IONISATION, DIFFUSION, ROTATION 
259 
/ 
in this range. (They had just completed their critical range in passing 
from Capella to the first star of the Table.) 
Fowler and Guggenheim comment on the very small effect of change 
of density on the ionisation, and give other examples of this. For Fe at 
5-6.10 6 degrees a change of density from 0-5 to 0-0004 alters /x only 2 per 
cent. This is an extreme case. The other extreme is represented by Br 
which undergoes L ionisation in this range, and p, increases 25 per cent. 
The slightness of dependence on density (i.e. electron density) justifies us 
in calculating the molecular weight for any element independently of 
other elements which may be present; for the degree of ionisation of the 
other elements affects the electron density by a factor which is evidently 
unimportant. 
Table 35. 
Average molecular weight in stars of the Main Series 
T= 26-36.10 6 181 
M 
P 
Molecular weight for 
Electrons retained by Iron 
(percentages) 
Oxygen 
Iron 
Silver 
1 
2 
3 
4 
2-13 
6-95 
1-97 
2-18 
2-40 
57 
37 
— 
— 
1-27 
15-2 
1-98 
2-21 
2-40 
40 
48 
10 
— 
0-75 
38-4 
2-00 
2-24 
2-40 
23 
53 
22 
— 
0-54 
71-6 
— 
2-25 
2-40 
15 
50 
31 
— 
0-36 
156 
2-06 
2-26 
2-40 
10 
44 
38 
6 
0-22 
392 
2-14 
2-28 
2-40 
9 
41 
41 
7-5 
181. Before leaving the subject of molecular weight reference should 
be made to “the correction for excluded volumes.” This is treated by 
Fowler and Guggenheim as a deviation from the laws of a perfect gas and 
discussed in that connection in their paper; but it is convenient here to 
amalgamate it with the molecular weight. Whilst the electrons and un 
excited ions can be treated as of infinitesimal volume in all stars except 
white dwarfs, the volume of the excited ions should be taken into account. 
Finite size of the atoms has the effect of increasing P for given p and T 
(in accordance with Van der Waals’ equation); this is equivalent to a 
decrease of the molecular weight or to an addition to the number of 
independent molecules. It is found that the volumes of the excited atoms 
have an effect on the pressure equivalent to the addition of phantom 
molecules in the proportion of f of a molecule per atom (or according to 
another, perhaps better, theory \ a molecule per atom). This constant 
limit is practically reached in all stars. Since it does not vary from star 
to star it can conveniently be taken account of in the molecular weight. 
17-2