SURVEY OF THE PROBLEM 
3 
/ 
showing strongly 
tars; it is rather 
¡nsity favourable 
ng the possibility 
ssary to account 
ume that in the 
e not connected 
f fixed composi- 
lown wards. Its 
pressure; hence 
l entirely on the 
led; and these 
gravity holding 
'ed into it from 
ere must adjust 
'face conditions 
lurface and the 
a conventional 
er unit area; it 
arly significant 
e of the stellar 
ve must expect 
ofold sequence, 
iensional order, 
ional order was 
qjes. But the 
due to Adams 
ctra transverse 
riterion follows 
e parameter g ; 
one time sup- 
ial distribution 
T e and g into 
coincidence of 
3 . Consider now the connection between our three pairs of parameters 
—g and T e ; Draper Type and Absolute Magnitude criterion; M and R — 
any pair defining a unique state of the star. The connection of the spectral 
criteria with g and T e is a problem of great importance in which much 
recent progress has been made; but it is not a problem of the stellar 
interior and lies outside the main lines of our investigation. As regards 
the connection of g and T e with M and R, the connection of g needs no 
comment; the main question is, How is T e , or equivalently the rate of 
outflow of radiation, determined by the mass and radius of the star? 
That is the central problem of this book. Various branches of inquiry 
will diverge from it; but it supplies the continuous thread in the dis 
cussion, so long as we are studying the stellar interior. 
This is essentially a problem of the stellar interior and not of superficial 
conditions. The sun does not radiate 6.10 4 * * * * * 10 ergs per square centimetre per 
second because its photosphere is at 6000° C.; its photosphere is main 
tained at 6000° because 6.10 10 ergs are streaming through it. It is under 
the temperature gradient in the interior that the radiant stream gathers 
way; the surface layers cannot dam the flow since their capacity for 
storing energy is insignificant; they can only adjust themselves to let it 
through. Qualitatively the radiant stream is greatly transformed in 
passing through the last few thousand kilometres of the star, and the 
actual waves that spread through space are born in the photospheric 
layers; but quantitatively it is one continuous stream passing from the 
interior into outer space. 
The intensity of this outward flow of energy through the interior 
depends on two factors, the one helping and the other hindering. Heat 
flows from a higher to a lower temperature, and the cause of the flow 
within the star must be a gradually increasing temperature from the surface 
to the centre. The hindering factor is the obstruction opposed by matter 
to the transmission of this stream of heat. We shall find that in a star 
the heat is transmitted almost entirely by radiation, and the obstruction 
to the flow of radiation is the opacity or absorption coefficient of the stellar 
material. Our problem is, therefore, firstly to find the distribution of 
temperature inside a star so as to determine the temperature gradient 
urging the flow; secondly, to determine the opacity of matter under the 
physical conditions prevailing in the interior. 
when we con- 
by considering 
>y stars of the 
; third way of 
tes is obtained 
•ameters. 
4 . Here at the outset we must deal with a criticism urged by Nernst, 
Jeans and others. It has been argued that this procedure for calculating 
the outward flow of radiation is necessarily doomed to failure, because 
the star’s output of heat energy is determined by entirely different con 
siderations. The supply of heat replenishing that which the star radiates 
into space must come from the conversion of other forms of energy ; and 
1-2