THE OUTSIDE OF A STAR 
353 
For the CN band which is most easily observable Av/c — 3-704. Hence 
Whnax. = 0-434 V T . 
At the centre of the sun’s disc we should expect the temperature of the 
reversing layer to be about 5200° (T 1 in (232-3)); at any rate it cannot be 
lower than T 0 = 4660°. The corresponding values of m are 
m max . = 31-2 and 29-6. 
Birge was not able to observe directly the maximum intensity in the 
solar spectrum, but by an indirect procedure (which is not very fully 
explained) he arrived at a much lower temperature 4000°*. The difference 
seems inexplicable; even granting that the region in the sun has no definite 
temperature (the radiation corresponding in quality to 5740° and in 
density to 4660°) it is difficult to see how any criterion could give a tempera 
ture lower than both. Further investigation seems to be required. 
Width of Absorption Lines. 
246. Whilst some lines in stellar spectra are extremely sharp, others 
may be of considerable width. In some stars the lines of the Balmer series 
of hydrogen extend over 30 Angstrom units on each side of the centre of 
the line. In the solar spectrum the H and K lines of Ca + extend 10 A on 
each sidef. 
Broadening of the lines can be produced by the disturbing effects of 
neighbouring atoms on the absorbing atom; these interfere with its perfect 
periodicity so that its quantum states are not entirely sharp. Broadening 
from this cause is usually referred to as pressure broadening though the 
name is now scarcely adequate. It is closely associated with the Stark 
effect of electric fields since the disturbing causes are electrical. 
Other conceivable causes of broadening are (a) Doppler effect of atomic 
velocities due to temperature, (b) Doppler effect of ascending and descending 
currents, (c) Doppler effect of rotation of the star, (d) Stark effect of regular 
electric fields, ( e ) Compton scattering by free electrons with different 
velocities, (/) Rayleigh scattering by atoms and ions, ( g ) great depth of 
the absorbing layer enhancing the importance of slight broadening due 
to other causes. Most of these have at one time or another been advocated 
as the main cause of width in stellar lines; but we do not think the argu 
ments will stand scrutiny. It is doubtful if they would have been proposed 
had not the operation of “pressure broadening” in stellar conditions been 
deemed inadequate. 
* Birge’s clue was that in a certain part of the band the relative intensities of 
the lines corresponded to that of a 4-ampere arc, and was at any rate intermediate 
between the furnace spectrum and the 13-ampere arc. The temperatures of these 
terrestrial sources of comparison were deduced from the observation of m max ,. 
f C. H. Payne, Stellar Atmospheres, p. 51. 
e 23 
/ 
J