FORESTS, RESERVOIRS, AND STREAM FLOW 251 
natural uniformity of flow, particularly in the summer time. The day 
of the small mill, which was so dependent upon such uniformity, is 
past. The modern water-power invariably seeks uniformity by artificial 
regulation, and the ups and downs of its sources of supply are abolished 
in its storage. Therefore it does not matter nearly as much that the 
run-off of the small streams be uniform as that it yield a good flow of 
water; and if forests diminish the total low-water supply, this fact 
more than offsets the gain in uniformity. Likewise, the great rivers 
swallow up and equalize the small irregularities of their head-waters 
and actually experience a somewhat larger low-water flow than if their 
water-sheds were still thickly forested. Thus, while forests may 
decrease somewhat the extreme range between maximum and min- 
imum run-off on very small water-sheds, they do not do so on great 
ones, which are combinations of very small ones. At the same time it 
seems certain that forests decrease somewhat the total run-off from 
water-sheds small or great.* 
Influence of Forests upon Snow-Melting—The second proposition 
—that forests have a beneficial effect upon the run-off from snow-melt- 
ing—is quite as firmly fixed in the popular belief as that just con- 
sidered, but has even less foundation in fact. It is a relation that 
can be definitely traced, and it can be demonstrated that the effect of 
forests upon the run-off from snow is invariably to increase its inten- 
sity. This results from two causes, one affecting the falling of the 
snow and the other its melting. 
In the first place, forests break the wind, prevent the formation of 
drifts, and distribute the snow in an even blanket over the ground. 
In the open country, the snow is largely heaped into drifts, their size 
depending upon the configuration of the ground, the presence of wind 
breaks, and the prevalence and force of the wind. These drifts form 
admirable reservoirs, and in the high mountains are the most perfect 
known. Forests prevent their formation entirely. 
The period of snow-melting begins in the open country much earlier 
than in the forests. -At first the melting is due mainly to the direct 
*This subject was ably discussed by Mr. Raphael Zon, of the Forest Service, 
Department of Agriculture, in Transactions, Am. Soc. C. E., Vol. LIX, pp. 494-495. 
He states among other things that “the quantity of water available for stream flow 
from forested water-sheds, all other conditions being equal, is less than from non- 
forested water-sheds”’; that “the forest soil receives least precipitation, next comes 
meadow land, and lastly tilled land’; that “in the forest, only the upper layer of 
the soil is moister than in the open, the lower layers being always drier.” This 
discussion is well worth perusal. 
