6J2 
Journal of Agricultural Research voi. xxvii, n©. 9 
held in other parts of the swamp. When evaporation is going on rapidly 
the water in some parts of the swamp may become much more concen¬ 
trated than in the channels through which the water is moving freely. 
In much the same way the water applied to a field by irrigation enters 
and moves through the underground water in the more permeable chan¬ 
nels and passes out in the drainage without necessarily mixing with the 
water held in the less readily permeable areas of the subsoil. 
The actual course and rate of the movement of underground water is 
not easy to observe. Under some conditions it has been found possible 
to estimate the rate and direction of the movement of what is called 
underflow by putting into it at some point, as in a well, a quantity of 
some substance, such as a dye, which could be identified in samples taken 
from adjacent wellp. Such investigations show that the actual movement 
of water by underflow is usually very slow. The fact as to whether or 
not there is a definite movement of underground water with its conse¬ 
quent mixing and something as to the rate and range of such movement 
may be established by the chemical examination of a series of samples 
of underground water from any given area. These observations as to the 
quantity and character of the dissolved substances, together with the 
fluctuations in the elevation of the underground water resulting from 
irrigation, give a due as to the rate and extent of the movement of such 
C#A4 
Fig. 9 . —Results of an experiment on the Frannie division of the Shoshone project, made in 
December, 1922 . 
water. Where it is found that the water taken at several points is similar 
with respect to its dissolved materials and that similar changes in level 
result from a common cause it may be inferred that there is a free move¬ 
ment. When these observations show pronounced differences it may 
be assumed there is very little movement or that the movement is con¬ 
fined to restricted zones or channels. 
As an illustration of a situation in which there appeared to be very little 
movement in the underground water, a series of observations made 
on the Shoshone project, Wyoming, may be used. These observations 
were made by putting down a line of wells across the valley of a coulee 
around the head of which the Deaver Canal is located. 17 The canal 
forms a loop at this point which is crossed by the line of wells as shown 
in figure 9. 
The wells were put down about 250 feet apart. The depth to water 
ranged from 2 to 10 feet below the land surface. The gradient of|the 
plane of underground water is such as to indicate an absence of equili¬ 
brium and consequently a potential movement toward the coulee. In 
17 The writer is indebted to J. R. lakish, of the United States Reclamation Service, for conducting the 
field work and collecting the samples of water. 
