42 BULLETIN 1059, U. S. DEPARTMENT OF AGRICULTURE. 
length, it is necessary to consider the constantly changing absorption 
by the earth's atmosphere. According to Very (75) , who cites Lang- 
ley (Professional Papers of the Signal Service, No. 15), there are 
" two different kinds of selective depletion which the solar rays suffer 
in traversing the earth's atmosphere. One kind is greatest for the 
rays of shorter wave length, and diminishes by perfectly regular 
gradations as one passes toward the longer waves of the infra-red. 
Its cause may be referred to selective reflection or diffraction of the 
shorter ether-waves by particles of excessive minuteness. The other 
kind of absorption produces irregular gaps or depressions in the 
spectral energy curve, which begin at the red end of the visible 
spectrum and grow in magnitude and frequency as the wave length 
increases. Researches by Abney and Festing, and by other investi- 
gators, have traced the majority of these depressions to the action of 
aqueous vapor." In the extreme infra-red there is shown to be al- 
most total absorption from this source. 
The light, principally of the shorter wave lengths, which is dif- 
fused by minute particles in the atmosphere, is not entirely lost, but 
may be measured as skylight, probably of greater wave length than 
the original direct rays. The infra-red rays which are so greatly 
absorbed by the vapor of the atmosphere, merely heat the upper at- 
mosphere, and to this extent, of course, are lost as solar radiation. 
2. Looking at the matter from another viewpoint, and accounting 
for the rather regular daily change in sunlight intensity at a given 
point on the earth's surface, Kimball (63) after showing the greater 
intensity of all wave lengths at midday when the light passes through 
minimum thickness of atmosphere, makes interesting comparisons 
of the total and luminous radiation under various circumstances. 
Radiation from an overcast sky is slightly richer, and radiation from 
a clear sky markedly richer, in luminous rays, than is direct sunlight. 
Direct sunlight decreases in luminous richness as the sun approaches 
the horizon. 
3. These few facts point to the uselessness of photometric methods, 
depending on the chemical action of rays of rather limited wave 
length, to measure the total radiation or any part of the radiation 
other than the few wave lengths which may be involved in the par- 
ticular reaction. Thus, for example, even if it were assumed that 
silver chloride was decomposed in proportion to the intensity of a 
given section of the spectrum, a certain reaction with silver chloride 
might be secured with other wave lengths varying through a very 
wide range. 
Again quoting Very (75), it is seen that photochemical processes 
are very complex and hazardous as a measure of energy : 
While luminous effects may be regarded as dependent on a certain photo- 
chemical action upon the retina, not all photochemical processes are equally 
