iyi SISSANO. 
the same whichever may win. ‘These are battles to extinction of the 
fighting men; the women go with the victory. If we may judge the 
past from the present, the Polynesian raider is of the better courage; 
he knows better his trade of fighting; the Melanesian advantage lies in 
numbers. If the Polynesian wins he takes the Melanesian women 
to his community; if the Melanesian wins the Polynesian women of 
the fleet remain; in either case the next generation presents a mixed 
blood and a mixed speech and it matters little in which order the mix- 
ture is produced. In case of such an onfall where the result has been 
a Melanesian victory, the cross of alien blood tends progressively to 
wane under the dilution of this mixture by successive matings with 
pure Melanesian strains. In case of Polynesian victory the victors 
may sail away and leave the cross behind to undergo the same absorp- 
tion, or the progeny of such mixture may be carried along with the 
Polynesian sires to undergo gradual absorption in that strain. In some 
cases we may see the probability of a settlement of victorious Poly- 
nesians upon land won from Melanesians, in which case the cross- 
mixture would tend to fairly equal replenishment from each source, 
and we should discover a race of hybrids with contamination of cus- 
tom and speech; and this is just what we do find at several points in 
the Melanesian chain. 
In the third class of such settlement along the route of migration 
we find an operative factor which tends to produce the permanent 
occupation. This is the yearly recurrence of the hurricane season, 
the months between December and March, when the trade wind is 
deranged. In this period of almost incessant rains, and rain is very 
uncomfortable to the bare islander, the wind either fails entirely in 
dull days of dreary calm or else reaches such violence of gale that only 
the stoutest ships may live afloat, and even ashore solidly constructed 
buildings plane through the air. When this season meets the canoe 
voyager a sojourn ashore is made necessary, no matter where he be. 
This factor, then, conditions the other reasons for such a halt. 
C. 
The third of these possibilities is that Melanesian communities 
which have come into possession at their home site of a given amount 
of Polynesian material, somatic or cultural or linguistic, may themselves 
migrate to communities thitherto untouched by the general Polynesian 
movement. Here we encounter strong probability of such Melanesian 
migration, but we lack definite detail. At the time when the islands 
of the western Pacific were densely populated there was lack of interest 
in preserving the record of their tradition history. Now, when interest 
has been attracted to the importance of this study, the islands have 
been depopulated by the infamy of the labor trade and the memory 
of the past has vanished. It is possible to feel the existence of at 
