60 
Research Bulletin No. 5 
six inches. The soil was devoid of available water at the last 
sampling-. Where the weeds were killed, a little more than 
one-half of the rain which fell was saved. The second six inches 
was moistened, and the soil was in good condition to plow. We 
attribute this difference more to the killing of the weeds than 
to the fact that the surface soil was disked. However, it was the 
disking that killed the weeds. 
Table 27. — Effect of iveeds on soil 'moisture, 1912. 
PER CENT MOISTURE IN THE SOIL 
Depth 
July 27 
July 29 
Aug. 2 
Aug. 5 
Aug. 12 
Weeds killed 
with disk 
Inches 
1-6 
7-12 
24 
36 
21.2 
9.4 
5.4 
5.4 
14.6 
9.8 
5.3 
5.7 
21.6 
9.6 
5.4 
5.3 
16.5 
11.4 
5.7 
5.9 
12.0 
13.0 
6.5 
5.7 
8.7 
7.7 
8.8 
8.5 
8.2 
Weeds growing 
1-6 
7-12 
24 
36 
Average 
17.2 
6.8 
5.8 
6.3 
9.6 
6.6 
5.7 
6.2 
19.0 
6.8 
5.8 
6.1 
13.6 
6.4 
5.8 
6.8 
6.2 
5.7 
5.6 
6.2 
8.0 
6.7 
7.9 
7.5 
5.9 
In Chart VI is graphically shown the water content, by foot 
sections to six feet, in the soils of three plats receiving different 
treatment. The data are shown in Table 17. The solid line 
represents the water content of a plat receiving the best tillage 
we could give it. The broken line represents the water content 
in a plat not cultivated but the weeds kept down with a hoe. 
The line designated by crosses represents the water content in a 
plat where weeds were allowed to grow. This chart shows that 
growing weeds were a much greater factor than lack of cultiva- 
tion in losing water from this soil when previously filled with 
water. After the weeds had been killed by frost so that they 
were no longer using so much water, rains partially filled the 
first two feet of soil. The difference at the last sampling is 
therefore not so great as earlier in the season. 
Chart VII shows graphically the use of water from the soil 
by a corn crop. For explanation of chart, see that given for 
Chart ITT. This plat was summer tilled in 1907 and cropped to 
