972 THE AMERICAN NATURALIST. [VoL. XXXIV. 
B. Distorted figures, caricatures, scare-faces. 
4. Grimaces to arouse mirth or terror; dance or war masks. 
C. Animal masks. 
D. Head-pieces. 
* As may be seen from this arrangement, it deals almost exclusively 
with the forms from whose nature only occasional conclusions as to 
their significance can be drawn. Such is an ethnographical view 
that would be very well in place where geographical relations had 
been already worked out and where only a collection of the forms 
of a place, district, or province had to be made. So such tables as 
Ratzel's are only desirable when convenient to a treating of masks 
like that of Haddon on those of southeastern New Guinea. For 
more esoteric work these systems have little value. The method 
that Dall has used might be considered, and in fact would not be at 
all bad, if one could convince himself of the truth of his premise, 
viz., that masks arose from weapons of protection." 
Dall’s searching classification is then given; it is allowed by the 
author to be clever, but condemned by the false premise just quoted. 
Nevertheless, he admits that Dall has shown that the “relations of 
form and significance, while various, are yet close." The introduc- 
tion continues : * Every experienced man must have observed from 
the ethnographic part (of the book) that the African masks compose 
no simple structure, and that custom and belief are continually 
influencing the form, changing it and developing it. 
* Masks are bound up with particular ceremonies, — they always 
appear at funeral feasts ; where the fraternities begin to disintegrate, 
masks gain in variety (Cameroon, Loango, Yoruba). 
“ The classes of head and body masks stand forth clear and 
distinct : here they are combined, here they are divided. We see 
derived forms on every hand, as in the bushy circumcision dresses. 
“In short, African masks, after examination, attain the appear- 
ance of a composite whole, a complicated development, whose 
several parts are to be traced back to various sources. So it is 
worth while to note our old point : the form corresponds to the con- 
tent. Plastic expression, custom, are the forms which owe their 
existence to the content, to beliefs. 
* If we now observe these beliefs and views in origin and growth 
from their roots up, we shall readily derive the understanding for 
the various points in our field of study. . . ." 
Frobenius then proceeds to derive nearly all African masks and 
customs from the cult of the dead. Even tree worship arises from 
