154 Profs. Liveing and Dewar. Spectra of the [Feb. 5,. 



instead of air, they state that an nltra- violet group of three fluted 

 bands appeared. They notice also certain red bands with shading in 

 the reverse direction, which are better seen when the flame is fed with 

 air than with oxygen. Other observers give similar accounts, noticing 

 the brilliance of the two series of bands in the blue and violet above 

 mentioned, and that they are seen equally in the electric discharge 

 through cyanogen. All do not notice the ultra-violet group, but this 

 is no doubt owing to the absorption of these rays by the prisms 

 employed, for photographs which we have taken of the spectrum of 

 the cyanogen flame in air show this group strongly developed. 



These three bright groups of bands of fine lines are as characteristic 

 of the flame of cyanogen and of the spark in that gas, as the four 

 hydrocarbon bands are of the flame of hydrocarbons. 



The flame of carbonic oxide shows nothing but a continuous spec- 

 trum. The series of bands seen in the spark in an oxide of carbon 

 are well known, and there is not, so far as we know, any dispute about 

 them. The brightest of them appears to have been seen by some 

 observers in the flame produced by the combustion of other compounds 

 of carbon. 



The contention of Angstrom and Thalen is that the channelled 

 spectra of the hydrocarbon and cyanogen flames are the spectra of 

 acetylene and cyanogen, and not of carbon itself, and that in the 

 flame of burning cyanogen we sometimes see the spectrum of the 

 hydrocarbon superposed on that of the cyanogen, the latter being tire- 

 brighter ; and that in vacuum tubes containing hydrocarbons the 

 cyanogen spectrum observed is due to traces of nitrogen. 



No chemist who remembers the extreme sensibility of the spectro- 

 scopic test, and the difficulty, reaching almost to impossibility, of 

 removing from apparatus and material the last traces of air and 

 moisture, will feel any surprise at the presence of small quantities of 

 either hydrogen or nitrogen in any of the gases experimented on. 



Mr. Lockyer (" Proc. Hoy. Soc," vol. xxvii, p. 308) has recently 

 obtained a photograph of the arc in chlorine, which shows the series 

 of fluted bands in the ultra-violet, on the strength of which he throws 



o 



over the conclusion of Angstrom and Thalen, and draws inferences as 

 to the existence of carbon vapour above the chromosphere in the 

 coronal atmosphere of the sun, which, if true, would be contrary to 

 all we know of the properties of carbon. We cannot help thinking 

 that these bands were due to the presence of a small quantity of 

 nitrogen. 



Our faith in the conclusions of Angstrom and Thalen on this 

 subject has been much strengthened by our own observations, which 

 we now proceed to describe. 



The following experiments were made with a De Meritens dynamo- 

 electric machine, arranged for high tension, giving an alternating 



