THE SOURCE OF STELLAR ENERGY 
293 
of hydrogen and helium, and most convincingly by F. W. Aston’s measure 
ments with the mass-spectrograph. The formation of helium from hydrogen 
has thus involved a loss of 0-8 per cent, of the mass; the corresponding 
energy must have been set free during the process of combination. There 
can be no doubt that the close approach of the electrons to the protons in 
the helium nucleus makes their mutual electrostatic energy less than in a 
state of infinite separation; and it is this loss of field-energy which is 
betrayed by the measurements of mass. After the first loss of -008 of the 
mass when the proton enters into the helium nucleus, the changes of energy 
in building higher nuclei appear to be much less significant. Aston has 
found another deviation from the “whole number rule” in the atomic 
weight of tin; but in proportion to the mass involved the liberation of 
energy is on a smaller scale. 
It appears then that not more than 1 per cent, of the mass is released 
as free energy by the processes I (6), so that if this is the main source of 
stellar energy the extreme time-scale is divided by 100, and the life of the 
sun (past and future) is limited to 1-5.10 11 years. (We are no longer 
troubled by the possibility of a large change in the rate of radiation since 
the mass changes no more than 1 per cent.) But this limit would only be 
attained if the sun originally consisted wholly of hydrogen. An initial 
proportion of 7 per cent, of hydrogen is necessary for a life of 10 10 years. 
We should be reluctant to admit a greater proportion even in the earliest 
stars*. The time-scale is thus rather cramped, but we cannot definitely 
say that it is insufficient. 
So far as we know, the processes I (a) give much less energy. To 
maintain a star’s energy by the breaking down of elements it is necessary 
to postulate elements of high radio-activity which are not known to 
terrestrial experience. 
By the third process involving destruction of protons and electrons the 
whole of the energy might be liberated and the extreme time-scale reached. 
This hypothesis supposes that when a proton and electron meet they may 
under exceptional circumstances coalesce; their positive and negative 
charges cancel and nothing is left but the energy which, released from all 
constraint, spreads out through the aether as an electromagnetic ripple. 
Or, instead of considering the two charges, we may fix attention on the 
field of force between them which involves something of the nature of a 
tube of discontinuity in the aether; this tube might slip back, healing the 
discontinuity, and at the same time starting a wave of radiant energy. 
If this last source of energy is operating the matter of the star will 
gradually disappear. The star burns away its mass. 
* Cf. § 169, where a mixture of 15 hydrogen atoms to 1 iron atom, or 21 per cent, 
of hydrogen by mass was considered with the idea of accounting for the opacity of 
Capella.