On the Classification of Stars of the 8 Cephei Class. 451. 



Reference to the Old Evidence of Two Series. 



In previous communications to the Royal Society I have stated the 

 evidence which shows that in stars of Group II, such as cc Orionis, 

 the occurrence of radiating carbon vapour is an indication that these 

 stars, like comets, consist of uncondensed swarms of meteorites. 

 Since the temperature of a condensing swarm of meteorites must be 

 increasing, in accordance with thermodynamical principles, stars like 

 cc Orionis must be placed on the ascendiDg arm of the temperature 

 curve. The photographs show that the spectrum of stars like 

 cc Tauri is almost identical with that of cc Orionis so far as the lines are 

 concerned, and since one of the flutings in the red in the spectrum of 

 a Orionis also appears in cc Tauri, this star must also be regarded as 

 one of increasing temperature, 



The discussion of the Kensington photographs led me to place 

 7 Cygni next to cc Tauri in the series of stars with increasing tem- 

 perature,* and we now know that S Cephei must be classed with this 

 star. I pointed out that the spectrum of 7 Cygni " has much in 

 common with that of cc Tauri, but there is less continuous absorption, 

 and many of the lines of cc Tauri thin out. The next step to cc Cygni 

 is rather a long one, but it seems very probable that if more photo- 

 graphs were available intermediate spectra would be found. It will 

 be seen, however, that in cc Cygni the hydrogen lines are intensified 

 as compared with 7 Cygni, and that all the important lines of cc Cygni 



agree in position with prominent lines in 7 Cygni In passing 



to Rigel the more important lines of cc Cygni are retained, and a few 

 new lines make their appearance." 



My argument was, then, that with an increase of temperature a 

 star like cc Tauri would develop into one like 8 Cephei, which, with 

 further increase, would pass through successive stages represented by 

 cc Cygni, Rigel, and Bellatrix. To justify this it is accordingly 

 necessary to show greater reason for associating S Cephei with 

 cc Tauri than with a star like the sun, which we know on other 

 grounds to be cooling. 



From the great similarity of the line spectra of cc Tauri and the 

 sun, it is clear that the argument will not entirely depend upon the 

 identity of lines in the spectra of a Tauri and 8 Cephei, but upon 

 general and specific differences between 8 Cephei and the sun. 



Evidence from Enhanced Lines. 



In a recent communication to the Royal Society,! I have shown 

 that the investigation of the lines enhanced in the spark spectra of 



* 'Phil. Trans.' (1893), A, vol. 184, p. 708. 

 f ' Boy. Soc. Proc.,' vol. 61, p. 381. 



