584 
Journal of Agricultural Research 
Vol. V, No. 14 
requirement investigations, and were grown in the sealed pots already 
described (Briggs and Shantz, 1913, p. 9), which practically eliminate the 
direct loss of water from the soil. The pots contained about 115 kgm. 
of soil and were sufficiently large to enable the plants to make a normal 
growth, a factor of importance in transpiration measurements (PI. TV, 
figs. 1-2). Apart of the transpiration measurements were made within 
the screened inclosure (PL LIV, fig. 1) used in the water-requirement 
experiments to protect the plants from hail and wind storms. Other 
measurements were made outside the inclosure where the plants were 
freely exposed, with no protection whatever (PI. LIV, fig. 2). 
MEASUREMENT OF PHYSICAL, FACTORS 
Soear radiation. —The solar-radiation measurements were made 
automatically with a mechanical differential-telethermograph already 
described by one of the writers (Briggs, 1913). The instrument has two 
independent cylindrical bulbs and records only the difference in tempera¬ 
ture of the two bulbs. When used for measuring radiation intensity, 
one bulb is blackened and surrounded by a spherical glass envelope 
(PI. LIU). This is so exposed to the sun that the longer diameter of the 
bulb is normal to the sun's rays at midday. This bulb rises in tempera¬ 
ture until the rate at which energy is lost is equal to the rate at which it 
is received. The other bulb follows the temperature of the air within 
the instrument shelter, through which the wind blows freely. The 
instrument was calibrated by comparison with an Abbot silver-disk 
pyrheliometer (Abbot, 1911). Such comparison shows that the difference 
in temperature, as measured by the telethermograph, is very nearly pro¬ 
portional to the intensity of the solar radiation falling on a blackened 
surface normal to the ray, as measured by the pyrheliometer. In other 
words, the §cale is linear and the loss of energy conforms to Newton's law 
of cooling. While the telethermograph includes the sky radiation as well, 
the apparatus can be calibrated in terms of the solar radiation on bright 
days, since on clear days the ratio of sun to sky radiation appears to be 
fairly constant and the latter at the elevation of Akron (4,200 feet) is small 
compared with the direct radiation from the sun. A comparison of the 
telethermograph with the pyrheliometer, when the former is used for 
measuring radiation, is given in figure 1. 
The radiation data given in this paper are expressed in terms of differ¬ 
ential temperatures and the mean values are converted to calories per 
square centimeter per minute on a surface normal to the sun's rays. 1 The 
radiation is integrated for hourly periods so that zero radiation is not 
recorded until the hour following the hour interval during which the sun 
set, or preceding the hour interval during which the sun rose. 
1 The magnification of the differential sunshine instruments was not the same in 1912 and 1914. To 
convert to calories per square centimeter per minute multiply the differential temperatures in the 1912 
observations by 0.0335; and in the 1914 observations by 0.028. In the 19x4 observations the instruments 
were so adjusted as to give differential temperatures in degrees Fahrenheit. 
