DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 517 
which this stream is said to derive its name; but its character as a Mr. belten 
“bad” river comes not at all from this regular annual glacier run- 
off, but from the great freshets that originate almost entirely below the 
glaciers. 
In the second place, the glacial area on the White River water- 
shed is barely 2% of the total area. Even if it contributed its full 
quota, based upon areas, and it could do no more, its influence would 
still be quite insignificant. The same consideration is equally true 
of the Skagit River, in which the proportion of glacial area must be 
even smaller. 
Finally, the simple fact is that there was no apparent difference 
in the run-off of those water-sheds in the flood of 1906. While the 
much steeper slope of the White gave its flow a more torrential char- 
acter, there was nothing to indicate that the yield per square mile 
from the water-shed was greater than from the Green. The overflow of 
Green River was extreme, and the destruction wrought was great. 
Even the Cedar River, a small stream, was quite as destructive as the 
White in proportion to its area of water-shed, washing away railroads 
and doing other extensive damage. 
As to the statement that the floods of White River get out of the 
way before those of the Green and Cedar come down, it is wholly 
erroneous in any general flood. Possibly a storm might affect the 
White River water-shed in some case in advance of the others, and 
this, with the steeper slope of that stream, might bring its waters 
down somewhat in advance, but this was certainly not the case in the 
flood of 1906, and in all the writer’s investigations he never before 
heard the idea suggested. The water-sheds of all these streams are 
within 24 hours (by river flow) of the lower valley, and this fact it- 
self makes it impossible that the floods of one should escape those of 
the others, except in a very slight degree. 
The writer would invite particular attention to Mr. Kuichling’s 
statement of the difficulties of treating these subjects with scientific 
precision, on account of the lack of data, and of the necessity of ac- 
quiring “a much larger volume of accurate observations” before these 
questions can be definitely settled. 
Mr. Labelle, in the same vein, touches upon an important matter 
when, in reference to the studies of the Forestry Service now in prog- 
ress for determining the various relations of forests to rainfall and 
run-off, he says “The Engineering Profession can never have full 
confidence in the results of work of this kind done by the Forestry 
Service alone.” The reason for this is that that service has a “cause” 
to promote, and it is human nature to bend everything toward the 
side of such a cause. The writer has found in his studies that in nine 
