Theory of Dissociation. 29 



" Consequently there is no iron in the sun, but only its 

 elements "*. 



Liveing and Dewarf have already controverted this rea- 

 soning, by showing that certain spectral lines of a substance, 

 e.g. X 5210 of magnesium and various calcium-lines, become 

 visible only when certain foreign substances are present — in 

 the case in question, hydrogen on the one hand, and iron on 

 the other, — that consequently the non-appearance of certain 

 iron-lines in the spectra of the spots or protuberances ought 

 not to be attributed to a dissociation, but to the absence of 

 foreign substances which produce the strong coming-out of 

 those lines. 



Lockyer, however, now relies upon another fact, which is 

 not accounted for by the experiments of Liveing and Dewar, 

 and which at any rate seems to give a firmer support to his 

 dissociation theory than those above mentioned. He says: — 

 " The last series of observations refer to the degree of motion 

 of the vapours in the sun-spots, which, as is known, is 

 indicated by changes in the refrangibility of the lines. If all 

 the lines of iron in a spot were produced by the vapour of 

 iron moving with a velocity of 40 kilometres in a second, this 

 velocity would be indicated by an alteration of the refrangi- 

 bility of all the lines. But we find that this is not the case. We 

 not merely ascertain different motions indicated by different 

 lines, but have observed in the degree of motion the same 

 changes as in the breadth of the lines. This fact can be easily 

 explained if we assume dissociation; and / know no simpler 

 way of explaining it " J. 



Lockyer mentions, as an instance, that in the spots on the 

 24th December 1880, and the 1st and 6th January 1881, a 

 certain number of iron-lines appeared curved, while others 

 remained straight. 



Now I believe I can explain these facts from numerous 

 observations in absorption-spectrum analysis, without needing 

 to have recourse to the hypothesis of dissociation. 



It is known that the position of the absorption-bands of a 

 substance essentially depends on the dispersion of the medium 

 in which it is dissolved or incorporated. It is often noticed 

 that in strongly dispersing media the absorption-bands of a 

 substance are displaced towards the red§. Now the remark- 



* Com})tes Rendus, xcii. p. 366. 



t Proc. Roy. Soc. xxx. p. 93 ; Wiedemann's Beiblatter, iv. p. 366. 



X I have here followed the rendering of Lockyer's memoir given in the 

 Naturforscher for June 4 ; 1881, in order to exclude any appearance of 

 individual colouring of the translation. 



§ Kundt, Pogg. Ann. Jubelband, p. 620. 



