130 
SKETCH OF THE PHYSICAL GEOGRAPHY 
ordinary action of tropical rivers, or inconsistent with the supposition 
that the line where the lower slopes of the hilly ground* when they 
attained their present elevation, met the sea, is distant from the line 
where the highest tidal wave now ceases. When the first alluvium was 
formed along the base of the lulls, its surface would incline to a level. 
By the accumulation of matter in the beds of the rivers they would be 
raised above the surrounding level, their courses would shift, and by 
these means, and the constant periodical inundations, the face of the 
land would gradually rise. In time, by the greater deposits nearer the 
hills of heavy detritus in the shifting channels, and of lighter matter 
from the inundations, the older portion of the alluvial tracts would 
acquire a slope. In the earlier ages of their history the inundations 
from the interior would probably rush down the valleys over broad 
areas, and, when they entered the plain, spread over them, and deliver 
their more bulky spoils over wide tracts. At this epoch the alluvium 
would tend to rise rapidly above the sea level and encroach on the 
valleys. In later ages the valleys would have become wider and 
lower towards the interior, and deeper channels have been excavated 
in them. As the level of these channels above the plain descended, 
the channels in the plain would also he scooped deeper. The slope 
of the plain, and the softness and want of cohesion of its materials 
would facilitate this operation. In the result, a broad and deep 
trough would he formed, capacious enough to contain the waters of 
inundation. The river would then cease to increase, and rather tend 
to abrade, the older part of the plain, and the detritus of the moun¬ 
tains would not be deposited until the loaded waters readied and 
overflowed a newer external tract of alluvium, less elevated above the 
bottom of the channel. It is also to he supposed that during the ear¬ 
lier periods of the physical history of a country like Sumatra, consist¬ 
ing apparently of plutonic and volcanic rocky masses elevated beneath 
or through stratified formations, the detritus would be more abundant 
than in after ages. The surface of the newest sedimentary strata 
would be often soft and easily yield to the rains and torrents. From 
the summits or higher levels of the plutonic ridges, where the abrad¬ 
ing action would be greater, all the softer rocks would gradually be 
