourth of spouses represented different language f
amilies. General hypotheses should be tested in the Boas data only wit
h great caution and after correcting for historical factors if possibl
e.