SURVEY OF THE PROBLEM 
23 
5 inches thick. 
3sorbed in the 
if we compare 
experimental 
Is of X rays to 
inations of the 
100 found for 
A g 
12 
82 
260 
580 
gnitude as the 
bion of stellar 
t first appears, 
pidly with the 
in the column 
3 wave-length. 
*her than any 
values for, say, 
at the opacity 
perature, since 
rtional to the 
servation; the 
iffering widely 
pacity. 
md laboratory 
tot difficult to 
ilarity is the 
it is a natural 
; much affect 
s individually 
Table III. 
each contributing its share independently of the presence of the others. 
But are we dealing with the same kind of atoms in the two cases? It is 
a question of definition of the word atom. The atomic nucleus which we 
regard as characterising the element is unaltered, and for that reason we 
can say that the same elements are present in the stars as on the earth. 
But if the atom is taken to mean the whole system with its satellite 
electrons, then we must say that the atoms in the stars are not the same 
as those on the earth. Most of that circulating system is broken away. 
The “billiard-ball” atoms, about lO“ 8 cm. in radius, so familiar in our 
picture of terrestrial gases, do not exist in the stars; and in particular 
those properties of terrestrial gases which depend on the considerable size 
of the atoms (limit to compressibility) have no immediate application in 
the stars. In considering absorption and opacity the mutilation of the 
electron system of the atom is of vital importance, because it is just this 
system which contains the mechanism of absorption. 
In elements of moderate atomic weight the X ray absorption is per 
formed mainly by the ten innermost electrons which are classed in two 
groups, viz. 2 .¿-electrons and 8 ¿-electrons. The outer electrons of the 
system are concerned in absorption of greater wave-lengths, including 
visual light; the loss of these will not seriously alter the atom’s power of 
absorbing X rays. But the mutilation extends to the inner ten electrons 
which would otherwise have been active in absorbing the X rays in the 
star; and this, of course, reduces the absorbing power as compared with 
terrestrial atoms. The absorption is itself the cause of the breaking away 
of electrons, so that if the circumstances are such that any absorbing 
mechanism is called strongly into play that particular mechanism will be 
especially broken down. There is in fact a saturation effect. 
We may look at the difference between terrestrial and stellar absorption 
from another point of view. When aether waves fall on an atom they are 
not absorbed continuously. The atom lies quiet waiting a favourable 
chance and then suddenly swallows a whole quantum at once. The mouthful 
is too big for the atom’s digestion; consequently the atom bursts. One of 
the satellite electrons shoots off at great speed carrying away the surplus 
energy which the atom could not assimilate. The atom is now done for 
so far as that particular absorption trap is concerned, and it has to let 
the quanta fly past without interfering. Evidently this bursting could 
not go on continually unless there were some counter-process of repair. 
The atom must capture one of the free electrons flying by, inducing it to 
stay and heal the breach. The absorption trap is then set again and the 
atom is ready for another quantum. The emission of radiation occurs 
during this process of repair, the free electron having surplus energy which 
must be radiated. For each burst there must be a repair; so for each 
absorption there must be an emission. 
)