( 892) 
The way of attacking the problem was 
as follows. Let an atmosphere of thickness 
t be irradiated by a surface SS, (fig. 3), 
which per unit emits a quantity of energy 
S within the limit of wave-lengths 2 and 
2 + dà, and uniformly distributed over all 
directions. Now ScHusreR begins with cal- 
culating the change which the total flow of 
Fig. 3 radiant energy suffers in a thin layer dz of 
that atmosphere. The layer receives from the left a quantity A per 
unit surface (which in general must be smaller than 5, although it 
includes, besides the radiation directly coming from .S,S,, also the 
radiation emitted by the part of the atmosphere lying between /S,S, 
and the layer dv). Of this quantity A the layer absorbs „Ade *), and 
scatters sAdr, the latter part not being lost as radiant energy of the 
given wave-length, but proceeding half to the right, half to the left. 
From the right side the layer receives a quantity of energy 5 per 
unit surface (composed of seattered light and proper radiation due 
to the outer part of the atmosphere); it absorbs z5dv and scatters 
sBdz, of which */,sBdx goes to the right and '/, sBde to the left. 
The layer also radiates energy in both directions, amounting to 
abide, if E represents the emission power of the black body within 
the chosen limit of wave lengths and at the temperature of the layer. 
Collecting these effects, one obtains the equations 
dA 
Fs (one ade te” 
av id 
dB Ean psd 
] —x(B— EE) + Fada . . . . . (15) 
aar 
If now the temperature and the composition of the atmosphere 
are supposed to be everywhere the same, so that /, #, and s may 
be considered as constants, A and B can be solved as functions of z. 
Let « be reckoned positive toward the right, and the origin of 
coordinates taken in the outer surface of the medium, then we shall 
find the emergent radiation R equal to the value which A takes for 
v=o, while at the same time we have 4—o0. Another condition 
is, that for c= — i, we have A= Db. 
Performing the calculations, ScHvsTER obtains 
-1) | am using here Scuusrer’s notation, and therefore must draw the reader's 
attention to the fact, that the above coefficient x is not quite the same as that 
An 
occurring in $2, S3 and § 4, but corresponds to the expression — . 7% of formula (12). 
5 7) 
