372 
DIFFUSE MATTER IN SPACE 
for wave-length À is equal to that of equilibrium radiation at temperature 
T K . The following results are found— 
The source of the radiation was taken to be as follows—5 per cent, from 
stars at 18,000°, 10 per cent. 12,000°, 20 per cent. 9,000°, 40 per cent. 
6 , 000 °, 25 per cent. 3,000°. The total density was taken at the round figure 
10~ 12 ergs per cu. cm. 
Suppose, for example, that we are dealing with material capable of 
absorbing and emitting only in wave-length 600. It is unaffected by the 
presence or absence of radiation of other wave-lengths. It therefore 
behaves as though it were in a field of equilibrium radiation of temperature 
T K = 4707° and takes up this temperature. This is an idealised illustration, 
and in actual matter no one process of energy transfer could be isolated 
so completely; indeed the notion of temperature is scarcely applicable 
unless a more general exchange is occurring. In natural material there 
will be a number of absorption and emission processes each striving to 
bring the temperature to the T x corresponding to the wave-length of the 
radiation concerned in it; and it is not possible to predict the result of the 
conflict without careful inquiry. 
257. Apart from direct astronomical evidence it is unlikely on general 
grounds that interstellar space is entirely void. Matter may escape from 
stars by radio-active emission, by radiation pressure or by the ordinary 
loss of high-speed molecules. Terrestrial magnetic storms are usually 
ascribed to some kind of corpuscular emission from the sun. At the same 
time space is continually being swept by the passage of stars which will 
pick up the atoms lying in or near their tracks; but it appears that the 
spring-cleaning of space takes at least 10 16 years so that it is unlikely that 
a steady balance of gain and loss has yet been reached. 
To fix ideas we shall consider a density of 1 atom per cubic centimetre. 
This density is probably about the maximum admissible, but it is more 
instructive to discuss an upper rather than a lower limit. We take a mean 
atomic weight 10 , since perhaps light elements are likely to preponderate. 
The mass in a sphere of 5 parsecs radius is then 128 x ©. This volume 
would contain about 50 luminous stars, most of them of mass less than \Q. 
Thus the assumed density gives a greater proportion of diffuse matter 
Table 48. 
Equivalent Temperatures of Radiation in Space. 
600 Â 
2000 
4000 
6000 
T K 
4707° 
1750 
967 
690