OCCURRENCE OF SMALL DESERT TRACTS. 17] 
(iv.) In texture the sedimentary beds are gritty and 
coarse. Their component grains and fragments 
vary considerably in size. They are generally 
more or less angular and are rather tightly packed. 
(v.) The red, fertile earth, on the contrary, is fine and 
regular in grain, and its individual particles tend 
towards a spherical shape. The packing is by no 
means tight. 
There can be no question that at one time the red earth 
cliff reached as far as the river bank, and it is probable that in 
those days precipitation was greater than now (see note at 
bottom of page 167). One may well imagine a gulley running 
off the escarpment to the river and being rapidly deepened 
during the rainy season. The intermittent stream would soon 
cut its way down to the sedimentaries below, and a small 
corrie might be started at its head by the collapse of the soft 
beds above, facilitated, perhaps, by a spring at their junction 
with the underlying series. Phenomena of this nature are 
by no means uncommon up-country, where the soft laterite 
overlies less permiable beds below. Whatever may be the true 
interpretation of the early beginnings of the desert tract, 
there is no doubt that exposure of the sedimentaries was the 
first landmark in its history, and (given a diminishing rainfall) 
from thence onwards the expansion of the barren area was 
merely a matter of time. 
The thickly forested red earth shows that the unequal 
distribution of rainfall throughout the year* is not the main 
determining factor in the formation of the desert floors ; but 
it seems equally obvious that the absence of moisture is the 
cause of the barren nature of the tracts, and one naturally turns 
to the consideration of soils as water carriers. 
The amount of water that can be held by soils and subsoils 
when saturated depends on the size and shape of the particles and 
stones (their mean diameter being termed the effective size), and 
on the consequent pore-spaces. Here...... we cannot do better 
than quote from Warington,} who states that, *‘ if a soil consisted 

* The yearly average for the part of the country is probably about 
forty inches (see ‘‘ Manual of the Puttalam District,’ 1908, p. 20). 
+ Referring to ‘* Physical Properties of the Soil,’ 1911. 
13 6(8)15 
