M. A. J. Angstrom on the Fraunhofer -lines 



found also thirteen manganese-lines closely approaching those of 

 iron. The identity of most of these lines with corresponding 

 lines of the solar spectrum is certain. 



As regards strontium and barium, the former has between H 

 and G two strong lines, which seem to correspond to lines in the 

 sun's spectrum, and the same seems to be the case with the line 

 F. At the same time we must add that the strong strontium- 

 line that lies about midway between G and F, has no strong 

 corresponding line in the solar spectrum, which circumstance 

 makes the presence of strontium in the sun somewhat uncertain. 



The 6anwm-spectrum exhibits four lines between F and D, 

 which seem to have their counterparts in the solar spectrum; 

 but this also requires further confirmation. 



Our examination of the red end of the spectrum, from D 

 onwards, is less complete, and I accordingly forbear to speak of 

 the lines that occur in that part of the spectrum. Moreover, the 

 dispersion being here considerably less, the difficulty of deter- 

 mining whether two lines in the solar and metallic spectra 

 respectively exactly coincide or not is proportionally increased. 



When we sum up what has been adduced with respect to the 

 Fraunhofer-lines, and assume that the line C belongs to hydro- 

 gen, we obtain the following general view of the lines in the 

 spectrum : — 



The two lines H belong to calcium. 



The line G belongs to iron. 

 „ „ F „ „ strontium and iron (uncertain ). 

 „ „ b „ „ magnesium and iron. 

 „ „ D „ „ sodium. 

 „ „ C „ „ hydrogen. 

 „ „ B „ „ potassium. 



As an example of our method of figuring the solar spectrum, 

 as space does not permit us to insert the drawings in their 

 integrity, we give in the accompanying Plate I. fig. 1, the lines 

 which lie between the two H's, as well as those which are found 

 in the neighbourhood of G, F, b and E. 



§ 4. That the theory of the reciprocity of radiation and absorp- 

 tion, to which we have more than once referred in the foregoing 

 pages, is not a mere conjecture, but a logical consequence of the 

 theory of thermometric heat published by me in 1848*, will, I 

 trust, be placed beyond doubt by the following short view of the 

 theory of thermometric heat, as expounded in the paper referred to. 



The possibility of a mathematical treatment of thermometric 

 heat always presupposes a greater or less number of hypotheses 



* Forsok till en mathematisk theori for thermometriska varmet 

 (Attempt at a mathematical theory of thermometric heat). Upsala, 1848. 



