DISCUSSION: FORESTS, RESERVOIRS, AND STREAM FLOW 375 
whom there is no more competent authority—in his paper on “Forests Mr. Swain. 
and Snow,”* comes to the following conclusions, among others: 
(a). “The greater the amount of forest cover the less violent the 
daily fluctuation, the more uniform the flow throughout the day and 
throughout the season, and the later the stream maintains its flow.” 
(b). “The loss of the forest cover means more violent fluctuation 
during the day, greater difficulty in regulating the headgates and keep- 
ing a uniform flow in ditches, and hence an additional difficulty in the 
economic distribution of water. Also the water runs off sooner, hence 
the streams drop earlier in the summer, and on account of the lessening 
of the springs, the smaller is the winter flow.” 
(c). “The preservation of the forest is an absolute necessity for the 
interest of irrigated agriculture.” 
The writer is not in a position to question the statement of the 
author with reference to Yellowstone Park. He has observed, however, 
that the greater part of the Park is covered with forests. What the 
condition of things would be if this area were deforested can only be 
conjectured, and the writer would like to inquire whether Fig. 1 is 
simply a fictitious diagram showing the author’s opinion, or whether it 
is based upon measurements of flow from approximately equal areas 
under similar conditions in every respect except that one is forested 
and the other not. The diagram does not coincide with the experience 
of the writer, which has convinced him that the snow run-off in 
forested areas extends later into the spring than the snow run-off on 
open ground. 
Of course, in this, as in the other matters discussed, it is possible 
to conceive of circumstances under which a particular forest might 
increase a particular flood. If the forests retain a thick blanket of 
snow late into the spring, and there comes a sudden warm rain which 
carries away all this snow, there may result a freshet of greater violence 
than if the sun had previously melted the blanket of snow; but it is 
equally clear that if the forest had not been present, and the snow 
had gone off with the snow from open ground, the earlier freshets 
would have been increased. If a given amount of snow has got to run 
off into the streams, it is certainly an advantage to have the period in 
which it runs off lengthened. 
With reference to the experience in the region of Puget Sound, the 
author states that the snow storms are followed by warm southerly 
winds and rains, which melt the snows and result in tremendous 
freshets and floods. A warm southerly wind and rain will easily melt 
18 in. of snow even on the ground, and the author has not shown that 
the condition now existing would be bettered if the forests were cut 
down and the snowfall of 18 in. rested entirely on the ground instead 
of partly on the trees. His remarks on this point are an excellent 
illustration of illogical reasoning. He says that the great flood of 1906 
“was a perfect demonstration of the vastly intensifying effect of forests 
* Bulletin 55, Agricultural Experiment Station of the Agricultural College of Colorado. 
