346 
ANTHROPOLOGY: R. H. LOWIE 
such marked divergence from the restricted method of fruit and seed 
production peculiar to other varieties of this species and to all other 
species of tobacco as far as known. 
The experiments reported upon above have, in part, been made pos- 
sible by an allotment from that portion of the Adams Fund of the 
United States Department of Agriculture granted to the Agricultural 
Experiment Station of the University of California. A more detailed 
report will appear in the University of California Publications in Botany, 
Volume 5. 
1 Thomas, MendelJ., 1 (1909). 
2 Howard, G. L. C. Mem. Dep't. Agr. India (Bot. Ser.), 1913. 
3 Hartley, U. S. Dept. Agric, Bur. Plant Ind., Bull., No. 20. 
4 East, Pop. Sci. Mon., 1910. 
^Wellington, Amer. Nat., vol. 47, No. 557. 
® Bateson, 4th Co?if. Inter. Gen. 
7 SetcheU, Univ. Cal. Pub. Bot., vol. 5, No. 1. 
8 Winkler, Prog. rei. Bot., Bd. 2, H. 3. 
9 de Vries, Bot. Gaz., Chicago, 59, 190. 
EXOGAMY AND THE CLASSIFICATORY SYSTEM OF 
RELATIONSHIP 
By Robert H. Lowie 
AMERICAN MUSEUM OF NATURAL HISTORY. NEW YORK CITY 
Presented to the Academy, April 24, 1915 
Lewis H. Morgan, in his Systems of Consanguinity and Affinity (Wash- 
ington, 1871), estabHshed the fact that in a large part of North America, 
in India, in Africa, and in Oceania the natives use terms of relationship 
that designate not individuals but groups of individuals, and accord- 
ingly he labeled these systems as 'classificatory.' Later E. B. Tylor 
and others advanced the view that the classificatory system and exog- 
amy — the rule that a person must marry outside of his own social 
group (clan or gens) — were merely two aspects of a single institution: 
that, in other words, primitive man classed together individuals belong- 
ing to the same exogamous division and separated individuals of differ- 
ent divisions. Quite recently this view has been advocated by W. H. 
R. Rivers. In his Kinship and Social Organisation (London, 1914) he 
correlates the classificatory system with exogamy, our own system with 
the family in the narrower sense of the term, and the descriptive system 
of, say, the Nilotic Negroes (in which a few primary terms designate the 
basic relationships and serve by their combination to describe all other 
relatives) with the patriarchal or extended family. The correlation of 
