
New York AGRICULTURAL EXPERIMENT STATION. 185 
contained in each cubic foot of the different plats during the four 
dryest weeks of the summer was for A 2.797 in., B 3.096 in., C 
3.170 in., D 3.275 in., and E 3.72 in., from which it appears that the 
excess of the other plats over A was for B .299 in., C .3787 in., D 
478 in. and E .923 in. 
Expressing these excesses in cubic inches and in liquid measure, 
to make them more intelligible, it appears that the excess for 
each cubic foot of soil was in the case of: 
B. 48.056 cubic inches, or very nearly 13 pints. 
C. 53.813 cubic inches, or nearly 1 quart. 
D. 68.832 cubic inches, or more than 1 quart. 
HE. 132.912 cubic inches, or more than 2 quarts. 
Expressing these excesses in barrels per acre, the number 
appears in the case of B 258 barrels, C 322 barrels, D 412 barrels 
and E 799 barrels. 
The true relations of these numbers appear more clearly in the 
annexed illustration, which represents the depth of water in inches 
retained in each series of plats. 
These figures clearly show that the amount of water retained 
in the soil in dry weather, by even a very shallow stirring of the 
surface, is by’no means inconsiderable. It should be remembered 
that they do not represent the total amounts of water thus retained. 
That is to say,the experiment does not show simply that B 
lost 258 barrels of water per acre less than A during the 
four weeks under consideration. Itshows rather that as the average 
of four observations, made during the four dryest periods of the 
summer, the soil of B was found to contain, incorporated within 
itself and available to plants, had plants been growing within 
reach of it, 258 barrels of water per acre more than that of A, 
while the other plats contained larger ‘amounts than this. We 
may infer, therefore, that) this excess of water was maintained 
throughout the periods of drought, despite the losses from evap- 
oration. | 
The figures also show clearly that the amount of water retained 
by the stirring of the surface increased with the depth of the 
stirring, but not in proportion to its depth. It would appear that 
the first half inch of stirring retained more water than any 
24 
