ew specific instances. In several islands of the Pacific, some of which
have had European settlers on them for more than a century, a most
important position in the community is occupied by the father's sister.
If any native of these islands were asked who is the most important
person in the determination of his life-history, he would answer, "My
father's sister"; and yet the place of this relative in the social
structure has remained absolutely unrecorded, and, I believe, absolutely
unknown, to the European settlers in those islands. Again, Europeans
have settled in Fiji for more than a century, and yet it is only during
this summer that I have heard from Mr. A. M. Hocart, who is working
there at present, that there is the clearest evidence of what is known
as the dual organization of society as a working social institution at
the present time. How unobtrusive such a fundamental fact of social
structure may be comes home to me in this case very strongly, for it
wholly eluded my own observation during a visit three years ago.

Lastly, the most striking example of the permanence of social structure
which I have met is in the Hawaiian Islands. There the original native
culture is reduced to the merest wreckage. So far as material objects
are concerned, the people are like ourselves; the old religion has gone,
though there probably still persists some of the ancient magic. The
people themselves have so dwindled in number, and the political
conditions are so altered, that the social structure has also
necessarily been greatly modified, and yet I was able to ascertain that
one of its elements, an element which I believe to form the deepest
layer of the foundation, the very bedrock of social structure, the
system of relationship, is still in use unchanged. I was able to obtain
a full account of the system as actually used at the present time, and
found it to be exactly the same as that recorded forty years ago by
Morgan and Hyde, and I obtained evidence that the system is still deeply
interwoven with the intimate mental life of the people.

If, then, social structure has this fundamental and deeply seated
character, if it is the least easily changed, and only changed as the
result either of actual blending of peoples or of the most profound
political changes, the obvious inference is that it is with social
structure that we must begin the attempt to analyze culture and to
ascertain how far community of culture is due to the b