426 DISCUSSION: FORESTS, RESERVOIRS, AND STREAM FLOW 
Mr. Pickett. goes faster than if in open areas. In the timber belt this does not 
hold good in June, July, or the first weeks of August. The snow in 
the forest that softens on top, or perhaps partly melts, is crusted over 
or frozen at night, so that the melting or softening is not resumed 
until noon or later the next day. 
His view is correct when referring to the valleys of 5000 or 6000 
ft. elevation, and there the drifts he mentions go very rapidly. These 
drifts commence to melt in April and disappear very rapidly during 
the wet snows and rains not infrequent in April and May. The 
writer thinks he has an exaggerated impression of the value of the 
small gulches or depressions in the plains below the mountains. Tlie 
writer’s observation is that sage brush (in evidence everywhere except 
on the high ridges which have been denuded of soil by high winds) 
effectually prevents the drifting of the snow when of ordinary 
density. 
Twice (1881 and 1882) the writer crossed the Yellowstone about 
May 1st at Benson’s Landing and passed around the foot-hills border- 
ing the Yellowstone Valley on the south, by the Upper Indian Lodge 
Pole trail, to Clark’s Fork, a distance of 60 or 75 miles. Though 
crossing on each trip numerous small gulches, that earlier were full 
of snow, there was no snow anywhere, but the streams of any size 
were high from the melted snows. The writer differs with Colonel 
Chittenden, as to the efficacy of snowdrifts in preserving snow at any 
altitude when subject to the direct rays of the sun. Pine forests, even 
at high altitudes, are more efficacious. 
Figs. 1 and 2, Plate XX XVII, confirm the writer’s ideas in reference 
to snow. The photograph of the forest shown on Plate XX XIX was 
evidently taken just after a snowstorm. A wind of greater or less 
power always follows such storms, and soon deposits the snow on the 
ground. The temperature of the air is usually too low to permit 
the snow to melt, even should the wind fail to dislodge it. If the 
wind blows parallel to the lay of the forest during a snowstorm, the 
forest will retain only its share of the snow; but in the high mountains 
it frequently happens that a forest is on the windward slope, and in 
that case the wind will drift the snow under the limbs of the trees 
and form large drifts well up into the timber. At the end of winter, 
such drifts are sometimes 6 ft. deep. In one locality the wind has 
such velocity in passing over a mountain, above the timber line, that 
snow is taken up and drifted a mile or more and deposited in forests 
at lower altitudes. This occurs every winter. 
The following is an index of the effects of the sun’s direct rays on 
snowdrifts in the plains below the mountains: On April 1st, 1893, 
the writer was delayed by the washing away of a bridge and trestle 
on Heart River near Mandan on the Northern Pacific Railroad. In 
passing over the reconstructed bridge the cultivated field in the river 
