SECT. iv. METALS IN THE SOLAR ATMOSPHERE. 153 



to the seven metals whose vapours M. Kirchhoff has 

 shown to exist in the atmosphere of the sun, but he 

 thinks it doubtful whether barium, zinc, or copper are 

 solar metals, for although their brighter lines corre- 

 spond with distinct dark solar lines, their weaker lines 

 do not. Strontium is doubtful also, for one of its 

 strongest bright lines is not coincident with any dark 

 line. Though both iron and nickel are decidedly solar 

 metals, yet as cobalt is doubtful, it cannot be presumed 

 that meteorites are of solar origin. 



The spectrum of luminous magnesium has many 

 green lines perfectly coincident with those in the solar 

 spectrum, so there is no doubt of that metal being a 

 constituent of the sun's atmosphere. But there are 

 magnesium rays as well as some of iron of such high 

 refrangibility that in Mr. Stokes's long spectrum they 

 are situated ten times as far from H as the whole 

 length of the visible spectrum from A to H. These 

 highly refrangible rays only become visible at the 

 exalted temperature of the electric spark, and as they 

 are not found in the solar spectrum, it is inferred that 

 the heat of the sun is inferior to that of the electric 

 spark. 3 Mr. Eoscoe observes that this conclusion would 

 only be legitimate if we knew that these rays of high 

 refrangibility are not absorbed in passing through the 

 atmosphere. 



These are some of the most striking results of the 

 numerous investigations that have been made since 

 M. Kirchhoff published his discoveries, for the subject 

 is anything but exhausted. 



The intensely vivid light of a magnesium flame is 

 rich in violet and extra-violet rays, partly due to the 



3 Dr. W. A. Miller hds shown that these invisible highly refrangible 

 rays exist in the vapours of all metals, and has obtained photographs of 

 their spectra (see Phil. Trans. 1862, p. 876), which correspond to the 

 spectra of fluorescence. 



