H9-122] The Chemistry of the Stars 133 
More generally any stripping of the electrons from the atoms of stellar- 
matter ought to inhibit the energy-generating capacity of the matter, since 
it increases the number of free electrons which are immune from annihilation. 
Broadly speaking, the atoms near the centre of a star are more highly ionised 
than those in its outer regions, and this must produce a tendency to inhibit 
the generation of energy in the central regions of a star and to throw the 
generation of energy into the star's outer layers. Other tendencies may of 
course operate to neutralise this, wholly or in part. We shall return to this 
question later. 
121. The conclusion reached in § 115 that the rate of annihilation of 
matter is unaffected by changes of density and temperature, was based on 
the supposition that the matter to be annihilated retained the same consti 
tution throughout all changes of temperature and density. Our conclusion 
implied, for instance, that a mass of matter in which each atom is ionised 
down to its A-ring but not below, will generate energy at precisely the same 
rate no matter how its temperature and density are changed, so long as these 
changes do not affect the ionisation of the matter. As soon as the ionisation 
is affected, the problem of the dependence of energy-generation on tempera 
ture and density becomes identical with that of the dependence of ionisation 
on temperature and density. 
An increase of temperature increases the degree of ionisation of stellar 
atoms because the quanta become more energetic; this in turn inhibits the 
generation of energy. On the other hand, an increase of density decreases the 
degree of ionisation because there are now fewer quanta per atom, and this 
in turn increases the generation of energy. Thus the rate of generation of 
energy per unit mass must decrease with temperature and increase with 
density. We have tentatively supposed this rate to be proportional to p a T&; 
it now appears that ¡3 must be negative and a positive. But a brief calcula 
tion, based on the known formulae connecting ionisation with temperature and 
density shews that a and /3 are both comparatively small. To a preliminary 
approximation which is adequate for most purposes, we may disregard both 
a and /3, and suppose the rate of generation of energv to be a constant for a 
given type of matter. 
122. Stellar radiation must either orginate in types of matter known tp 
us on earth or else in other and unknown types. When once it is accepted 
that high temperature and density can do nothing or almost nothing to 
accelerate the generation of radiation by ordinary matter, it becomes clear 
that stellar radiation cannot originate in types of matter known to us on 
earth. Terrestrial matter as we have seen (§ 104) generates almost no energy, 
and as increasing its temperature to stellar temperatures and decreasing its 
density to stellar densities could not increase its generating capacity, we 
must conclude that other types of matter are responsible for the radiation 
J 
9