COLORADO IRRIGATION WATERS AND THEIR CHANGES. 43 
sider, as suggested in the observations made on the irrigation of 
1898, that the ground water as taken after irrigation represents a 
mixture of equal parts of irrigation water and ground water we 
find that to produce this change in the amount of total solids 
12,856 pounds of salts must have passed into solution. 
§ 69. For the experiment of 1898 we found that 4,411 
pounds went into solution or, assuming a mixing to the extent of 
equal parts, 12,664 pounds per acre-foot; for 1899 we have 4,941 
pounds and 12,856. When we attempt to find how this gain was 
distributed between the different salts we find the same order, 
namely sodic sulfate, magnesic sulfate and calcic sulfate. In 
1899, however, the calcic sulfate shows a greater increase than in 
1898. This is accounted for by the influence of well D, which in 
1898 could not be included, because being near the point of 
onflow it was not looked after as carefully as it should have been, 
and the water getting advantage of us ran into the well from the 
surface. In 1898 the percentage of calcic sulfate in the residue 
from the ground water was lower after irrigation than before, ex¬ 
cept in the case of well D, which showed an increase of five per 
cent. The result is probably correct and represents what actually 
took place, but it is contrary to our observations. While it mod¬ 
ifies our general results, it does not reverse them. 
§ 70. The potassic oxid in an acre-foot of the ditch water 
used was only 11.6 pounds, in the ground water before irrigation 
5.8 pounds, in the ground water after irrigation 18.3 pounds, or if 
we consider the ground water after irrigation as representing 
a mixture of equal parts, as before, we have 19.2 pounds of 
potassic oxid brought into solution by the application of an acre- 
foot of water. 
§ 71. The water that flowed over and off of the plot was not 
large in quantity but we collected samples as near the beginning 
and end of off-flow as was feasible. The salient features of the 
results will be seen upon an examination of the analyses. 
§72. The off-flow took place at two points, one near the 
center of the north side of the plot, the other at the east end, the 
water flowing from west to east. 
§ 73. The samples obtained of the off-flow on the north side 
showed a very marked difference in the quantity of total solids 
present in the first and second samples. The first sample con¬ 
tained 3,390, the second 1,431 pounds per acre-foot. The sample 
taken at the east end of the plot showed the same fact but much 
less markedly; the first sample containing 1,847, the second 1,641 
pounds per acre-foot. This difference is accounted for, I think, 
by the fact that we failed to get the first portion of the off-flow at 
the east end, while we succeeded in getting it at the north side. 
The decrease in the total salts carried in solution by such water is 
