1889'.] of the various Species of Heavenly Bodies. 261 



In the first of these, X Geminorum, the companion has probably 

 been added since the primary condensed, for we cannot place the two 

 components on curves which begin at the same point. 



With regard to 7 Leonis, there is a difficulty as to what spectrum 

 should be associated with the greenish-yellow component, so for the 

 present it cannot be stated whether both have condensed from the 

 same nebulosity or not. 



We cannot include Herculis in Class 2, because the difference 

 between the magnitudes of the two components is too great, but we 

 can represent the case by starting the companion curve a little later 

 than the primary curve. We may therefore conclude that we have 

 here to deal with an added companion. 



. X. Conclusion. 



From the foregoing lists and discussions it will be seen that in 

 nearly all cases the components of a binary can be shown with much 

 probability to have had their origin in doable nebulae. There are 

 exceedingly few cases in which it seems at all likely that the com- 

 panion is an addition of a cometary nature, and it is possible that 

 even these few exceptions may be due to errors of observation. 



This, then, strengthens the view that in the case of regular variable 

 stars of Group II we are in presence of the formation of a double star, 

 at an early period in its history when the two swarms are at times, 

 so to speak, in contact. When the variability is not regular we are 

 in presence of the formation of a multiple system. 



I cannot omit to point out how very admirable the colour observa- 

 tions must have been to stand the strain to which the foregoing 

 generalisation has subjected them, and that if equal skill be now 

 applied to observation of the spectra of these bodies, a considerable 

 advance in our knowledge may be looked for. 



In the discussion included in this paper, I have been aided by 

 Messrs. Fowler, Gregory, Baxandall, Porter, and Coppen. Mr. 

 Fowler made the observations of the spectrum of the large air 

 vacuum tube, and of the spectra of manganese nodules and iron 

 spherules. He also classified the binary stars, Mr. Coppen assisting 

 him in preparing the tables. 



Mr. Gregory has been responsible for preparing the various tables 

 in connexion with comets and auroree. 



Messrs. Baxandall and Porter have prepared most of the maps 

 and drawings, for the careful reproduction of which I have to thank 

 Mr. Collings. 



I wish, as before, to tender my thanks to them -for the unflagging 



