Stellar velocity distribution 



25 



of class B stars at other observatories are in good accordance with those obtained 

 at Mount Hamilton and in Chile. The most probable explanation however seems to 

 him to be found in the atmospheric conditions of the stars. In another paper 1 

 he calls attention to the slight relative displacements towards red, found by Mr. 

 Jewell for many of the spectral lines in the solar spectrum. Analogous pheno- 

 mena may exist for the class B stars, with the difference, however, that on the sun 



Fig. 9. 



we observe the vapours of the heavier elements, whereas in class B stars we observe 

 brighter gases, principally helium and hydrogen. In this way however the large 

 values of K for the spectral classes K and M will not be explained. 



20. The assumption of the existence of a systematic constant K seems strongly 

 to influence the relative velocity of the sun, as well as the apex direction, at least 

 when K has a large value. It is however remarkable, that the change of the apex 

 direction, when K is put equal to zero, corresponds to an increase of the galactic 

 latitude, the galactic longitude being unaltered. 



1 Note on the evidences of rapid convection in Stellar atmospheres. Lick. Obs. Bull. VIII. 

 No. 257, page 82. 



Lunds Universitets Årsskrift. N. F. Afd. 2. Bd 11. i 



