16 BULLETIN 291, U. S. DEPARTMENT OF AGRICULTURE. 
regions crops are selected without regard to their water economy, but 
in the drier sections of the country those crops which show a low 
water requirement are the most likely to mature in years of deficient 
rainfall. 
In the water-requirement measurements of Briggs and Shantz x at 
Akron, Colo., they have found that millet and sorgo are the most 
efficient in the use of water of any of the numerous crops compared 
at that place. In their summary of water-requirement determina- 
tions in 1911, 1912, and 1913 (table 34, op. cit.), the authors men- 
tioned give the water requirement of Dakota Kursk millet (S. P. I. 
No. 34771) as 265 and that of Minnesota Amber sorgo (A. D. I. No. 
341-13) as 305. The average water requirement of 11 varieties of 
corn was 368, of 7 varieties of wheat 513, and of 4 varieties of alfalfa 
831. On this basis, the water required to produce 100 pounds of dry 
matter in Dakota Kursk millet would produce only 87 pounds of 
Minnesota Amber sorgo, 72 pounds of corn fodder, 52 pounds of wheat 
(grain and straw), and 32 pounds of alfalfa hay. The results con- 
firm the experience of farmers, who find that sorgo and millet are 
among the most dependable crops grown in the Great Plains. 
In 1912 the water requirement of Dakota Kursk millet was 187, 
being ' l the lowest water requirement so far recorded for any crop at 
Akron." In comparison with other crops this was a remarkable 
showing. German millet had a water requirement of 248, the proso 
millets 207, Red Amber sorgo 237, corn 280, Sudan grass 359, and 
alfalfa 659. The water required in 1912 to produce 100 pounds 
of Dakota Kursk millet would produce only 28 pounds of alfalfa, 52 
pounds of Sudan grass, 67 pounds of corn fodder, 79 pounds of Red 
Amber sorgo, and 75 pounds of German millet. 
WATER REQUIREMENT OF SELECTED STRAINS OF MILLET AT NEWELL. 
The plant breeder who is working for increased drought resistance 
should determine the relative water requirement of his selected 
strains, as the method is a simple and comparatively rapid one and 
the differences, if significant, afford one of the best indications of 
differences in adapt ability to drought conditions. 
In 1912 and 1913 determinations were made of the water require- 
ment of two strains of Kursk millet, Dakota Kursk (A. D. I. No. 3) and 
Kursk (A. D. I. No. 13-3), in comparison with a strain of Siberian 
millet (A. D. I. No. 4-3). In 1912 common millet also was included 
in the experiment. In 1914 only Dakota Kursk and Siberian 
millet (A. D. I. No. 4-3) were compared. The water requirements 
were as shown in Table X. 
1 Briggs, L. J., and Shantz, H. L. Relative water requirement of plants. In Jour. Agr. Research, v. 3, 
no. 1, p. 1-63, 1 fig., pi. 1-7, 1914. 
