o 



4 M. A. J. Angstrom on the ¥raunhofer -lines 



or the gaseous envelope that immediately surrounds that body, 

 even though some of these lines should belong to the atmosphere 

 of the earth. As, moreover, the gases which immediately enve- 

 lope the photosphere must be at a very high temperature, we 

 are fully justified in applying the principle, that these gases ab- 

 sorb just the same kinds of light which they emit in a state of glow- 

 ing heat. Accordingly by seeking out the bright lines in the 

 electrical spectra of different metals which have corresponding 

 dark lines in the solar spectrum, we can with considerable pro- 

 bability determine what metals in gaseous form enter into the 

 composition of the solar envelope. I say only " with consider- 

 able probability ;" for, from the circumstance of two lines coin- 

 ciding in both the spectra of the sun and of a given metal, it 

 by no means follows as a necessary consequence that this sub- 

 stance is to be found in the sun, because, on account of the 

 enormous number of dark lines in the sun's spectrum, such 

 coincidence may be accidental; nevertheless the probability of 

 such an assumption increases in proportion to the number of 

 such coincident lines and their phenomenal peculiarities. 



§ 3. Fraunhofer had already observed that the two soda-lines 

 coincide with the double line D of the solar spectrum; and 

 Brewster has made the interesting observation that the combus- 

 tion of potassium gives lines which correspond with those of the 

 extreme red end of the solar spectrum. 



Kirchhoff, to whom belongs the honour of having first by direct 

 experiments with sodium and lithium proved the truth of the 

 law of correspondence between absorption and radiation, which 

 1 had laid down in the memoir so often referred to, has, in a 

 memoir inserted in the Philosophical Magazine, and in the Bi- 

 bliotheque de Geneve, stated that he had found no less than sixty 

 iron-lines having their corresponding lines in the solar spectrum 

 between the lines F and D of the latter; moreover, that mag- 

 nesium displays lines answering to the group b, and that nickel 

 and chrome have corresponding lines in the sun's spectrum. 

 These statements we have verified and can confirm. 



The number of iron-lines between F and D whose identity 

 with the Fraunhofer-lines we can venture to guarantee, is not 

 indeed so great as that affirmed by Kirchhoff, but it is quite suf- 

 ficient to establish beyond a doubt the presence of iron in the 

 solar atmosphere. The iron-lines are the most characteristic 

 in the whole solar spectrum ; and if a sufficient magnifying 

 power be employed, or the light be refracted through two 

 prisms, these lines, or at any rate the stronger ones among 

 them, appear perfectly black. 



The same remark, which I have already made with respect to 

 the lines of the electrical spectrum, is in general true of the 



