24 BULLETIN" 1139, U. S. DEPARTMENT OF AGRICULTURE. 
Certainly the greater quantity of moisture stored by either method 
in any section is not enough to counterbalance other factors such as 
timeliness of work, control of weeds, and prevention of soil blowing. 
Plat C or D has been markedly superior to the other two plats in 
moisture storage at all stations except those with a limited water- 
storage capacity. At Edgeley, it will be remembered, the soil is so 
shallow that practically all of it has been filled to capacity for all 
methods of cultivation each year between harvest and seeding; con- 
sequently, little difference between methods has existed. At some 
other stations, such as Archer and Ardmore, the difference in favor of 
plat C or D has been small. The superiority of C or D over A and B 
in individual years has depended upon the quantity of rainfall as 
well as the character of the soil. The demand of the wheat crop for 
moisture differs in the several portions of the Plains, 7 and a given 
quantity of stored water may be of more value at one station than 
another. When translated into bushels of yield 4 inches of water 
stored at Assinniboine would probably be much more valuable than 
the same quantity of water at Amorillo. The superiority of plat 
C or D over the other plats in conserving moisture must be con- 
sidered in terms of increase in bushels per acre to determine the 
value of fallowing as a farm practice. 
Alternate cropping and summer fallowing practiced in these 
experiments has in all but the driest years stored moisture in the 
entire zone of natural development of the roots of the wheat crop. 
It is evident from this that any extension of the fallow period for 
the purpose of increasing the quantity of moisture stored would not 
often be effective in increasing the yields of wheat. The more likely 
result in most years would be the storage of moisture at a depth 
where its recovery by the wheat crop would be extremely doubtful. 
The three plats taken together indicate that moisture is the con- 
trolling factor in crop production in the Great Plains and that differ- 
ences in yield generally are due to differences in the water supply. 
Plats A and B have produced yields almost or quite equal to those 
on plat C or D in years when they have contained the same quantity 
of water. 
GENERAL CONCLUSIONS. 
Under dry-farming conditions there is present in the soil no 
ground water or other source of free water. Below the zone of a 
few feet near the surface, which the present study shows may be 
wetted and dried during the cycle from harvest to harvest, the soil 
is either dry or does not contain water above its field carrying 
capacity. Under these conditions no water moves upward through 
appreciable distances to replace the water removed by the roots. 
Water is supplied to the roots only by such part of the soil as they 
occupy, and only that part of the soil suffers exhaustion or reduction 
of its water content. 
The development of the roots of the wheat plant is indicated by 
the depth and extent to which the soil water is used. The usual 
depth of development is indicated by the results given in detail in 
Table 1 and summarized in Table 2. It appears that at stations 
7 Cole, John S., and Mathews, O. R. Use of water by spring wheat on the Great Plains. 
U. S. Dept. Agr. Bui. 1004, 34 p., 10 fig. 1923. 
