CELESTIAL DISSOCIATION. 199 



deeply imbued with the idea that an element could have 

 one, and only one, spectrum. They held this idea because, 

 while not formally saying so, they believed that the atom 

 which caused the spectrum was a simple undecomposable 

 thing. This idea of one element one spectrum was a 

 serious mistake, for it has been possible to show that an 

 clement may have two or three spectra under different con- 

 ditions, though each spectrum is characteristic for that 

 substance. Thus, Plucker and Hittorf in 1865 announced 

 that " there is a certain number of elementary substances 

 which when differently treated furnish two kinds of spectra 

 of quite a different character, not having any line or band 

 in common.' 7 This was the first blow to the current con- 

 ception though it took many more before the fact won the 

 concensus of scientific belief. The conditions under which 

 a substance must be placed in order that its spectrum may 

 change from one form to another are as follows: 



a. The temperature of a flame. 



b. The temperature of the electric arc. 



c. The vibration due to an electric spark of very high 

 potential. 



These three conditions, which amount to three stages in 

 temperature, give rise, in the case of many elements, to 

 three or more different spectra. 



Let us take the case of iron, according to Lockyer: 



1. The flame spectrum consists of a few lines and flut- 

 ings, only, including several well-marked lines, some of them 

 arranged in triplets. 



2. The arc spectrum consists, according to Rowland, of 

 2,000 lines or more. 



3. The spark spectrum differs from the arc spectrum in 

 the enhancement of some of the short lines and in the 

 reduced relative brightness of others. 



