374 SCIENCE PROGRESS 



maximum frequency down to that observed : investigation of 

 this point shows that, without any great impossibility, this 

 might happen even for negatively charged particles, and that 

 the belt required for this purpose will not exert an appreciable 

 magnetic force on the earth's surface. Further investigation 

 may be expected to throw more light on this point. 



The Temperature of Space. — It is not generally realised that 

 the notion of temperature in free space has in itself no meaning. 

 In order to make the matter definite we must suppose a small 

 test body introduced, and we can then define the temperature 

 of free space as the temperature, of thermal equilibrium of this 

 body, in accordance with our usual method of defining tempera- 

 ture. M. Ch. Fabry, in Astroph. Joum. 45, 269, 191 7, points 

 out that matters are not so simple as they appear, for the 

 temperature attained by the test body will depend on its 

 emissive and absorptive powers at various wave-lengths. The 

 temperature attained may vary enormously, and be either 

 lower or higher than that attained by a black body. If the 

 body absorbs only long wave-lengths its temperature will be 

 low ; if only short wave-lengths, it will be high. Thus, sup- 

 posing the radiation is produced by the sun (taken as equivalent 

 to a black body at temperature 6,ooo°), and that the testing- 

 body is at the earth's distance and has one absorption band, 

 then the temperature attained by it will be 1980 if the absorp- 

 tion band has wave-length 0*4 microns, and only 130 if the 

 wave-length is 10 microns. It is, therefore, quite possible that 

 gaseous masses may have selective absorption for short wave- 

 lengths and will therefore reach a high temperature. The 

 high temperature found by M. Fabry for the Orion nebula may 

 thus be explained, as also may the emission spectra of comets' 

 tails. 



The Dependence of Stellar Motions on Absolute Magnitudes. — 

 A paper of the highest importance by W. S. Adams and G. 

 Stromberg dealing with this subject has appeared in the Astroph. 

 Joum. 45, 293, 191 7. It is well known that stars of large proper- 

 motion have high velocities in space. This may be due either 

 (1) to the fainter stars moving more quickly than the brighter, 

 or (2) to the nearer stars moving more quickly than those more 

 distant. These two alternatives are investigated with the aid 

 of hypothetical parallaxes determined by Adams's method from 

 the spectral characteristics. It is found, in the first place, that 



