RECENT ADVANCES IN SCIENCE 15 



gathered. A first list of parallaxes of 500 stars has been pub- 

 lished in the Astroph. Journ. 46, 313, 1917- The stars chosen 

 were for the most part those whose parallaxes had been deter- 

 mined directly in order to provide a check on the results. The 

 agreement in general is surprisingly good and it is evident that 

 the new method — which at present is limited in application 

 to certain types of stars — is capable of turning out results by 

 the hundred, of a high degree of accuracy and invaluable for 

 statistical researches. 



We are now within sight of the time when accurate paral- 

 laxes of several thousand stars will be available. In 191 5 the 

 number of reliable determinations was only about 200, and 

 in 1900 only about 60. 



Einstein's Theory of Gravitation. — The astronomical conse- 

 quences of the new generalised relativity theory of gravitation 

 advanced by Einstein were referred to recently in these notes 

 (Science Progress, 11, 623, 191 7). It was mentioned that 

 the theory required the spectral lines of the sun to be displaced 

 to the red as compared with the positions of the corresponding 

 lines obtained from a terrestrial source. The displacement is 

 equivalent to the Doppler displacement produced by a radial 

 velocity of 0*63 km. per sec. The existence of this displacement 

 has been carefully investigated by St. John at the Mount 

 Wilson Observatory (Astroph. Journ. 46, 249, 191 7) : the 

 amount of the displacement in Angstrom units is only 0*008 A, 

 but so accurate are the Mount Wilson measurements that the 

 existence of systematic displacements of this amount can 

 be detected. St. John attacked the problem by choosing 

 lines in the band spectrum of cyanogen which show no pres- 

 sure shift ; displacements in the line of sight were eliminated 

 by observations of the same lines at the Sun's polar limb. 

 Wave-lengths of these lines were measured at the centre and 

 at the limb and also in the carbon arc in terms of identical 

 iron standards ; the solar wave-lengths were redetermined in 

 the Rowland system. The displacements sun minus arc were 

 determined at the centre directly and by three indirect 

 methods. The mean displacement was not in any case greater 

 than 0*002 A. The results of the investigation were that, 

 although there existed a limb effect not due to motion, in 

 which pressure, level, and line intensity appeared to be in- 

 volved in varying degrees, there was no evidence of a 



