626 SCIENCE PROGRESS 



including the terms introduced by the new theory, and (2) from 

 the tables without these terms, is given by de Sitter. 



Observed. Theory. Difference (i). Difference (2). 



The agreement of the new theory is thus very satisfactory and 

 generally better than the old. 



The influence of the new terms introduced by Einstein's 

 theory into the theory of the moon's motion have also been 

 discussed by de Sitter. No results decisive either for or against 

 it are obtained. He finds a slight influence on the motions of 

 the lunar perigee and node of amount +i // '97 and +i"*9i 

 respectively per century. The observed values are : 



Brown, Cowell . . . d« =■■ + 14643536^2" 

 Newcomb .... = + 14643530" ±2'* 



Newcomb, Brown . . dSi = - 6967944" ±2* 



The new theoretical values are : 



d£> = +14643534" ±2" 

 dSi = - 6967939" ±2" 



The residuals 



.« = ( +2" ±3" 



X -4" ±3" 



A 8 = - 5" ± 3" 



are of the order of the probable error and thus give no infor- 

 mation, this being also the order of the terms added by the 

 new theory. When the theoretical motion can be more accur- 

 ately determined, some information as to the truth of the 

 theory may be provided in this manner. 



De Sitter also considers the effect of the gravitational field 

 of the fixed stars on the stellar spectral lines and shows that 

 the result will be to produce a slight systematic displacement of 

 the lines towards the violet. In this way it may be possible 

 to derive an approximate upper value for the total mass of 

 the stars within a given distance from the sun. This may 

 provide a rough indication of the relative numbers of the bright 

 and dark stars, since the latter contribute to the gravitational 



