THE OUTSIDE OF A STAR 
355 
/ 
or about 6 electrostatic units. The Stark effect of a steady field of 6 units 
resolves H^ into components extending over 0-7 A. The observed width 
of the line should be greater—say 2 to 3 A—since the Stark components 
will be shifting about over a wider range owing to the fluctuations of the 
field. The actual width of H p in stars corresponding to the assumed con 
ditions is about 15 A. 
I do not think that the fluctuating Stark effect is in itself sufficient 
to account for the great widths often observed. It seems likely that 
another point is involved. In the foregoing conditions the average speed 
of the electrons is 6-2.10 7 cm. per sec. so that the electron traverses a 
distance r 0 in 2*4.10 -13 secs, or 24 periods of the 4-quantum orbit. Some 
electrons will approach much closer and some will have greater speeds so 
that there will be a considerable amount of disturbance which changes 
completely in the course of 3 or 4 periods. Now the Stark effect represents 
the disturbance by a steady field, which during one half-revolution 
counteracts to a large extent the change of phase produced by it in the 
other half-revolution; the theory of the effect is based on the “adiabatic 
hypothesis” that the field is established slowly in comparison with the 
time of revolution. The quickly fluctuating field should give much larger 
effects, and it seems possible that the great width of the lines is produced 
in this way. 
247. Although the general flow of heat through the photosphere is to 
some extent dammed back by line absorption and emission, the main 
obstruction is likely to be continuous absorption and emission, just as in 
the deep interior. 
Our formula for the continuous absorption coefficient at temperatures 
of some million degrees was (158-2) 
The theory of continuous optical absorption at photospheric temperatures 
must be fundamentally the same as that of continuous X ray absorption 
at high interior temperatures; but we should scarcely expect the formulae 
derived under the simplifications permissible at high temperature to apply 
to the photosphere. Milne*, however, has found reason to believe that, 
whether by coincidence or by actual appropriateness, this extreme extra 
polation of (247-1) gives results not far from the truth. 
We apply (247-1) to a typical photospheric region with 
Continuous Absorption. 
/*№(i -f) 
(247-1). 
P 
T — 6000°, po = 200 dynes/cm. 2 
* Monthly Notices, 85, p. 768.