206 
The Philippine Journal of Science 
1917 
southwestern rim of the main crater. It will be seen that there 
is a considerable tendency to form deep, narrow canons with per- 
pendicular walls. Plate XVI, fig. 1, represents such a canon on 
the southern side of Volcano Island. The depth of this canon 
can be calculated from the size of the men in the picture. Such 
canons occur on all sides of the volcano. On the upper slopes 
where these canons originate, the surface consists of small 
rounded mounds separated by shallow depressions; the latter 
unite, resulting in the formation of ravines which enlarge 
rapidly. As these canons reach the lower and more gradual 
slopes they widen and coalesce to form the large deltal fans pre- 
viously discussed. The method of erosion, just described, has 
a very great retarding effect on the vegetation. 
The volcanic materials of which the upper layers of the soil 
of Volcano Island are composed have apparently not been weath- 
ered into a good soil. The surface is so loose that when one 
walks across it the feet sink in it to a depth of several millimeters. 
Equally striking is the comoposition of the soil, which in many 
places is composed largely of fine pebbly material. In Table 
II are given the percentages of soil particles in the different 
Table II . — Analysis of soils from Volcano Island. 
[Numbers give percentages based on total dry weights.] 
Source of sample. 
Depth 
of 
sample. 
Total 
humus. 
Total 
nitro- 
gen. 
Part- 
icles 
not 
passing 
1-mm 
sieve. 
Sulph 
Total. 
uric anhydride 
(SO3). 
Soluble in cold 
water. ^ 
Acidity 
in terms 
of Ca 
CO3. 
After 
stand- 
ing 12 
hours. 
After 
shak- 
ing 3 
hours. 
cm. 
0-10 
0.35 
0. 02 
46. 50 
13.29 
3.10 
7. 20 
0-10 
0.98 
0. 06 
28.58 
5.00 
1. 28 
1.96 
0-10 
1.38 
0. 13 
38.40 
1. 97 
0. 15 
0. 61 
Crater slopes 
5-25 
0. 37 
0.01 
45.00 
12. 56 
1.82 
5. 22 
0. 155 
Grass area 
7-25 
0. 38 
0. 02 
3.76 
4.42 
0. 09 
0.80 
0.022 
^ Soil particles not crushed ; determinations were made on particles passing through a 
1-mm sieve. 
samples that did not pass through a 1-millimeter sieve. The 
surface soil at the crater rim showed 46.5 per cent of such 
material. Preliminary examinations indicate that the water- 
holding capacity of the soil is low. 
The chemical properties of the soil have probably also had 
