XII, C, 4 Brown, Merrill and Yates: Volcano Island 209 
shows that this water does not contain an unusual amount of 
dissolved material; it has a total solid content obtained by evapo- 
ration amounting to 370 parts per million. The water of the 
lake contains a very much higher percentage of dissolved mate- 
rial as will be seen from the analysis in Table III. Near Ambu- 
long the total solids amount to 1,220 parts per million and near 
Voleano Island to 1,540 parts per million. The large amount 
of dissolved material in Lake Bombon is probably derived from 
volcanic ejecta, and a considerable proportion of it may have 
come from Volcano Island. 
The analysis (Table III) by Cox,?* of water flowing into the 
crater lake in 1911 shows an unusual high content of dissolved 
mineral matter. A comparison of this analysis with the water 
in the crater lake in 1917 shows that the lake contains a much 
higher percentage of dissolved material than the water of the 
stream flowing into it in 1911. The amount of chlorides is 
about three times as great, while sodium shows an even greater 
increase in concentration. Calcium sulphate forms layers of con- 
siderable extent over the soil at the edge of the lake. 
The amount of solid material in the water of the lake is very 
high, there being 40,000 parts per million. This high solid con- 
tent shows that a great deal of soluble material has been taken 
out of the soil of Volcano Island, as the crater lake is of consider- 
able size and depth. Pratt? says of this lake that after the 
eruption of 1911, the crater was occupied by a single lake about 
1 kilometer in diameter, the surface of which was 70 meters 
below sea level when the first descent was made to it. Later 
it rose until, at the time at which he wrote, it stood at about 
sea level. 
Another factor that will probably have considerable influence 
on the development of the vegetation is grass fires. In October, 
1916, and January, 1917, there was no evidence of any consider- 
able burnt areas on Volcano Island. By the first of April, 1917, 
fires had swept over a large portion of the north end of the 
island, including considerably more than half of Mount Binin- 
tiang Malaqui, much of Mount Tibag, the northern and eastern 
slopes of Mount Mataas-na-golod, and the northern slopes of 
Mount Ragatan. All these fires had occurred during the early 
part of the dry season; so that it may very well be that before the 
end of the season nearly all of the areas, in which the grass is 
thick enough for fire to spread from clump to clump, will have 
* Cox, A. J., Philip. Journ. Sci. A 6 (1911) 96. 
” Pratt, W. E., loc. cit. 
150107-——3_. 
