RANGE PRESERVATION AND EROSION CONTROL. 11 
In correlating the meteorological data with stream flow from melt- 
ing snow it was found that the stream flow follows very closely the 
fluctuations in temperature. The relative behavior of these factors 
is shown in figure 2. The flow increases with the rising temperature 
of the day and reaches the maximum at practically the same time 
that the mercury is at the highest point. The flow practically ceases 
between 8 p. m. and 10 p. m. and remains practically at zero during 
the cooler hours of the night, only to rise again with the increased 
temperature of the following day. High winds are found greatly to 
increase the rate of evaporation of the snow cover, but they affect the 
run-off relatively little. 
Any medium, such as an effective vegetative cover, which serves 
to insulate the heat from the snow cover, breaks the effect of high, 
dry winds, and at the same time intercepts the run-off more or less, 
will tend to conserve the snow, regulate run-off, and make possible 
the absorption of a larger amount of water. This fact has been 
demonstrated on the wooded portion of the selected areas as well as 
on the extensivety denuded, sparsely vegetated, and timbered lands 
on the forest generally. 
To sum up the facts concerning the action of melting snow : Ero- 
sion from melting snow is a more serious factor than generally sup- 
posed when the vegetative cover is sparse and the slope steep. Both 
run-off and erosion from melting snow vary in intensity more or less 
directly with the character of certain climatic factors, especially tem- 
perature. In general, the soil is not frozen under a cover of a few 
inches of snow if the latter falls before cold weather early in the 
winter; so whenever melting takes place erosion may occur unless the 
soil is held firmly in place. The most rapid melting of snow and 
the most serious erosion occur where there is a lack of vegetation. 
In general, snow lies the longest on timbered lands. 
RAIN. 
An examination of the accompanying tables showing rainfall and 
the resulting run-off, or lack of it, disclosed several interesting facts. 
In the first place, out of the 26 rainstorms for the year 1915 (Tables 
1, 2, 3), distributed over the four months from June to September, 
inclusive, only one storm — that of July 21 — produced run-off. At 
this time, according to the record of the four rain gauges, 0.70 and 
0.71 of an inch of rain fell on area A and 1.48 and 1.38 on area B, 
within a period of 65 minutes. From area B the run-off was 335 
cubic feet and it carried 94 cubic feet of air-dry sediment, as com- 
pared to 3,019 cubic feet of run-off on area A and 717 feet of air- 
dry sediment (fig. 3). It should be kept in mind that the run-off 
from area A was enormously greater than on area B in spite of the 
