104 



Mr. J. Norman Lockyer. 



Taking Arcturus as a representative star of the solar type,* it will 

 be seen that although the spectra of 7 Cygni and 6 Cephei resemble 

 it in showing a large number of dark lines, they differ considerably 

 from it in point of detail. I shall return to the subject of these 

 differences in a future paper. 



On the meteoritic hypothesis, then, the variables of this class are 

 to be regarded as meteor-swarms not yet completely condensed, and 

 this fact may prove of great importance in connexion with explana- 

 tions of the variability. 



(3.) There is a general ivedkening of the continuous spectrum as the 

 light of the star decreases. 



This is shown by the fact that longer exposures are required to 

 photograph the spectra when the variables are near the minima. 

 With equal exposures, the negatives taken near minima are very much 

 less dense than those taken at maximum, and even with the longest 

 practicable exposures, when the highest dispersion is employed, the 

 photographs obtained nearer minimum are relatively feeble. This has 

 been especially noted in the case of 8 Cephei, one photograph of 

 which was taken exactly at maximum and another only a short time 

 after minimum. 



(4.) There are no indications of bright -line radiation at the positions 

 occupied by the lines of hydrogen or helium at any part of the 

 period in the case of these variables. 



The lines of hydrogen are dark in all cases, and no changes of 

 intensity such as might be brought about by an increased radiation 

 at any time have been found in the photographs. A special search 

 has also been made for bright lines in the positions of the lines of 

 helium and the associated gases, but none have been traced. Since 

 all of these lines occur in the spectra of such stars as Bella/trix, this 

 part of the inquiry was easy — it was only necessary to make direct 

 comparison of the negatives. 



(5.) There is no visible doubling of the lines in any of the photographs. 



Unlike /3 Lyra?, the variables now under consideration show no 

 doubling of any of the lines at any part of the period, so far as the 

 Kensington photographs give any information on the point. The 

 photograph of 8 Cephei (see fig.) , taken exactly at maximum, shows 

 that there is no doubling at the time when such appearances might 

 be expected to be most marked. 



Hence, the conditions appear to be entirely different from those 

 which exist in the case of j3 Lyrae, and on the meteoritic hypothesis 

 the stars are less condensed. 



* ' Phil. Trans..' 1893, vol. 184, A, p. 699, 



