﻿Flames of Gases containing Carbon. 215 



Having already described this in my previous papers on this 

 subject, I will only add here those remarks which are suggested 

 by a comparison with other spectra of gases containing carbon ; 

 they may be summed up in the following points : — 



1. The lines of a flame of carbonic oxide (Bessemer-flame) 

 appear on a continuous spectrum, and contain several groups of 

 bright lines and some dark absorption-streaks, which are irre- 

 gularly distributed from the red up to the violet end. 



2. The groups of lines coincide neither with those of coal-gas 

 and olefiant gas, nor with those of cyanogen. 



3. The strongest lines are situated, as in the spectrum of 

 coal-gas, in the green and blue-violet part. 



4. The increase in the intensity of light of the individual 

 lines of the groups, when such an increase is perceptible at all, 

 takes place always in the same direction : but this is the oppo- 

 site direction to that observed in all other spectra of gases con- 

 taining carbon; for the most refrangible line of each group is 

 the brightest, and those succeeding it invariably become fainter. 



5. The spectrum shown by a Geissler's tube filled with car- 

 bonic oxide is not the same as that of the flame of carbonic 

 oxide, since both the position and the distribution of the bands 

 and lines are different. 



The spectrum, then, of a flame of carbonic oxide is such that it 

 must unquestionably be considered as one peculiar to itself — the 

 spectrum of ignited carbonic oxide. 



Spectra of the flames of compounds of carbon and hydrogen, 

 of cyanogen and of carbonic oxide, never show the lines of hy- 

 drogen relatively to those of nitrogen and oxygen ; but from 

 this we can only infer either that the lines of these last-named 

 gases do not appear on any ground of such a kind (as is the case 

 in the spectrum of chloride of potassium, in which also the po- 

 tassium lines alone are visible), or that the molecules of the 

 three first-named kinds of gas as such form the luminous matter 

 when in the state of most intense ignition. Now, as the spectra 

 are of different characters, they cannot be referred to a common 

 origin, namely, to the carbon alone ; and from this we are enti- 

 tled to argue that every gas has its own peculiar spectrum, in 

 so far as it possesses a difference of quality in regard to its con- 

 stitution. 



Lastly, as to the comparison of the spectra of the flames of 

 gases with the spectra given by them when in the condition of 

 greatest rarefaction and made luminous by the induced electric 

 current, I believe I ought not to conceal my opinion that it is 

 really inadmissible ; if the electric current is able to decom- 

 pose so many bodies in their natural condition of density, how 

 much more must it be able to do so when that density is so 



