26 
BULLETIX 835, U. S. DEPARTMENT OF AGRICULTURE. 
the moisture an inch in light soils than in heavy soils. In flume 31 
more moisture was required per inch of advance during the first few- 
days than during the fifth day. but after the fifth day there was a 
gradual increase in the moisture required. This same condition was 
found hi flume 200. It is probable that this results from the fact that 
the moisture percentage changes to a very much greater extent near 
the tank end of the flume than it does toward the other end. and 
especially is this true the first few days. It is also noted from the 
results of the vertical flumes that for a distance of 14 inches above 
the surface of the water the moisture moves rather slowly upward. 
It is probable, therefore, that during a period along about the fifth 
day there is not sufficient moisture near the top of the flume to permit 
a maximum evaporation. After that time evaporation takes place 
more rapidly and hence the increase in water consumed. Another 
fact that will be brought out in the distribution of the soil moisture 
in these flumes is the gradual increase in the percentage of moisture 
throughout the wetted area of the flume from day to day, this con- 
stantly increasing percentage continuing until very near the point of 
capillar} 7 saturation. 
In a review of Table 13 it is found that the order of the water 
requirements of these flumes at the end of the twentieth day, begin- 
ning at the one requiring the least water, is flume TO, 200, 20, 31, 90, 
and 50. Comparing this order with the order of the moisture equiva- 
lents of these soils, and beginning with the' least moisture equiva- 
lent, we find the order just the same as above, except that flumes 70 
and 200 are reversed. This interchange of place is probably ac- 
counted for hj the fact that flume 200 would permit of more evapora- 
tion per square inch than would flume 70. 
From the data in Table 13 it is possible to calculate for the hori- 
zontal flumes the quantity of water removed during any period of 
time, in cubic inches. Assuming the same rate of use in nature as in 
the flumes, the use of water by the soil in place can be calculated in 
acre-inches. In preparing Table 11 such assumptions were made. 
Table 14. — Loss of water from tanks in different periods of tune. (Depth ex- 
pressed in inclies on an area of 100 square inches. 
Flume. 
Xumber 
of davs. 
20 
31 
50 
70 
90 
200 
Inches. 
Inches. 
Inches. 
Inches. 
Inches. 
Inches. 
1 
4. 58 
6.10 
4.88 
3.67 
7.32 
6.10 
3 
7.93 
9.76 
7.93 
6.10 
12.21 
9. 15 

10.07 
11.59 
10.76 
7.34 
15. 87 
11.59 
10 
14. 34 
15.87 
12.81 
10.37 
22.58 
15. 25 
15 
18.00 
18.22 
14.95 
12.82 
28.68 
19.53 
20 
21.82 
22.27 
16. 48 
15. ^7 
34.78 
2}. 40 
30 
26.85 
27.46 
19.12 
20.14 
42. 72 
