﻿Flames of Gases containing Carbon, 209 



Spectrum of Coal-gas. 



When Swan published* his investigations on the prismatic 

 spectra of the flames of compounds of carbon and hydrogen, 

 which are remarkable for their completeness in the then (1855) 

 state of our knowledge in spectrum-analysis, the only known 

 observations were those of Fraunhofer, Brewster, and Draper 

 on the spectrum of the cone of the blowpipe-flame, and those 

 of Fraunhofer, Herschel, and others on the spectrum of a wax- 

 or oil-flame ; moreover the means of making this kind of ob- 

 servation had not at that time attained its present high degree 

 of perfection. Under these circumstances, the service rendered 

 by Swan must not be underestimated — in examining the spec- 

 tra of the flames, not only of carbon combined with hydro- 

 gen, but of the various compounds of carbon, hydrogen, and 

 oxygen, in doing this with an accuracy not surpassed even in 

 recent times, and in arriving at conclusions which are still con- 

 sidered true in their fundamental principles, viz. (1) that the 

 position of the bright lines in the spectra of the different com- 

 pounds of carbon and hydrogen is independent of the relative 

 proportion of those two substances contained in them and is in 

 all cases the same, and (2) that compounds which contain 

 oxygen as well as carbon and hydrogen give spectra identical 

 with those of compounds of the latter two substances only. 



Since the time to which I refer, this subject has been again 

 investigated ; and particularly Plucker and Hittorf f, as well as 

 H. C. DibbitsJ, have annexed to their publications very excel- 

 lent drawings; those indeed of the last-named investigator 

 leave nothing to be desired in their repetition. But not one of 

 these authors has mentioned, in treating of the spectrum of 

 coal-gas or olefiant gas obtained by combustion with oxygen, a 

 group of five red lines, which I, in my oft-repeated experi- 

 ments, have always been able to observe with uniform distinct- 

 ness, and which decidedly belong to the spectrum of coal-gas. 



For the production of this spectrum, 1 made use of aDaniell's 

 cock of the ordinary construction, with an escape-aperture, nar- 

 row in bore compared with that of the tubes by which the coal- 

 gas and oxygen are brought into communication, regulating the 

 quantity of the gases to be combined in such a manner that I ob- 

 tained a small, nearly globular flame, only narrowing into a very 

 short point at the top, which was of a faint bluish-white colour 

 and an intense brilliancy. A flame thus produced, and brought 



* Transactions of the Royal Society of Edinburgh, vol. xxi. part 3, 

 p. 411. 



f Phil. Trans, vol. civ. part. 1. p. 15. 



% De Spectraal- Analyse, Akademische Proefschrift : Rotterdam, E. H. 

 Tassemeyer, 1863. 



Phil. Mag. S. 4. Vol. 37. No. 248. Mar. 1869. P 



