138 MISS L. S. GIBBS : A CONTRIBUTION TO 
With regard to cultivation, to cite Fiji alone, which comprises much the 
largest area, Seemann (3, p. 216) writes that there was no virgin forest to speak 
of in 1862, and certainly anyone who has travelled through Viti Levu will 
corroborate the justness of this observation. Every year fresh patches of 
forest are eut down by the natives for the cultivation of their crops, and 
these are worked for a year or two and then left to go wild again. These 
patches may be seen on the highest ridges up to 3500 ft., and if they are pro- 
nounced now, must have been more so when the island carried a much larger 
population depending on home supplies. The tribes were divided, as in all 
the large Pacifie islands, into mountain and eoast tribes, invariably at deadly 
enmity with each other. The coast tribes were then the most numerous, and 
still line all the islands with an uninterrupted fringe of villages, nestling 
behind the belt of strand trees. They were in constant communication with 
the other islands, and disposed of the fertile littoral areas for the cultivation 
of their crops. Space being limited, that cultivation was necessarily more 
continuous and not so sporadic as in the mountains, whilst climate and con- 
ditions were more favourable to the cultivation and exchange of all the 
natural products showered so lavishly by nature. The mountain tribes, on 
the other hand, were restricted to the mountain districts of one island. 
Many tropical fruits do not flourish at high altitudes, so, eut off from the 
abundant coast-supplies, river-fish, land-shells and many forms of inverte- 
brates, wild yams and wild fruits would form their chief supplies in a country 
lacking all quadrupeds, and consequently animal food. In times of scarcity 
they would be reduced to the wild yams, to obtain which large areas of 
country are still burnt off. Trees producing any form of edible fruit would 
be carefully planted as new “ towns" were established and fresh land brought 
under cultivation, and these species would naturally spring up where old 
cultivation had been. 
The South Sea Islander is also a born gardener. All trees and plants are 
known by name and their utility or beauty appreciated. Their villages are 
a blaze of colour from the many-hued foliage plants grown round their 
houses, and magnificent specimen trees shade the levelled grass areas on 
which their “towns” are built. Favourite species for beauty or use will be 
planted along their roads or tracks, and the wandering native will preserve 
seed for future planting of any particular species that may strike his fancy. 
Given these conditions, which have prevailed for centuries throughout the 
Pacific, it will be seen, as Seemann says, how hopeless it is to construct 
theories of distribution or to expect a pronounced endemic element where 
natural conditions have never prevailed in recent times. Where man has 
cleared the dense forest growth, there the wind’s agency is well seen in the 
secondary протом. On a patch over 3000 ft. alt., I found Spireanthenuum 
vitiense and S. Graeffei in full flower, species apparently not represented 
among the surrounding trees, and. not noted elsewhere. In a neighbouring 
