530 SCIENCE PROGRESS 



potential for giants is small compared with that for a dwarf 

 star, and consequently that it is highly unlikely that there 

 are any stars for which measurement of the Einstein effect 

 is at all accessible as compared with the Sun. 



The Spectroscopic Determination of Stellar Parallaxes.— 

 Our knowledge of stellar parallaxes has been considerably 

 revolutionised of late years by the spectroscopic methods of 

 determination which have been developed mainly by Adams 

 and others at the Mount Wilson Observatory. The method, 

 which has been previously explained in these notes, depends 

 on determining the absolute magnitude of the star from 

 measurements of the relative intensities of certain lines in its 

 spectrum which are particularly sensitive to luminosity. The 

 method as originally developed was restricted to stars which 

 are redder than those of type A, as the A-type stars do not 

 show the large variations in the intensities of the lines which 

 appear in the redder stars. On the other hand, it is found 

 that they show considerable differences in the appearances 

 of the lines, which are sharp in some stars and diffuse in others. 

 This characteristic has been made the basis of a method of 

 obtaining the parallaxes of A-type stars by Adams and 

 Joy {Astroph. Jour., 56, 242, 1922). Using stars whose absolute 

 magnitudes are accurately known, it is found that there is 

 a good correlation between magnitude and spectrum, the 

 magnitude being brighter the earlier the spectral type and 

 brighter for stars having spectra with sharp lines than for 

 those having spectra with diffuse lines. Extending the method 

 to all A-type stars of known spectral type and parallax, it 

 was found that there was an average difference of only 

 ^h o''-oo77 for 82 parallaxes determined by group motion, 

 and of diO"'Oi3i for io4 trigonometric parallaxes. The 

 method is to be further developed and refined, but mean- 

 while a preliminary list of the parallaxes of 544 stars of 

 types B7 to F,2 is given by Adams and Joy in the paper 

 referred to. 



The preliminary results of an investigation by Edwards at 

 the Norman Lockyer Observatory of the spectroscopic parallaxes 

 of B-type stars is given in M.N., R.A.S., 13, 47, 1922. In this 

 investigation the intensities of all measurable lines were referred 

 to the average intensity of the hydrogen lines as a standard. 

 In this way a certain number of lines were found whose inten- 

 sities varied with either type or absolute magnitude. Such 

 lines were compared with neighbouring unaffected lines by 

 means of a wedge photometer and the variations were stan- 

 dardised by using stars of known absolute magnitude. In 

 the paper is given a preliminary list of 49 parallaxes of stars 

 of B-type. It is thus seen that with these extensions of the 



