8 



Walter Gyllenberg 



4. In the following years the systematic work on radial velocity determinations 

 was continued, especially at the Lieh, YerJces, Allegheny and Mount Wilson Observatories. 

 In many cases, where the velocity, due to a binary character, was found to be variable, 

 the velocity of the centre of mass was determined. 



In the year 1911, in two papers, »On the motion of the brighter class A stars» 1 

 and »Some peculiarities in the motion of the stars» 2 Campbell makes a complete 

 summary of the results up to that time drawn from the radial velocities of the stars. 

 In the first paper he discusses the motions of 225 class B stars mainly brighter 

 than magn. 5.0 whose velocities in the line of sight are determined. At the same 

 time he publishes the individual observations. From these stars he derives the 

 position of the apex and the relative velocity of the sun. Moreover he makes here 

 the discovery of the remarkable constant K the explanation of which seems to him to 

 be found in certain atmospheric conditions of the stars of this class. In the same paper 

 he points out the very small avarage velocity of the class B stars. Frost and 

 Adams had however as early as 1904 called attention to the small peculiar velocities 

 of these stars. 



Observing the dependence of spectral classes upon radial peculiar velocity, lie 

 makes a new investigation of the correlation between visual magnitude, spectral 

 type and velocity. Dividing the stars into two spectral classes, Secchi's types I and II, 

 he gets the following table, reprinted from his paper. 



Average velocities. 



Vis. Mag. 



Type I 



Type II 



Types I 

 and II 





No. 



V km. 



No. 



V km. 



No. 



T km. 



Above 1.50 



1.51 to 2.50 



2.51 to 3.50 



3.51 to 4.50 

 "4.51 to 5.50 . 

 5.51 to (5.50 . . 

 Below 6.50 



9 

 26 

 40 

 116 

 126 

 9 

 4 



10.5 

 9.2 

 9.4 

 8.9 

 11.3 

 19.1 

 11.3 



8 

 22 

 74 

 239 

 336 

 18 

 8 



16.8 

 12.3 

 15.0 

 15.1 

 15.1 



18.5 

 12.5 



17 

 48 

 114 

 355 

 462 

 27 

 12 



12.2 

 10.7 



13.0 

 13.0 

 14.1 

 18.7 

 12.1 



Means 



330 



10.25 



704 



15.08 



1034 



13.51 



The influence of spectral class upon velocity was shown more clearly when the 

 classification of the stars was made in greater detail. The strong correlation between 

 magnitude and velocity is here quite obliterated. 



In order to test the two-drift theory of Kapteyn, Campbell computes the 

 average residual velocity in zones at different distances from the vertex found by 

 Kapteyn a — 93 u S = -f- 12°. Concerning the class B stars the difference of the 

 average speed in the zones seems to give no evidence for the existence of a vertex. 



1 Lick. Obs. Bull. VI. 195. p. 101. 

 - Lick. Obs. Bull. VI. 1%. p. 125. 



