XII. c. 4 Brown, Merrill and Yates: Volcano Island 189 
The very open nature of the stand of grass on most parts of 
the island is certainly not due to competition and apparently 
not to a lack of seed. Saccharum, hnperata, and Miscanthus 
have adaptations for the distribution of the seed by wind, and 
the first two certainly produced large quantities of seed on the 
island as early as three years ago. 
The failure of grass to produce dense stands even in most places 
where it occurred as scattered patches in 1914 would indicate 
unfavorable environmental conditions. This view is supported 
by the reduced height of the clumps of Saccharum. 
TREES 
Trees are very scarce except in limited areas near the northern 
coast and at the extreme tip of the peninsula that extends from 
the southwestern corner. Even in these localities grasses are 
very much more prominent than trees Elsewhere the trees 
occur only as very widely separated individuals or as clumps 
of two or three individuals. Plate XI, fig. 1, is from a photo- 
graph taken at the northwestern end of Mount Balantoc and 
shows an exceptional development of trees. 
The only area in which trees predominate is on the northern 
slopes of Mount Pirapiraso, at the northwestern corner of the 
island, where the second-growth forest reaches its greatest de- 
velopment and covers more than half of the area, the remaining 
space being largely occupied by grass. 
The specific composition of the tree flora of Volcano Island 
is extremely varied when the small number of the individuals 
is considered. 
The most abundant species is Acacia farnesiana (aroma), 
which is scattered over the whole island except on the main cone 
and Mount Tabaro. This plant is very common on the mainland 
around Lake Bombon, and in the Philippines generally it is 
nrominent in the early stages of invasion of grasslands by trees. 
Its success in the latter situation is due to its ability to re- 
generate after the aerial portions of the plant have been killed 
by Are. In view of this habit and its present prominence on the 
island it seems not unlikely that a considerable number of plants 
of this species may have escaped destruction during the eruption 
of 1911. On the other hand the seeds of aroma are inclosed in 
a woody pod which floats so that it may readily have been washed 
ashore ; in which case, however, we must account for its distribu- 
tion in places distant from the shore. The method by which 
