184 REPORT OF THE HORTICULTURIST OF THE 
plats. The bottom line of the table shows that, on the average, 
this gradation was uniform and marked. 
The influence of surface treatment assumes its chief importance 
in time of drought. In periods of abundant rainfall, it matters 
little to the farmer whether soil stirred one-half inch or four 
inches deep contains the more moisture. But when his crops are 
suffering from a dearth of water, he would gladly know how he 
may best mitigate the effects of the drought. 
If we observe the average moisture content of all the plats 
during the different weeks, as given in the right hand column of 
the table, it appears that the four dates showing ‘the lowest 
percentages were June 19, July 31, August 8eand September 4. 
During the week preceding each of these dates, then, if at all, the 
crops must have been suffering from drought. 
In order to avoid needless repetition of words, the plats will be 
designated by the letters applied to them at the head of the 
columns in the table. Averaging the moisture percentages of the 
different plats for the four dates above mentioned, we have for 
A 14.55, B 15.625, C 15.925, D 16.575, and E 18.45. 
If now we compute the excess of moisture in the other plats 
over A, we find that the per centis for B 1.075, C 1.875, D 2.025 
and E 3.9, from which it appears that keeping the surface stirred 
one-half inch deep retained 1.075 per cent of water that would 
otherwise have evaporated; that keeping it stirred two inches 
deep retained 1.375 per cent, four inches deep 2.025 per cent, and 
keeping it mulched one inch deep retained 3.9 per cent of water 
that would otherwise have been lost by evaporation. These 
differences may at first thought seem slight, but’ when we recall 
that the percentages are calculated upon the weight of the soil 
itself, their significance more readily appears, as computations 
given later will show. 
On September 22, an exact cubic foot of soil was taken up on 
opposite sides of the main plat devoted to this experiment. By 
drying generous samples from these measured bulks of earth, it 
was ascertained that the dry weight of a cubic foot of the soil under 
experiment was approximately 89.7 pounds. Calculating on this 
basis, and reducing the water contents to depth inches, as we 
measure rainfall, it appears that the average amounts of water 
