INDONESIAN ANNOTATIONS ON THE VOCABULARY. 137 
starch-cells induces a much greater flow of saliva. Within the mouth 
in the molar grinding of the root the starch-cells are not only separated 
from the intractable woody fiber, but are themselves bruised open, 
with the result that the saliva may reach the starch itself. We may 
conclude that when the starch is brought freely into contact with the 
saliva, and this for a much longer period than is utilized in the mastica- 
tion for the stomach, the process of ptyalization causes normal changes 
in the starch which can not be without their effect in the maceration 
which follows. ‘That the method in Polynesia was uniformly that of 
chewing is substantiated by all the discovery record. We have further 
evidence of the great antiquity of the method, for in Samoa we find 
the word ‘aumaga employed to designate the young men in their cere- 
monial position in society, and this word is merely ‘the chewers.”’ 
I assume the validity of the statement that in the interior of Viti Levu 
the original practice was to pound the root, but I have never seen it in 
the country of the Kai Colo. On the Fijian littoral, even of Viti 
Levu, pounding was distinctly understood to be a concession to foreign 
prejudices; the introduction of chewing from Tonga, if such were 
indeed the case, could hardly have reached the mountaineers. 
In connection with the mixture of the two cultures on the Polynesian 
island of Tikopia I should make a note on the Verge Islands to the 
north (The Polynesian Wanderings, 4) even as far as Kapingamarangi 
north of the equator. While the inhabitants are almost pure Poly- 
nesians and might be expected to have the kava custom, we lack 
definite evidence in the scanty memoranda of their speech which has 
been published. I regret that in the loss of my field-notes I am unable 
to report positively upon the use of kava in that chain of islands. It 
is for that reason that in distributing upon my culture chart the bound- 
aries of the kava and the betel I have left the Verge Islands out of 
the scheme. 
To the kava people Dr. Rivers (p. 125) assigns the secret societies 
of the Banks Group and, by extension throughout the area of matri- 
lineal culture, of the Fijian nanga. Before this determination can 
meet with whole acceptance it will be necessary to consider the Poly- 
nesian culture in this respect, for it is not to be denied that the Poly- 
nesians represent the highest attainment of that culture group. With 
the exception of the arioz of Tahiti we have no information as to the 
existence of any fixed body within the community which might be 
regarded as a society in the restricted and technical sense here involved, 
and even in the case of these joyous pagans we find naught of ritual 
and arcana which might suggest the society structure. Such being 
the case, it is difficult to accept this assignment of the societies to the 
kava culture, for it is inconceivable that a ritual observance should 
have persisted when the whole purpose of the ritual has vanished so 
completely as not to leave a trace. 
