244 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1931 



distance, 34 km per sec. for 2,000 parsccs, and so on. This provides 

 what may ultimately prove to be a valuable means of finding the 

 mean distance of a class of objects when it is not determinable by 

 older methods; for if we measure the magnitude of the rotational 

 effect we can at once write down the corresponding distance. To 

 illustrate this I will refer to a remarkable investigation by Plaskett 

 and Pearce. 



Their research dealt with the radial velocities of about 250 stars 

 of the most distant type known. They wished to sort these into 

 groups according to distance; but since the stars were far beyond 

 the range of ordinary methods of distance determination this separa- 

 tion presented some difficulty. It is not much use to sort them accord- 

 ing to apparent brightness, because brightness is a poor criterion 

 of distance. The authors availed themselves of a method developed 

 recently by Otto Struve. We are looking at these stars through a thin 

 veil of cosmical cloud. The cloud leaves its mark on the light, pro- 

 ducing certain narrow absorption lines in the spectrum of the star. 

 If the absorption is intense it is a sign that we are looking at the star 

 through a great thickness of cloud — that the star is very remote. By 

 this criterion Plaskett and Pearce divided the stars into three groups 

 showing low, medium, and high absorption, respectively, which must 

 correspond presumably to small, medium, and great distance. 



In the following table the third column gives the magnitude of 

 the rotation term for each of the three groups and the fourth col- 

 umn gives the deduced distance (the proportion being 17 km per 

 sec. per 1,000 parsecs as already stated.) It will be seen that Struve's 

 criterion has been successful; or at least that Oort's and Struve's 

 methods of estimating distance (both of which must be regarded as 

 on trial) confirm one another. 



Turn now to the fifth and sixth columns, in which the same analysis 

 is applied not the stars but to the motions of the cosmical cloud. 

 The velocity of the cloud can be measured in the same way as that of 

 a star from the Doppler shift of the spectral lines which it absorbs ; 

 but, of course, our measurement refers not to the whole cloud but 

 to the particular part of the cloud responsible for the absorption 



