Mr. Colling- 
wood. 
-etc., have continued the forest destruction. 
. 
820 DISCUSSION : FORESTS, RESERVOIRS, AND STREAM FLOW 
remained unmelted long after every bit of snow had disappeared from 
the cultivated portions of the valley. It was a common remark that 
there was enough left to add perceptibly to the flood usually expected 
in June. Exposure and the kind of forest may have had something to 
do with this, but the fact remains. 
Passing to the third proposition—that forests increase the precipi- 
tation of moisture—it is not believed that this idea is now held to 
any extent by meteorologists. While admitting this, it does not follow 
that they have no effect upon storms. Probably no person would ques- 
tion the statement that a chain of mountains causes most of the 
moisture in the clouds to be deposited on their windward side, so that 
the rainfall is always greater there, and less on the opposite or leeward 
side. If this be true, are we not forced to believe that a belt of forest 
must have a similar effect, but in less degree? 
When there is nothing to deflect them, the several moisture-laden 
currents of air tend to travel parallel to each other. Let them meet a 
forest, and the lower current is deflected upward, and, by mingling 
with the upper, causes the formation of clouds and rain. 
The contention, then, is that forests tend to cause a more uniform 
distribution of the rainfall, lessening the mighty downpour and giving 
more frequent gentle showers; with this would follow fewer severe 
winds and an amelioration of the climate. It is probable, also, that 
forests act electrically to induce precipitation, by restoring equilibrium 
between earth and clouds. ; 
The author’s fifth conclusion in summing up is that “Soil erosion 
does not result from forest cutting in itself?’ and that “The natural 
growth which always follows the destruction of a forest is fully as 
effective in -preventing erosion.” ; 
The writer’s contention is: That forest destruction is always fol- 
lowed by a more rapid flood discharge, a consecutive increase in the 
flood volume, a diminution in the low-water volume, and a greatly in- 
ereased erosion of the banks and damage of the flood plane in the lower 
reaches of every stream. His opinion is based largely on his life-long 
knowledge of a single river, he having been born in 1839 at Elmira, 
N. Y., on the banks of the Chemung. The conditions in his boyhood 
days were these: The stream had a fairly uniform width; the banks 
were largely grassed; the channel was comparatively uniform; there 
were very few bars, and all through the summer there was a good 
flow of water.* 
On May 31st, 1889, and the following day, there was a freshet, 
exceeding in height any previously known, this being the result of a 
very heavy and extensive rain. The writer was employed to make an 
* This whoje region, including the head-waters of the streams forming the Chemung. 
had been heavily wooded, and the cutting and rafting of pine lumber, for a market on the 
Lower Susquehanna, formed one of the chief industries for many years. Rafting ceased 
somewhere in the early Fifties, but lumbering and the cutting ms haranecit for ties, fuel, 
