974 The American Naturalist. [October, 
. these fifty persons, independently writing one hundred words from the 
many thousand with which they are acquainted, all in all, select from 
the same 2,024; 7.e., of the five thousand words written only 2,024 are 
different. Again, 1266 words occur but once in the aggregate lists, 
and omitting these we find that about three thousand of the words are 
formed by the repetition of only 758 words. Passing to an analysis 
of this ** mental community," it becomes clear that it is greatest at 
the beginning of the list, and becomes less towards the end; 7.e., the 
habit is to write first the most common, and when these are exhausted, 
the more unusual words. A very interesting point is the comparison 
of men and women in their tendency to repeat one another's thoughts. 
The evidence is unmistakable that the lists of words drawn up by the 
women are much more like one another than are those written by the 
men. The women use only 1,123 different words, the men 1,376; 
the women write but 520 words that occur but once in the lists, the 
men write 746 such. words. 
(2) A study of the processes involved in these lists bases itself upon 
a careful analysis of the ideas therein represented. The relative sizes 
of such classes, in a measure, indicate the prominence of different 
classes of objects in the minds of the writers. It may be interesting 
to mention that the five best represented classes (of the twenty-five 
adopted in the paper) are ** Names of Animals,” ** Articles of Dress," 
* Proper Names," ‘‘Actions,’’ ‘‘Implements and Utensils." The 
sexes present characteristic preferences for the various classes. The 
women contribute most largely to “Articles of Dress," writing 224 
such words, while the men write but 129; they show an equal favor- 
itism for ** Articles of Food," writing 179 such words to but 53 for the 
men. The men, on the other hand, show fondness for ** Implements 
and Utensils," ** Names of Animals," ** Professions," ‘ Abstract 
Terms,’’ etc. 
Of the various links by which the one word suggests its successor, it 
may suffice to indicate as prominent types, (@) association by sound, 
in which words are rhymed, or begin with the same letter; (4) by be- 
longing to the same class, as when a series of animals or articles of 
dress is formed ; and (c) by more general but not briefly describable 
relations. One may combine the two inquiries (1) and (2) to ask 
how often the same word is associated with the same word in different 
lists. If we take the twenty words most frequently occurring, we find 
over 500 mentions ; and if we examine in each case the word 
preceding the given word we find it to be the same in 111 cases, and 
the succeeding word the same in 145 cases,—certainly a remarkable 
