28 
Research Bulletin Xo. 3 
In the 1911 experiment conditions were much more unfavor- 
able on account of the extremely high temperatures prevailing out 
of doors. In that year the ventilation of the greenhouse had 
been so much improved that for ordinary seasons it might have 
proved adequate, but during the very hot weather of June and 
July the temperature out of doors on many days rose above 
100° F., while on July 5 it rose to 110° F. On that day it reached 
130° F. in the greenhouse. This temperature, which did not last 
long, was not sufficiently high to injure well-watered plants in 
the same greenhouse. 
Suspecting the correctness of the remarkably high tempera- 
tures recorded by the maximum thermometer, it was tested at 
various times, both in 1910 and 1911, by placing, during the 
hottest part of the day, a number of accurate thermometers in 
various parts of the greenhouse and protecting them all both from 
the direct rays of the sun and from the heat reflected by the brick 
walls of the greenhouse. However, nothing was found to justify 
any assumption of inaccuracy in the temperature data. The 
high maximum temperatures during the early part of each sea 
son were due to the presence of the steam-main mentioned above 
DETERMINATION OF THE SOIL MOISTURE. 
Each 6-foot cylinder was opened as soon as convenient after 
removing the last of the plants from it. About two-thirds of 
the cylinders were opened on the same day that the last plant 
was removed, but with the others there was an interval varying 
from 2 to 16 days. It is probable that no appreciable amount of 
moisture was lost from below the first foot during this interval, 
and that the moisture conditions were practically the same at 
the time the cylinders were opened as they were when the last 
plant died and was removed. 1 It is also improbable that any con- 
siderable loss of moisture occurred even from the first foot, as in 
all the cylinders bearing plants the soil of the surface foot was 
already very dry and was underlaid by at least one foot of com- 
paratively dry subsoil. 
In order to remove the soil from a six-foot cylinder for mois- 
ture determinations the cylinder was placed on a table and split 
lengthwise by shears, 3 inches at a time. ( Figure 1, &.) The 
soil from this section was quickly and thoroly mixed and a sam- 
ple for the moisture determination placed in a pint jar which was 
then sealed. After the samples had been secured from the whole 
of the cylinder, moisture determinations were made, using 100- or 
1 Alway, F. J., and Clark, V. L. A Study of the Movement of Water in 
a Uniform Soil under Artificial Conditions. Twenty-fifth Annual Report of 
the Nebraska Experiment Station, 1912, p. 255. 
