Dec. 4, 1916 
Evaporation of Moisture from the Soil 
447 
20 to 240 per cent. The muck consisted almost entirely of vegetable 
mold gathered from accumulated pockets in the brush swales of Logan 
River. It was known to have high water-holding capacity from previous 
experiments. 
The pans were kept several inches back from the edge of the tables on 
which they rested. This almost entirely eliminated the effect of air 
currents which had, in previous tests, caused some variations in the rows 
of pans set close to the edge. The loss was made up each day after the 
weighing. The winter losses were much higher in this case than with 
the loam previously reported. 
Figure 2 gives the results for sand and shows an increase in evaporation 
as the initial moisture is increased, up to 33 per cent. The most rapid 
increase is up to about 
7 per cent. Above this 
point the increase in 
loss is not so great. 
Figures 3 and 4 give 
the curves for clay and 
muck. The results for 
clay show a more grad¬ 
ual ascent in the curve, 
and a higher point 
before there is any 
break, than do the re¬ 
sults for sand. This 
might have been ex¬ 
pected from the great 
water-holding power 
of the clay. Muck with high percentages and differences at greater 
intervals produced about the same kind of curve as did loam. 
More tests are necessary, however, to establish points as nearly exact 
as was done in the case of loam. 
HUMIDITY OF THE AIR 
To study the effect of a saturated atmosphere on evaporation, a set of 
wet soils in Petri dishes was placed in an air-tight copper germinator. 
Forty gm. of dry soil were put in each of these dishes, which were about 
3 inches in diameter. Duplicate vessels were made up with soil for each 
1 per cent from 1 to 35, and for each 2 per cent from 35 to 45 per cent of 
moisture. They were then set on the shelves of the germinator, which 
is a hollow box 22 inches wide, 37X inches deep, and 47^ inches long. 
The whole was surrounded with a water jacket, except at the doors, 
which were made of two panes of glass inclosing a dead-air space about 
half an inch thick. The perforated copper shelves were covered with 
Fig. 2.—Evaporation from sand containing different initial per¬ 
centages of moisture. 
