1190 
Journal of Agricultural Research 
Vol. XXXI, No. 12 
quantity of irrigation water applied to the land during the growing 
season. In speaking of the wheat grown in Colorado on land irri¬ 
gated with 1 and with 2 feet of water, he states that no results were 
secured which showed conclusively that the different amounts of water 
applied made any difference in the weight of the wheat per bushel or 
in the composition of the grain. Jones, Colver, and Fishburn ( 4 ) 
also made a study of the factors influencing wheat protein formation. 
In their conclusions they state that growers and millers are wrong 
in assuming that low protein wheats necessarily result from the prac¬ 
tice of irrigation and that, as a matter of fact, irrigation is not the con¬ 
trolling factor in determining the protein content of the harvested 
grain. They state further that a better quality of grain is possible 
when wheat is brought into rotation with alfalfa or red clover, since 
the activity of nitrifying organisms in the sod of these legumes when 
turned under provide a substantially larger supply of available 
nitrogen for growing wheat plants. 
Neidig and Snyder ( 5 ) made an extensive investigation of the 
factors which influence the formation of protein. Their studies 
were confined to pot experiments with different types of soil and the 
application of different fertilizers. They found that the available 
nitrogen in the soil directly affected both the quantity of protein in 
the grain and the yield of wheat under the conditions of their experi¬ 
ments. They state that climate also plays an important part in the 
quality and yield of wheat. 
In some of the publications on the influence of moisture upon 
protein formation there seems to be a lack of data relative to the 
quantity of nitrates in the soil during the period over which the 
experiments extended. 
EXPERIMENTS IN MONTANA 
From 1911 to 1917, inclusive, an investigation was conducted at 
the Montana Experiment Station for the purpose of studying the 
factors which innuence nitrate formation m the soil. As will be 
shown, the results obtained have a direct bearing on the subject of 
protein formation. The experiments were made in a dry-land section 
of Gallatin Valley having an average rainfall of nearly 20 inches. 
The soil is a silt loam, classified as Yakima silt loam. It is of alluvial 
origin, at one time the bottom of an old lake bed, and is especially 
rich in all of the plant food elements. The total nitrogen averaged 
about 0.25 per cent. This soil was cultivated first in 1910, when it 
was laid out in twentieth-acre plots. The data used in this article 
are from plots cropped continuously with both winter and spring 
wheat and from plots growing winter and spring wheat under alter¬ 
nate fallow and crop. The plots growing continuous crops of wheat 
were plowed either in the fall or early spring. The plots summer- 
fallowed were plowed in the early spring and cultivated throughout 
the growing season for the purpose of conserving moisture and Keep¬ 
ing the land free from weeds. 
SAMPLING THE SOIL 
Soil samples were taken about every two weeks for the purpose of 
determining the moisture content and the quantity of nitrates and 
nitrites present. These samples represented the first, second, third, 
