594 DISCUSSION ON THE 
very slowly falling intensity with increasing frequency, instead of a fairly 
sharp, well-marked band such as experiment reveals. 
No other known mechanism can soften an original beam of photons of 
the gamma ray type. If such a beam is in the least inhomogeneous, our 
universal experience is that it is hardened, not softened, by passage through 
matter. If, on the other hand, the original beam is monochromatic, 
it is also at first apparently hardened in the process of getting into 
equilibrium with its secondaries, and when that condition has been 
attained it has regained its original absorption coefficient. If annihilation 
rays maintain the temperatures of the stars at all, the mechanism by which 
they do it is not that of a continuous degradation of wave-length until the 
wave-lengths corresponding to heat and light waves are reached. The 
process is rather that the penetrating power of the beam of photons is 
maintained wntil all of these photons have been picked off by Compton 
encounters and complete absorption of the beam has thus been brought 
about, the temperature of the matter traversed having been thus slowly 
raised by this straight attrition process. Our third fact, then, of a banded 
structure in the particular region in which the chief cosmic ray band is 
found, appears to me to be the last arrow that pierces the heart of the 
already twice fatally wounded annihilation hypothesis of their origin. 
Also, I should like to present one more reason why the hypothesis of 
cosmic electrical fields as an agency for directly imparting energies of from 
25,000,000 volts to, say, 400,000,000 volts to electrically charged particles 
cannot be admitted. It is not only that the most fantastic assumptions 
would have to be made to make such fields produce cosmic ray bands of 
the type observed, but also, more than that, to make fields of any such 
intensities symmetrical with respect to the earth and thus account for 
the uniformity of distribution of cosmic rays would involve, as it seems to 
me, something very like a return to the geocentric theory of the universe— 
a return scarcely acceptable to any scientific worker who has lived since 
A.D. 1500. 
With cosmic electrical fields and the annihilation of protons both 
completely unacceptable, what is then left to furnish the energy, first, of 
the great cosmic ray band which carries 90 per cent. of the energy of 
these rays? The answer seems to me to be as follows :—If the Einstein 
equation H=mce’, and the actual facts of isotopes, as accurately worked 
out first by Aston, are taken as guides, then this answer is unambiguous. 
But first, how dependable is this Einstein relation? Note first that it is 
a purely thermodynamic equation, stating merely energy relations. In 
using it, therefore, we are completely independent of any assumption as 
to the nature of the cosmic rays. Whether they are photons or high- 
speed charged particles is, so far as it is concerned, quite immaterial. 
Again, as to its dependability, I think that most physicists would say that 
it is about as safe a guide as any theoretical equation which we now have — 
in physics. It not only rests on excellent theoretical foundations, but 
also it has predicted quantitative relations which have stood the tests of 
all the careful experiments which we have as yet been able to apply to it. 
It states with entire definiteness that there is no atomic transformation 
whatever that can furnish the necessary energy except an atom-building 
process. 
| 
: 
: 
| 
| 
