620 
on the other hand, for large values of 2, an intermediate temperature 
would be found lying more to the side of 1500°. 
(The two imagined intermediate curves of radiation of black 
bodies have been drawn dotted, so that each of them again 
embraces the same area). 
The temperatures caleulated for different values of 2 would have 
been still more divergent, if I had not been the original energy 
curve of a black body, but had presented a much greater slope 
towards the violet side on account of molecular scattering. 
Though for the sun everything is of course much more compli- 
cated than in these examples, the conclusion. remains valid that 
T must be found dependent on 4, if our supposition should be 
justified that every layer radiates as a black body. But though 
this hypothesis accounts to a certain extent for the variation 
of the found values of 7’ with 2, and is preferable in so far to the 
undoubtedly untenable supposition of Deranr and others, that the 
radiation of the sun would issue from one single absolutely black 
photospheric surface — yet the hypothesis of the “partial photo- 
spheres” cannot be considered either as satisfactory. 
Until by other means, some insight has been obtained into the 
power of emission of the successive layers of the sun’s mass, and 
the degree in which they scatter and absorb the different kinds 
of light, hardly anything can be derived from the distribution of 
energy in the solar spectrum concerning temperatures on the sun. 
The conception “effective temperature of the sun” has little value. 
This temperature varies in fact greatly according to the way in 
which it is defined, and none of the definitions warrant in any way, 
that by means of them an approximation is found of temperatures 
that actually prevail on the sun. 
