FORESTS, RESERVOIRS, AND STREAM FLOW 279 
and more safely than a similar stream in the lowlands. The banks are 
almost always stable and the bottoms rocky or composed of heavy gravel 
or boulders; in fact, floods do less harm on such streams than on any 
others. In the low land, where the streams have smaller slopes and un- 
stable banks, much smaller run-off produces greater floods and more 
destruction. Moreover, Nature to a large degree adapts streams to 
the work required of them. ‘The channels of the tributaries of the 
Ohio have been carved out through long ages to carry in safety the 
average flood flow. Area for area of water-shed, their cross-sections 
are much larger than those of streams in climates of less rainfall. 
The normal section of the Ohio at Wheeling is over 2 sq. ft. for every 
square mile of water-shed, while that of the Kaw River at Kansas City 
is less than 4 sq. ft. per square mile. It is therefore wholly erroneous 
to conclude that the streams of these mountains are more subject to 
over-bank freshets than those of the low lands or that the freshets 
themselves are more destructive. Considering the conditions growing 
out of settlement, the reverse is unquestionably the case. 
There is one other consideration of prime importance in this for- 
estry argument, and that is the fact that no possible development of 
forestry can increase the present percentage of forest-covered areas. 
At least as much ground as is now devoted to agricultural purposes 
must continue to be so used. The utmost admissible expansion of 
National forests will never require a greater area than is now occu- 
pied by forests and second growth or logged-off lands, which, so far as 
run-off and erosion are concerned, are just as effective as the virgin 
forest itself, and more effective than will be the groomed forest of the 
new régime. There may be a shifting of areas devoted to forests, but 
possible expansion, compared with the present area, is so small that its 
influence upon the great rivers, even admitting the full force of the 
forestry argument, would be wholly inappreciable. 
The fact just dwelt upon should make us thankful that the forestry 
theory as to the stream flow is not correct. Whatever the value of 
forests, we cannot have them everywhere, and by far the greater por- 
tions already cleared away must always remain deforested. If this 
fact of deforestation has brought with it in greater degree than of old 
the calamities of high and low waters, then, indeed, we are in an 
unfortunate case. But it has not done so. Nature has decreed no such 
penalty for the subjugation of the wilderness, and on the whole these 
