340 RESUME AND THEORETICAL DISCUSSION. 
the region of sufficient strength to sweep the material away after it has fallen 
and been sufliciently ground up by the action of the water. But, again, a 
sinking of the land will also bring fresh surfaces into contact with the waves, 
and thus allow the work of erosion to be continuously carried on, even where 
the abraded material accumulates at the base of the elevation from which the 
supply has been obtained. In this way, probably, the larger portion of the 
work of the ocean in levelling off the surface of the land has been performed. 
The tidal wave has beaten incessantly against an ever-sinking mass of land. 
Again, second, in regions of great precipitation, there, other things be- 
ing equal, the erosion will be most rapid. This statement is one of which 
the truth is so evident that it is not necessary to enlarge upon it. It needs 
only to recall what devastation streams swollen to very much enlarged 
dimensions have been capable of effecting. 
Third, there are to be taken into account all those varied conditions, with 
regard to the character of the rocks themselves, by which rapidity of erosion 
is affected. As a general rule, the accumulation of débris is likely to bea 
somewhat slow process; the sweeping of such material away may, under 
favorable circumstances, be very rapidly accomplished. Much will depend, in 
regard to the amount of time required for detrital material to accumulate, 
on such circumstances as these: the character and position of the joints, 
cleavage planes, and lines of stratification by which the mass is intersected ; 
the tendency of the rock to undergo decomposition on exposure to the at- 
mosphere; the amount and rapidity of the variations of temperature; the 
violence of earthquake shocks, ete. 
And, finally, the original form of the depression in which the water begins 
to run, as the result of the preceding orographic disturbances, will have a 
most marked effect on both the character and the rate of the erosion. 
In applying the above considerations more specially to the resulting forms 
of valley sections, we arrive at the following conclusions. _ 
As a general rule, large streams run in correspondingly broad valleys, with 
cross-sections closely approaching straight lines; the amount of the depres- 
sion is but trifling compared with the linear extent included within the edges 
of the water-shed. The higher we ascend into the mountains, and conse- 
quently the smaller the stream,— because nearer its source,— the more the 
valley acquires a pronounced form, capable of being represented in cross- 
section. To this, there are, of course, striking exceptions; but these have 
usually special orographic causes for their existence. 
