DISCUSSION: FORESTS, RESERVOIRS, AND STREAM FLOW 341 
presses, dwellings, and many other improvements deliberately located mr. Todd. 
and built on ground which is sure to be overflowed again; and, what 
is more surprising, only in a very few instances have steps been taken 
to protect these improvements when high water comes. 
The writer does not mean to condemn reforestation as applied to 
forestry, having observed too many examples of wanton waste of 
valuable timber throughout the entire Mississippi Valley. It is doubt- 
ful if there has been anywhere a greater waste of timber than along 
the shores of that river. A recent magazine writer deprecates the fact 
that he saw “squashes and pumpkins” caving into the river with the 
banks and going to waste! The loss of growing crops by caving banks has 
been very small, compared with the great amount of merchantable timber 
which has been lost in the river in that way during past years; and 
even now there is a great annual loss. These trees, as pointed out by 
Colonel Chittenden, after they fall into the stream, vie with the sand- 
bars as to which can cause the greatest obstruction to navigation. 
Again, the writer has seen areas of hundreds of acres of the finest 
timber land cleared by deadening the trees, which are thus utterly 
ruined. Timber firms have acquired the timber right on lands, and 
have cut everything large enough to make a railroad tie with two 
sides flat; however, there is usually such a variety of timber growing 
that the land is seldom entirely cleared from this cause. Timber 
conservation, local or otherwise, however, would not have the slightest 
effect upon the regimen and flow of the lower river; and, as Colonel 
Chittenden well states, “no engineer could honestly advise lowering in 
height by a single inch the levees of the Mississippi” even if the 
fondest expectations of the forestry advocates could be realized. 
The whole matter, of the benefit that the forestry experts expect 
to contribute to the ultimate improvement of navigable rivers by the 
application of forestry, rests on theory built up on certain rainfall data 
and stream gaugings collected in various sections of the country. 
Many of the data, though most carefully observed and compiled, are 
often very discrepant and misleading, owing principally to the lack of 
a sufficient number of gauging stations properly located and efficiently 
observed. 
The paper by John C. Hoyt, Assoc. M. Am. Soe. C. E., on rainfall 
and stream flow,* and the discussions thereon bear out the foregoing 
statement. . 
It is a peculiar fact that nearly all writers and investigators of 
rainfall and stream flow neglect to take into consideration the influence 
of slope on run-off percentages. The author states: 
“Tt is perfectly true that more rain falls on the hills than on the 
lowlands, that a greater percentage of rainfall runs off from steep than 
from flat slopes, etc.” 
*Transactions, Am. Soc. C. E., Vol. LIX, p. 431. 
