APPENDIX II 
399 
Molecular Weight. 
The numerical results are very sensitive to changes in the adopted molecular 
weight. In No. 7 this was taken to be 54. In consequence of arguments for 
strong ionisation urged convincingly by Jeans, it was reduced to 2 (or, having 
regard to correction No. 9, to 2-8) in No. 8. At first it was only intended that 
Nos. 7 and 8 should indicate the two limits between which the results must lie, 
but gradually the latter came to be regarded as approximately correct. The 
first thermodynamical study of ionisation in the stellar interior is 
23. John Eggert. Ueber den Dissoziationszustand der Fixsterngase. PhysiJcalische 
Zeitschrift, 20, p. 570 (1919). 
This appeared to indicate a molecular weight 3-3, or perhaps higher; and in some 
of the author’s subsequent papers values of 3-5,4, 4-5 have been used tentatively. 
It was shown by 
24. E. A. Milne. Statistical Equilibrium in relation to the Photoelectric Effect 
and its Application to the Determination of Absorption Coefficients. Phil. Mag. 
47, p. 209 (1924). 
that more refined calculation indicated much stronger ionisation, the same con 
clusion being reached by the author a few weeks later (No. 16). There has con 
sequently been a reaction to low values ¡x = 2-1 or 2-2. The latest calculations 
are given in 
25. R. H. Fowler and E. A. Guggenheim. Applications of Statistical Mechanics 
to determine the Properties of Matter in Stellar Interiors. Monthly Notices, 85, 
p. 939 (1925). 
but I understand that Mr Fowler is not yet satisfied and believes that in the 
smaller stars ¡x is to be increased appreciably. 
Absorption Coefficients. 
The following additional references are of fundamental importance— 
26. H. A. Kramers. On the Theory of X Ray Absorption and on the Continuous 
X Ray Spectrum. Phil. Mag. 46, p. 836 (1923). 
27. S. Rosseland. Note on the Absorption of Radiation within a Star. Monthly 
Notices, 84, p. 525 (1924). 
The first contains the physical theory of absorption here accepted as most 
satisfactory; the second points out the distinction between absorption and 
opacity which we have called “Rosseland’s correction.” Other papers are 
28. S. Rosseland. The Theory of the Stellar Absorption Coefficient. Astrophys. 
Journ. 61, p. 424 (1925). 
29. J. Woltjer, Junior. Line Absorption and Absorption Coefficients inside a 
Star. Bull. Astr. Inst. Netherlands, No. 82 (1925). 
30. E. A. Milne. The Stellar Absorption Coefficient. Monthly Notices, 85, p. 750 
(1925). 
Connected with the same subject is 
31. R. H. Fowler. On Statistical Equilibrium and the Mechanism of Ionisation 
by Electronic Impacts. Phil. Mag. 47, p. 257 (1924). 
I may here acknowledge indebtedness to Dr C. D. Ellis who guided me among 
the literature and helped me to a working knowledge of X-ray absorption, etc. 
when the astronomical researches led in that direction.