74 bulletin 82. 
ago, i. e., that the soil retains potash salts more tenaciously than it 
does others. 
43. The drain waters, as indicated by such data as we have been 
able to gather, though we have not been able to study this subject as we 
desired, differ materially from the ground waters. They contain a 
smaller quantity of salts in solution, and are more uniform in this con¬ 
tent than the ground waters. The salts present stand in a different 
order, especially in regard to their relative quantities, sodic sulfate 
sometimes disappearing entirely. Calcic sulfate is uniformly first in 
quantity; magnesic second, sodic carbonate third, and sodic chlorid 
fourth, with sodic sulfate quite irregular, but usually less than the sodic 
chlorid. 
44. The first significance of these facts is that our drains benefit 
our lands by removing the surplus water, rather than the useless or 
deleterious salts, from the soils. This is by no means a small service. 
Indeed, it is the most important service to be rendered to nearly all of 
our alkalized land. Of the salts removed, the most injurious one, when 
present in sufficient quantity, is the sodic carbonate. Relative to the 
amount of this salt present in the drain and ground waters, a compari¬ 
son of the analyses of the drain waters with those given of ground wa¬ 
ters in this Bulletin, and also with those in Bulletin No. 72, pages 23-26, 
it will be seen that the grains per gallon remain quite constant. In 
other words, the sodic carbonate does not seem to be retained by the 
soil, or removed from solution by passing through it, while the sodic 
sulfate, or white alkali, is retained to a very marked extent. 
45. The only samples of drain and ground waters taken on the same 
date, are those taken April 17, 1900. The samples of ground water are 
unusual, as set forth in preceding paragraphs, but the features to which 
I wish to call attention are so bold that thev will not be hidden, or even 
distorted by these facts. In the ground waters we have 452 and 470 
grains respectively, in the drain water 114grains of total solids. The sodic 
carbonate in the ground waters amounts to 26 and 22 grains respectively 
in the drain water 23 grains. The range of this salt in the ground waters, 
given in this Bulletin, is from 10 to 23 grains per imperial gallon, and in 
those given in Bulletin No. 72, it is from 9 to 18 grains, while the range 
of the same salt in the drain waters given in this Bulletin is from 11 to 
23 grains. Returning to the samples of April 17, we have in the ground 
waters 76 and 115 grains of calcic sulfate per gallon, for the drain water 
45 grains. We have 152 and 137 grains magnesic sulfate per gallon for 
the ground water and 24 grains in the drain water. Still more marked 
than either of these is the case of the sodic sulfate, of which we have 
105 and 89 grains respectively in the ground waters, and 5 grains in the 
drain water. The sodic chlorid is also retained within the soil, but in a 
less degree than some of the other salts. The ground waters on this 
date, April 17, 1900, carried 56 and 64 grains respectively, while the drain 
water carried 7 grains per gallon, or one-eighth as much as one of the 
samples and one-ninth as much as the other. 
46. The analyses of the ground water before and after irrigation 
show that one of the effects of irrigation is to rather increase the rela¬ 
tive amount of sodic chlorid in the ground water, so that the above 
figures appear more favorable to my statement than the facts as they 
are found, under less extreme conditions, might appear. Reference to 
the analyses of ground water, given on pages 30-33 and 38-4'> of this 
Bulletin, and to those given on pages 21-26, Bulletin No. 72, will show 
that the sodic chlorid in the drain waters is less than in the ground 
waters, under the wide range of conditions represented by these numer¬ 
ous samples. Some of the analyses referred to, especially some of those 
in Bulletin No. 72, suggest very pointedly that the character of the soil 
has a decided influence upon these points; the indications being that the 
soil experimented with permitted the respective salts to pass through 
more freely than soils freer from alkali salts, and in better mechanical 
condition, would have done. 
