Apr. 9,1917 
Water-Retaining Capacity of Soil 
43 
were refilled with soil J, carrying 15.2 per cent of water, and allowed to 
stand for 185 days before opening. The soil was tamped in as usual, 
this being continued to the top in the case of one and to within 1 inch of 
the top in the case of the other, the last inch in which was filled with 
air-dry soil. This, not being tamped, formed a shallow mulch. Both 
cylinders were so situated that the sun could shine on them nearly all 
day long; but, as whitewash was used on the windows, the maximum daily 
temperatures were not much higher than those in the open air. Thus, the 
mean maximum temperature in the plant house from July 5 to the end 
of the month was 93 0 F., while that in the open air was 91 °. The corre¬ 
sponding average daily mean temperatures for the same period were 
84° and 79 0 , respectively. For the month of August the mean maximum 
temperature in the plant house was 84° and in the open air 75 °. While in 
the greenhouse the temperatures were somewhat higher, there was much 
less wind movement; and accordingly the soil in the cylinders was 
exposed to somewhat the same, or slightly less, drying influences that it 
would have experienced had it been in the open air but entirely protected 
from rainfall. 
During the first three months of the experiment the cylinder with the 
compact surface was daily examined for the presence of minute cracks, 
and both cylinders were examined for the presence of crevices around 
the walls. As soon as cracks or crevices appeared they were filled with 
dry soil from the surface, so that all water lost would have to pass through 
the surface layer, instead of part of it escaping through such ventilating 
fissures. As, long before the experiment was concluded, cracks and 
crevices had ceased to form, examinations were made much less fre¬ 
quently after the third month. 
The distribution of moisture found at the end of six months is shown 
in Table XI. The loss of water was greatest at the surface and least 
toward the bottom of the cylinders. All portions had suffered a loss of 
4 per cent or more. A uniform upward movement, similar to that 
mentioned in the case of the earlier experiments, is to be observed in the 
case of the portion below the first 12 inches. The soil below this depth 
became as dry as in the earlier experiment (Table XII), but the latter 
had been exposed less than half as long. It is probable that the lower 
portion of the soil in both experiments had practically ceased to lose 
water. The loss of water from the two cylinders was similar, averaging 
for each 5.7 per cent. The final moisture content of the portion of the 
column below the twelfth inch varied only from 10 per cent at the top 
to 11 at the bottom, with an average of 10.7, or 1.9 times the hygroscopic 
coefficient. 
78370°—17 - 2 
