[CALLENDAR] OBSERVATIONS OF SOIL TEMPERATURES 69 
thermometers between 2 and 3 p.m. The surface temperature, however, 
is mainly dependent upon sunshine and radiation. Its variations are the 
cause rather than the effect of the variations of atmospheric temperature. 
The dependence of the range of surface temperature on radiation is illus- 
trated by the diminished amplitude of the curves on Wednesday, May 
8th. The night was cloudy, after a brief thunderstorm, and the sun did 
not come out strongly till about 1 p.m. The daily range of surface tem- 
perature was reduced from 18° to 8° F. 
The curves show, as might naturally be expected, that the rate of 
rise of surface temperature due to sunshine, at this time of year is much 
more‘rapid than the rate of fall at night. The maximum in the case of 
the 4-inch thermometer occurs at about 6 p.m., and the minimum at 
about 8 a.m. ; the rise occupying 10 hours and the fall 14. 
Fig. A shows the diurnal variations, for the same periodyon a slightly 
larger scale, of the 4-inch, 10-inch and 20-inch thermometers. The 
curves illustrate, more clearly than the previous example, the retardation 
and diminution of amplitude of the waves of heat as they are propagated 
downwards into the soil. They illustrate, in addition, the changes of 
form which these waves undergo. The curve of temperature variation 
at the surface is not a simple harmonic or sine curve, but must be 
regarded as made up of a series of such simple curves, which are propa- 
gated according to the laws of conduction of heat at different rates and 
with different decrements. Ata depth of 10 inches the form of the wave 
is considerably modified. The fall is now more rapid than the rise, and 
the crest of the wave shows, in some cases, signs of breaking into two. 
At a depth of 20 inches the form of the wave is very considerably affected 
by the variations of two or three preceding days. For instance, the 
small surface variation of the 8th, combined with the large variation of 
the 9th, produces a double wave in the curve on the 9th and i0th at a 
depth of 20 inches. To attempt, however, to analyse these changes com- 
pletely would require a much more extended series of observations. 
RECORD OF RADIATION. 
The amplitude of the diurnal variation of the 4-inch thermometer 
may probably, as we have seen, be made to give a very convenient mea- 
sure of the mean clearness of the atmosphere, or the intensity of radia- 
tion, for each day and night. It would not be difficult from a sufficiently 
long continued series of observations to disentangle the probably small 
effect of changes of air temperature from those of radiation pure and 
simple. 
In a similar manner, the daily observations of the 20-inch thermo- 
meter, or of one buried at a convenient depth at which the amplitude of 
the diurnal wave begins to be negligible, might be made to give a con- 
veniently integrated record of the rate at which the soil is gaining or 
