1894.] The Spectrum Changes in /3 Lyrcv. 281 



The greatest separation of the dark lines occurs about the time of 

 second maximum, and the relative velocity, as determined by 

 measurements of three of the doubles in the photograph of August 

 24, 1893, is that stated above. The individual measurements are as 

 follows : 



H>y = 155 miles per second. 



H = 154 

 X 4025 =158 



5. One of the dark line components bears a strong resemblance to Rigel 

 and the other to Bellatrix. 



The spectra of the two components can readily be separated, for 

 the reason that only lines common to both will be doubled. Among 

 these are. the lines of hydrogen. Lines special to either component 

 are always single, and they retain the same relative positions with 

 respect to one group of hydrogen lines throughout the period. 



In Plate 2 photographs are given to facilitate an analysis of the 

 compound dark line spectrum. At the bottom of the diagram is a 

 reproduction of a photograph taken near the time of second maximum 

 (August 24, 1893), and the spectra of Rigel and Bellatrix are in- 

 cluded in the same plate. The compound character of the dark line 

 spectrum of j3 Lyrse at this time is shown by the fact that one group 

 of lines corresponds very closely with those which appear in the 

 spectrum of Rigel, and when these are subtracted from the whole 

 spectrum, a spectrum closely resembling that of Bellatrix remains ; 

 the latter spectrum being displaced in this photograph to the more 

 refrangible side, as shown by the short lines drawn beneath the 

 spectrum. The resemblance of the two components to Rigel and 

 Bellatrix respectively, the spectra of which I have described in a 

 previous paper,* is further shown by the following tabular com- 

 parison, the two dark line components of ft Lyrse being called R and 

 B respectively (p. 282). 



It is not intended to suggest that the spectra of the two dark line 

 components are quite identical with those of Rigel and Bellatrix. 

 These are simply the best known stars which they most closely re- 

 semble, and the similarity is pointed out as an indication that we 

 have not to deal with bodies of an unfamiliar type. Throughout the 

 paper I shall refer to the two components as R and B respectively. 



The conditions at first maximum, as shown in Plate 1, are not so 

 simple as those at second maximum, though there is evidence to 

 show that at this point of the light curve the component B is reced- 

 ing with respect to R. As will be seen on reference to the photo- 

 graph of March 13, 1894, the hydrogen lines are broadened, and the 



* ' Phil. Trans.,' 1893 (A), vol. 184, p. 693. 



