The Ground Water. 
29 
§ 82. It is difficult to present this subject without reproducing 
all of the analyses representing the different sections of the plot, for 
they differ so much in character that one is not really representative 
ot the plot. The suggested difficulties are still greater than the 
simple lack of representativeness, for it suggests that the chemical 
reactions taking place within very limited areas of soil may be but 
partially or not at all comparable. This difference is made 
strikingly evident by the difference in the salts present in the water- 
soluble portions of the first and second two inches of these soils. 
Whether I have adopted the proper order of combination or not 
does not matter. I have adopted the same method of interpretation 
in all cases, which in itself may be an error, still it brings out sev¬ 
eral important and scarcely questionable differences. 
§ 83. I shall select section C for my present purpose, because 
it is less favorable to my presentation of this subject than B or D, 
and rather more favorable than section A. The reader who wishes 
to compare the results obtained for the other sections can find the 
analyses of the water-soluble portions of the soil in Bulletin 65, 
pages 36, 37 and 38. 
§ 84. The samples of soil were taken, one in May and the 
other in June. The sample of water was taken in June. It would 
have been better for the present purpose had they been taken at the 
same time as well as from the same place, but I have chosen these 
from the samples taken, as being the nearest together in the point of 
time of collecting. 
§ 85. The alkali which I use in this case was also collected in 
June, but nearer to well A than to well C. This, however, does not 
detract from its value for the purpose of comparison, for other sam¬ 
ples show that the differences in the alkali incrustations of this plot 
do not lie in the salts of which they are composed, but in their 
relative quantities. I have a sample taken nearer to well C, but it 
was taken in January during freezing weather, which, owing to the 
deportment of sodic sulfate at low temperatures, might make it less 
comparable than the one chosen. 
The arrangement of the analyses is evident. 
