’ 
DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 491 
Mr. Snow has called attention. Anyone who has observed the condi- mr. cnitten- 
tion of the forest bed under deciduous trees just after the snow has 4e. 
disappeared will understand this. Those leaves virtually form an oil- 
cloth separator between the snow and the ground, and whatever melt- 
ing there may be from beneath is drawn up into the open sponge above 
rather than through the leaves into the ground beneath. This fact 
should be borne in mind by those who lay so much stress upon the 
frozen condition of the ground on our eastern water-sheds when the 
floods of winter and early spring take place. This condition in the 
open country is partially, and possibly fully, offset by that of the forest 
bed above described, particularly when the bed of leaves is itself sealed 
up by frost. 
The curves (Fig. 1), to which Mr. Leighton refers so slightingly, 
will stand as a rational expression of a scientific truth which will be 
fully recognized as this subject comes to be better understood. The 
time scale has more immediate application to the great central Rocky 
Mountain region about the sources of the Yellowstone.. The vertical 
scale is, of course, only a most general expression. As shown in the 
figure, it represents the writer’s estimate of an extreme case, like the 
flood of 1899. In average years the two crests are much closer together. 
The great saving factors in these mountain floods are difference of 
altitude and changes of weather. The lower altitudes yield up their 
waters first, and these are safely carried away before those from the 
higher altitudes come down. The melting at the same altitudes is 
often interrupted by a sudden drop in temperature. On more than one 
occasion has the writer taken measures to patrol the roads in the Yel- 
lowstone when conditions seemed to make a flood imminent, but has 
been spared any trouble by a sudden change of weather which stopped 
melting, tightened up the little streams and allowed the accumulated 
waters to run away. This, in fact, is the general rule; but every five 
or six years a season will come in which warm weather and rains 
continue for several days without interruption, and it is then that 
disasters follow. 
The writer will note here a common, but very erroneous, impression 
to the effect that delay of forest melting is a benefit to irrigation. This 
point has been urged by Colonel Pickett and by the Editor of E'ngi- 
neering News. Far from being a benefit, this delay, by concentrating 
and intensifying the run-off, is a positive injury by the damage it 
causes to irrigation works. The water comes in such volume that it 
cannot be used, and all the excess which constitutes the flood is abso- 
lutely lost to irrigation. Storage reservoirs alone can save it. Par- 
ticular attention is invited to the fallacy of popular thought in this 
matter, because it has a bearing upon the main question here con- 
sidered. 
Professor Swain takes exception 1o the writer’s statement that the 
