974 REPORTS ON THE STATE OF SCIENCE.—1914. 
moisture content of the mulched plat did not differ markedly from the 
plat on which the weeds were kept sliced off with a sharp hoe; while 
the plat on which the weeds were allowed to grow was dried out to 
a depth of 3 feet. 
A striking example of the loss of moisture from weeds is also 
shown in experiments by P. V. Cardon, conducted at Nephi, Utah.'® 
Winter wheat was grown on four plats by the summer fallow system, 
one-half the plats being in wheat each year. Two plats were fall- 
ploughed each year, and during the following summer, one plat was 
cultivated to destroy the weeds, while the other was left untouched 
except to clip the weeds in time to prevent the seeds maturing. In 
the autumn both plats were sown to winter wheat. The experiment 
was conducted for four years, and during this time the yield from the 
cultivated plat averaged four bushels more per acre than from the 
weedy plat. 
The loss of moisture in these plats as the season advanced, due to 
the demand made by the weeds, is illustrated in the accompanying 
graphs, fig. 4. That this loss is primarily due to the weed cover and 
not to direct evaporation is supported by the fact that in other experi- 
ments at this station spring-ploughed uncultivated fallow on which the 
weed-growth was slight was practically as effective as cultivated fallow 
in conserving moisture. The average moisture content (6 feet in 
depth) of the weedy Nephi plat was at the time of the spring sampling 
0°8 per cent. below the cultivated plat, and at the time of the Fall 
sampling 4°5 per cent. below the cultivated plat. This loss in moisture 
during the summer is equivalent to 3°5 inches of rainfall stored in the 
soll. This amount of water is sufficient, according to the water- 
requirement measurements of Briggs and Shantz,!7 to produce ten 
bushels of wheat per acre at Akron, Colorado, where the evaporation is 
the same as at Nephi. In 1911 the actual increase in yield of the 
cultivated plat over the weedy plat was eleven bushels per acre. 
During the other years the yield was reduced by winter killing, so that 
the water-supply was not the primary factor in determining production. 
Surely no more convincing proof is needed of the necessity of keeping 
fallow land free from weeds in regions where the moisture supply is 
of primary importance. 
Growth-water. 
It has long been known that a part of the soil-moisture is held so 
tenaciously that it is not available for the growth of plants. Sachs 
in 1859 appears to have been the first to recognise that the percentage 
of non-available moisture varies greatly with the type of soil. This is 
# matter of fundamental importance in the interpretation of soil- 
moisture observations, for the water unavailable for growth ranges 
from 1 per cent. or less in sand to 80 per cent. or more in the heaviest 
16 Office of Cereal Investigations in Co-operation with the Office of Biophysical 
Investigations. See Tillage and Rotation Experiments at the Nephi sub-station, Utah, 
U.S. Department of Agriculture, Bulletin 157, 1914. 
VW Briggs, L. J.,and Shantz, H. L., ‘Relative Water-requirement of Plants,’ Journal 
of Agricultural Research, U.S. Department of Agriculture, 8, 1, 1914. 
