452 DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 
Mr, Snow. froth the soil. Fine needles act in much the same manner. Spruce, 
hemlock, cedar, fir, and larch furnish an absorbent bed, but it takes 
more than a gentle rain or common shower to get through the dense 
foliage of these trees before it is dissipated by evaporation; but little 
water reaches the ground, under these conditions, except that which 
runs down the trunks. Where the surface is covered with a thick 
moss growth, water is absorbed and retained better than on grass-land, 
but this condition obtains but sparingly in true forest. 
As to forest effects on snow-melting: The author’s position is 
certainly correct under certain conditions that are always possible, but 
seldom occur in the region here considered. Those who are responsible 
for railroad way and structures always dread a heavy burden of snow 
in the woods during late spring. If there is much snowfall, the 
normal condition is that a heavy burden will remain in the forest 
after the open ground is bare, except for drifts. If, under these condi- 
tions, a long rain occurs after a day or two of cloudy warm weather, 
the whole burden in the woods will go. Frequently the quiet reaches 
of the streams will still be covered with heavy ice, and all sorts of 
things may happen. 
_ Snow upon open land frequently melts largely from the bottom, 
if the snow blanket comes on early enough to prevent deep freezing 
of the ground and remains on continuously through the winter. In 
this case the water is very largely absorbed by the ground. In heavy 
forest, on the other hand, this does not occur to nearly the same 
extent. Although the ground seldom freezes at all, the bed of humus, 
leaves, or moss, seems to act as an insulator to prevent the ground 
heat from melting the snow. Snow sometimes evaporates without 
melting, if the air is unusually dry for a long period. Water, as we 
know it, is a meteorological phenomenon, and meteorological conditions 
govern the whole problem. Forest effects are very subordinate and 
occasional, as far as ordinary run-off is concerned; under special condi- 
'tions, such as described, the effect of large forest areas is distinctly 
bad. 
In one way, not noted by the author, forests conserve the low-water 
flow of streams of moderate size, especially that of mountain streams 
on rocky beds. The evaporation on a hot summer day from a stream 
flowing over a bed paved with cobbles exposed to the full glare of 
the sun is very great. The cobbles and boulders are alternately wet 
and dry from the splashing of the water and the drying effect of: the 
sun. Many miles of streams exist in New England that are now 
exposed to these conditions, whereas a century and a half ago they 
were largely protected from the sun’s pumping effect by overhanging 
trees and water-loving bushes along their banks. This feature of shade 
and exposure probably measures the principal advantage of forest 
cover over open land in conserving stream flow. 
