682 
Journal of A gricultural Research voi. xxvii, no. <> 
per million of salt enough should be used so that io per cent of the quan¬ 
tity applied may percolate below the root zone. 
In those irrigated sections where the subsoil is saturated, that is, 
where there is a water table not far below the surface of the ground, it 
is readily possible to determine the concentration of the soil solution. 
The accumulated underground water may be taken as representing that 
part of the solution that has percolated from the root zone. Where there 
is a drainage system which relieves the underground water, samples of the 
drainage discharge are 'usually more dilute than the average of the soil 
solution because there is nearly always some waste of irrigation water into 
the drains, and, furthermore, the downward percolation of water through 
the soil is not uniform. Where the soil is more permeable there is freer 
downward movement and less concentration through evaporation and 
transpiration. 
It has been shown earlier in this paper that the detailed exploration 
of a continuous body of underground water shows great differences 
in concentration even within short distances. In general it seems safe 
to assume that where the concentration is high the movement of the 
water is slow, and, conversely, that low concentration indicates freer 
movement. Occasionally it is found that the salt content of the under¬ 
ground water is very high even where the surface conditions give no 
indication of it. Such a situation has been observed on the Huntley 
Experiment Farm in Montana. In one of the fields which has been 
continuously very productive a well was put down in the spring of 1922 
for the purpose of sampling the underground water and noting its change 
of level in connection with a series of similar wells in other fields of the 
farm. In this field there were no surface indications of abnormal under¬ 
ground water conditions. The crops had been thrifty and uniform in 
growth and the yields had been large. In the early spring, when the 
well was put in, the water table was found about 10 feet below the surface. 
It was very salty, about 4 per cent. Repeated samples were taken and 
the well was pumped out to obtain fair samples. As the season advanced 
the level of the water rose until it stood within 3 feet of the surface yet 
the crop of silage com on the field continued to thrive. Toward the 
close of the irrigation season the water receded, and during the following 
winter it stood again about 10 feet below the surface. The course of 
these changes in level and the salt content of the water as determined each 
month are shown in Table XXVI. It seems remarkable that good crop 
growth was possible on land where the water so close to the surface 
was more concentrated than sea water. The inference is that the under¬ 
ground water at this point represents the accumulated leachings for a 
long time and possibly from other places and that the rise of the water 
in this area resulted from pressures due to seepage or percolation in 
higher lands. The soil solution in the root zone must have remained 
more dilute than the water obtained from the well. The irrigation water 
used on this field contains only about 400 parts per million of salt and 
this with the rain and melted snow doubtless served to keep the solution 
in the surface soil below the critical limit, which for com is probably not 
far from 1.5 per cent. The character of the solution obtained from the 
well in this field is shown in Table XXVII. Except for its higher salt 
content, this water, resembles that obtained from the wells in adjacent 
fields and from a drain which serves the district in which the farm is 
located. 
