Stellav velocity distribution 



7 



Like Kempf, Rjsteen makes an exclusion of stars apparently belonging to 

 the same clusters and puts in their place only one star. From the remaining 42 

 stars he gets elements of the solar motion that do not in any high degree differ 

 from the results of Kempf. His value of the relative velocity of the sun is 

 17.6 km per sec. 



3. The successful measurement of stellar radial velocities at the Potsdam Obser- 

 vatory with the aid of photographic methods soon led to many attempts on the 

 same subject by other observatories, and it was in the first place at the Lick and the 

 Yerlces Observatories that the measurements of stellar spectra were put on the 

 working programs. The sensibility of the instruments of these observatories made 

 it not only possible to get radial velocity determinations of fainter stars but also 

 gave the measurements a precision that was much greater than could be obtained 

 at other observatories. 



It was chiefly at the Lick Observatory, under the direction of Prof. W. W. 

 Campbell, that the radial velocity measurements were persued and in the year 1901 

 Campbell had at his disposal 280 of the brightest stars mainly of north declination 

 whose velocities in the line of sight are well determined. 



In a paper published in Asirophysical Journal 1 he gives the following results 

 for the solar motion: 



p ... , «= 277° 30' ±4°.8, 



Position of the apex: 



p §= + 19° 58' +5°.9, 



The relative velocity of the sun: 19.89 ± 1.52 km per sec. 



These results are in good agreement with those later obtained. Campbell 

 makes, however, a further examination of his observations. The values for the apex 

 direction and the solar velocity being known, it is possible to make a special exa- 

 mination of the peculiar velocities of the stars in the line of sight. To this end he 

 corrects every observed radial velocity for the influence of the motion of the sun, 

 and of the thus remaining 280 peculiar velocities he determines the average value 

 without regarding the signs. This average he finds to be 17.05 km per sec. To 

 get an idea of the influence of spectral type upon average velocity he makes a 

 rough separation of the stars in two groups approximatively representing white and 

 red stars. The difference, however, in the average velocities of the two groups seems 

 hardly sufficient to justify any statement as to the effect of spectral type upon velocity. 



There seemed, on the contrary, to be a larger correlation between the visual 

 brightness and the velocity, and he says that the progression is in no way due to an 

 increase of probable error of the velocity determination with decreasing magnitude, 

 but must certainly have a real basis. As was later shown, however, these results 

 were exclusively due to the definition of the spectral types and the distribution of the 

 spectral types within the different stellar magnitudes. 



1 A preliminary determination of the motion of the stars by W. VV. Campbell. Astr, Phys. 

 Journal, XIII. p. 80. 



