ANTHROPOLOGY: R. H. LOWIE 
347 
the classificatory nomenclature with exogamy now requires empirical 
verification in the several areas of the globe, and the following is an 
attempt to make this test for North America. 
Before undertaking this inquiry, however, the concept * classificatory' 
must be supplanted by one that more adequately represents the phenome- 
non under discussion. For any particular 'classificatory' system is not 
moulded by a single factor but by a series of factors, and these are de- 
veloped in varying degree in different systems. Hence the test must 
be applied to that common element which, consciously or unconsciously, 
differentiated the primitive terminologies in question from those current 
among ourselves in the minds of investigators. An examination of 
Morgan's earliest expressions on the subject indicates that it was the 
merging of lineal and collateral relatives — the use of a single term, e.g., 
for mother and mother's sister, for father and father's brother — that 
impressed this pioneer investigator, and this is the feature that actually 
characterizes the classificatory systems of all the regions of the globe. 
Our query is thus reduced to this form: Is the confusion of collateral 
with lineal relatives a function of exogamy? 
The first question is, of course, how the exogamous tribes compare 
with those 'loosely organized,' i.e., those lacking exogamous divisions. 
One of the principal exogamous areas of North America is found in the 
United States east of the Mississippi. Practically throughout this im- 
mense territory the custom of exogamy is associated with a terminology 
that fails to distinguish collateral and lineal relatives. Among the 
Northwest Coast tribes of Canada the same association holds, and this 
applies likewise to those of the Plains tribes that possess a clan or gentile 
organization. The one doubtful exogamous region is the Southwest, 
for which we have practically no data except from the Tewa, where the 
correlation does not hold. An inquiry into the as yet unknown systems 
of the Keresan, Hopi, and Zuni pueblos is of the highest theoretical 
importance. When we turn to the loosely organized tribes we meet 
again with one exceptional region, that of the Mackenzie River, and sev- 
eral sporadic cases outside, where non-exogamous tribes are reported to 
possess a kinship terminology that is ex hypothesi to be expected together 
with exogamy. On the whole, however, the agreement with the Tylor- 
Rivers theory is highly satisfactory. The Eskimo, the Plateau Indians, 
the Calif ornian tribes are loosely organized; and all of them tend to em- 
phasize the distinction of such relatives as father and paternal uncle, 
mother and maternal aunt. It is important to note that these terminolo- 
gies by no means resemble those of European languages. Among the 
Shoshonean tribes and the Kootenai, for example, relatives distinguished 
