JANUARY 12, 1912] 
eradually outlined in the two Americas 
had unquestionably influenced one another, 
but they felt that although all American 
cultures might thus be in some degree in- 
terrelated, they had had no connection 
with any of the cultures of the old world. 
The similarities observed were thought to 
be coincidences due to the effects of similar 
environment and to the fundamental unity 
of the human mind. In short, so far as the 
history of culture was concerned, the new 
and old worlds formed two closed circles, 
tangent but not intersecting, within each 
of which different cultures had modified 
each other, but between which little or no 
interaction had taken place since human 
culture had attained to any real differen- 
tiation. 
As our knowledge of different cultures 
both in America and elsewhere has come 
to be more detailed and their historical 
relations have become more and more ap- 
parent; as the methods of investigation 
have grown more exact and the criteria of 
relationship become more defined, the feel- 
ing has grown that after all, perhaps, the 
similarities between old and new world cul- 
tures might have a deeper meaning. 
I may be permitted perhaps at this point 
to outline very briefly some of the methods 
and what seem to me to be some of the most 
reliable criteria of this so-called historical 
as opposed to the evolutionary school of 
anthropology. Such statements must in- 
deed be trite, but will make clearer per- 
haps my argument further on. In essence 
the method depends upon the realization 
of the complexity of culture, that that of 
any given people is probably made up of 
elements derived from many sources, and 
that analysis must therefore precede a 
elear understanding. The analysis com- 
pleted, the separate elements must be 
‘traced step by step and from tribe to tribe 
to determine their distribution. This 
SCIENCE 47 
necessarily involves much careful consid- 
eration of apparently insignificant details, 
close comparison of archeological material 
with modern forms, the sifting and weigh- 
ing of historical traditions, and the search- 
ing analysis of dialectic and archaic forms 
of speech. If as a result of such study, 
similarities are found to exist between dif- 
ferent tribes: or peoples, this is regarded 
as evidence of a real historic relationship ; 
either one tribe has borrowed directly or 
indirectly from the other, or both have 
felt the influence from a common source. 
But in drawing such conclusions we must 
proceed with caution, and constantly test 
our conclusions by reference to certain gen- 
eral principles. Perhaps the most obvious 
of these is the possibility or probability of 
such relationship on geographic or histor- 
ical grounds. Thus resemblances between 
one tribe in South Africa and another in 
Chile would afford little basis for assuming 
relationship in their cultures, as from their 
very wide separation and known history, 
no reasonable ground exists for assuming 
any possible connection. On the other 
hand, similarities between the people on a 
remote island in eastern Polynesia and 
others in Assam or southern China may be 
significant, as there exists a possible route 
by which cultural influences or even migra- 
tions might have taken place, and there is 
historical evidence of movements of popu- 
lation on a tremendous scale in just this 
direction. 
Another factor of importance is that of 
continuity of distribution. If striking 
similarities are observed between two 
widely separated peoples and no trace ap- 
pears, either among the living peoples in- 
tervening, or from archeological evidence, 
of any connecting links, and there is no 
evidence of migrations, we must be cau- 
tious in assuming any genetic or historical 
relationship, and adopt at least as a work- 
