ANTHROPOLOGY: R. H. LOWIE 
349 
the father's sister's soti's daughter, since with exogamy and maternal 
descent she cannot belong to the father's clan. And actually we find that 
among the Hidatsa and Crow this relative is no longer classed with the 
father's sister but with the sister, this latter relationship following 
from the fact that her father is classed with my father. But between 
the status of a father's sister's daughter's daughter and a father's sister's 
son's daughter there is no ascertainable difference except that of clan 
affiliation. Therefore the terminological classification is a function of 
the exogamous group, which was to be proved. 
The same conclusion may be established by ehminating the hypotheti- 
cal factor in another way. The Siouan family embraces a number of 
tribes with patrilineal descent, of which the Omaha are the best known. 
In such tribes the father's sister's descendants are no longer, as among 
the Crow or Hidatsa, members of the father's exogamous group; and we 
find, as a matter of fact, that her son and daughter are classed not with 
her but with the sister's son and daughter. With paternal descent my 
father's sister is my group sister, and while the Omaha have a distinct 
term for the father's sister it seems that in some ways she is still regarded 
as a sister — both as regards her children and as regards her husband, 
who is classed with the brother-in-law. On the other hand, the mother's 
brother's son, mother's brother's son's son, and so forth, are all members 
of the same exogamous group if there is paternal descent, and the Omaha 
actually designate them by a single term. And again, as soon as we pass 
out of the exogamous group, the terminology varies: my mother's 
brother's daughter's son is my brother, not my mother's brother, since 
he no longer belongs to my mother's gens but is related to me solely 
through his mother, who is my ' mother' because she does belong to my 
mother's gens. 
In short, passing from tribes with matrilineal to tribes with patri- 
lineal descent within the Siouan stock, we find precisely those differ- 
ences that logically follow from the assumption that the exogamous 
group lies at the basis of kinship classification; and passing within a 
particular tribe from relatives within the same group to relatives of other- 
wise similar status outside the group we at once find a difference in 
nomenclature. Hence the exogamous factor rhust have been a real 
cause in moulding the kinship terminology of at least some so-called 
classificatory systems. 
