COMPOSITION OF THE SUN RUSSELL. 203 



faint) in the sun, and carbon, oxj^gen, nitrogen, and sulphur have 

 thus been added to the list. 



The second clue to the riddle is found in the fact that the same 

 element may give difl'erent spectra under different conditions. 



The importance of this was first realized by Sir Norman Lockyer 

 more than 50 years ago. When a compound of a metal is volatilized 

 in a Bunsen or hydrogen flame, the spectrum shows lines (usually 

 only a few) characteristic of the metal, and often also bands (re- 

 solved with high dispersion into a multitude of close-packed lines) 

 which vary with the compound present and are evidently due to 

 undecomposed molecules. At the higher temperature of the electric 

 arc, many (though not all) of the bands disappear, in.dicating that 

 the compounds have been dissociated by the high temperature, while 

 the lines — due evidently to free atoms of the metal — become stronger 

 and more numerous. Band spectra, if present, are strongest in the 

 light from the relatively cool outer flame of the arc, for obvious rea- 

 sons. With the condensed spark between metallic poles as source, 

 new lines make their appearance, and others which are faint in the 

 arc are greatly intensified or " enhanced." 



Lockyer suggested, more than 50 years ago, that the elements 

 familiar to the chemist were themselves compounds which, at the 

 very high temperature of the spark, were decomposed, and that the 

 enhanced lines came from the products of this decomposition while 

 the arc lines were from the unaltered element. He supported his 

 contention by the observation that, in white stars like Sirius — which 

 are undoubtedly hotter than the sun — only the enhanced lines of the 

 metals appear, while in the sun both arc and enhanced lines are 

 present. For silicon, he pointed out four groups of lines appearing 

 in hotter and hotter stars as representing successive stages in the 

 process of decomposition of the familiar element. 



Lockyer was in advance of his times; and his bold theory, though 

 generally doubted at first, was essentially sound. We have now 

 every reason to believe that atoms are complex structures, composed 

 of electrons circulating in some fashion about a nucleus. By dis- 

 turbing the atoms with sufficient violence, one after another of these 

 electrons may be removed, a process called ionization. Both neutral 

 and ionized atoms can emit and absorb line spectra, but the two 

 are ciitirely different. Spcctroscopically, an ionized atom behaves 

 like an entirely new element, and each successive ionization again 

 leads to a wholly different spectrum. The arc lines belong to the 

 neutral atoms, ordinary enhanced lines to atoms which have lost one 

 of their electrons, and so on. Lockyer's four sets of silicon lines 

 correspond to atoms possessing all their electrons, or deprived of one, 

 two and three. The detailed theory of the process was first applied 



