140 ANNUAL OF SCIENTIFIC DISCOVEKY. 



but an electric spark gives with the greatest distinctness the spectrum 

 of the metal of which the electrodes consist. A large Ruhmkorff's in- 

 duction-coil yields electric sparks in such rapid succession that the 

 spectrum can be observed as easily as that of the sun. 



A very simple arrangement permits the comparison of the spectra 

 of two sources of light. The rays of one of the sources may pass 

 through the upper half of the vertical slit, while those of another pass 

 through the lower half. When this is the case, one of the two spectra 

 is seen immediately beneath the other, and it is easy to determine 

 whether coincident lines occur in both. 



In this manner the author satisfied himself that all the bright lines 

 peculiar to iron correspond to dark lines in the solar spectrum. In 

 the portion of the spectrum between D and F, about seventy par- 

 ticularly well-marked lines occur, resulting from the iron in the sun's 

 atmosphere. 



Iron is remarkable on account of the great number of distinct lines 

 which it produces in the solar spectrum. Magnesium is interesting be- 

 cause it produces the group of Fraunhofer's lines lying in the green 

 denoted by Fraunhofer by &, and consisting of three very strong lines. 

 Very distinct dark lines in the solar spectrum correspond to the bright 

 lines produced by chromium and nickel, and we may, therefore, regard 

 the presence of these substances in the sun's atmosphere as proved. 

 Many other metals appear, however, to be wanting in the sun's at- 

 mosphere. Silver, copper, zinc, lead, aluminum, cobalt, and antimony 

 have extremely brilliant lines in their spectra ; but no distinct dark 

 lines in the solar spectrum correspond to these. 



Many metallic compounds do not give in a gas-flame the spectrum 

 of their metal, because they are not sufficiently A'olatile. In these cases 

 the spectrum may be made to appear by means of the electric spark. It 

 is true that* in this case the spectrum of the metal of which the elec- 

 trodes consist and that of the air in which the spark passes is also seen. 

 To avoid the difficulty arising from the very great number of bright 

 lines of which the spectrum of every electric spark consists, it is neces- 

 sary to have recourse to a particular arrangement. The electric spark 

 is allowed to pass at the same time between two similar pairs of elec- 

 trodes, the light of one spark being allowed to pass through the upper, 

 that of the other through the lower half of the slit, so that one spec- 

 trum is seen above the other. When the two pairs of electrodes are 

 clean, the two spectra are perfectly similar ; when, however, a metal- 

 lic compound is placed upon one pair, the corresponding spectrum 

 immediately shows the lines belonging to the metal introduced. The 

 author has satisfied himself that in this manner even the metals of the 

 rare earths, yttrium, erbium, terbium, etc., may be recognized most 

 quickly and certainly. It is, therefore, to be expected that, by the 

 help of Ruhmkorff's apparatus, the spectral method of analysis may be 

 extended to the detection of all metals. The researches which the 

 author has undertaken, in connection with Bunsen, will, it is hoped, 

 determine this point. Journ. far Prakt. Chimie. 



