74 
Gaseous Stars 
[ch. Ill 
diameter 10 -8 cms. are replaced by crowds of free electrons of diameters about 
4 x 10 -13 , of nuclei whose diameters are less even than this and of atoms 
whose rings of electrons have nearly all been stripped away. 
Even at a density of 50,000 times that of water, the electrons and nuclei 
are at mean distances apart which are of the order of 10~ 10 cms., and this is 
some hundreds of times the probable diameter of both free electrons and 
nuclei. If then all atoms were stripped absolutely bare of electrons, so that 
stellar matter consisted solely of electrons and bare nuclei, it would seem 
reasonable to assume the ordinary gas-laws to hold, even at these monstrously 
high densities, at any rate as a first approximation. Anderson* has pointed 
out that the electron pressure in stellar interiors is that of a “degenerate” 
gasf, so that the pressure can in no case be given exactly by Boyle’s law, 
the divergence increasing from one of a few per cent, in ordinary stars to 
a factor of about 2 for Kruger B and Sirius B. 
We shall, however, find that, although the mean density of Sirius B may 
be 50,000, the density in its central regions is probably to be measured in 
millions, so that here Boyle’s law can hardly hold, even as an approximation. 
Furthermore, the temperatures at the centres of most stars are not adequate 
to break the atoms down to their ultimate constituents. Stellar matter 
consists in part of free electrons and perhaps some bare nuclei, but these are 
mixed with miniature atoms in which some rings of electrons remain to clear 
a space about the nucleus, and when such atoms are present it is not obvious 
without further investigation whether Boyle’s law will give a valid approxi 
mation to the pressure-density relation or not. 
For the moment we shall proceed on the supposition that the pressures in 
stellar interiors are those given by Boyle’s law, but we shall soon find it 
necessary to discard this supposition. For the two reservations just men 
tioned prove to be of the utmost importance in stellar physics, and probably 
dominate the whole dynamics of stellar interiors. 
The Pressure of Radiation. 
70. We have been led to picture the far interior of a star as consisting of 
a crowd of dissociated electrons, of bare nuclei and of atoms stripped of 
electrons almost down to their nuclei. The electrons move about from one 
broken atom to another, seldom staying with any one atom for long, since the 
majority of the atoms are either completely or almost completely bare of 
electrons. As we approach the surface, we come to regions of lower temperatures 
where the disintegration is far less complete, and here we find semi-formed 
atoms. Finally, close to the surface we come upon ionised atoms from which 
only a few of the outer electrons are missing, and possibly fully formed atoms. 
In the iH-type stars, the surface layers are so cool that even fully formed 
molecules are found, such as those of titanium oxide and magnesium hydride. 
* Zeitsch. f. Physik, l. (1928), p. 874. + E. Fermi, Zeitsch. f. Physik, xxxvi. (1926), p. 911.