296 Scientific Proceedings, Royal Dublin Society. 
quartz prism photographs obtained in a similar manner; and it 
will be noticed that the difference between them is not very 
remarkable, particularly in the violet and ultra-violet. 
That these photographs contain a greater number of lines than 
those of Eder and Valenta may be accounted for—first, by the 
difference in the flame employed and the part of it examined ; 
secondly, by the composition of the gas not being exactly the 
same. It has been observed by Lecocq de Boisbaudran (and this 
has been mentioned in my original paper) that even a Bunsen 
burner may offer, under slightly different conditions, spectra which 
materially differ. With compressed oxygen used in a blow-pipe 
with a large flame such differences may be greatly increased, and 
this is believed to have been the case in consideration of the fact 
that a whole group of lines, not recorded by Eder and Valenta, 
are in close agreement with a group peculiar to the spectrum, which 
I have since recognised as belonging to cyanogen. 
Eder and Valenta examined the inner cone of a Bunsen burner, 
fed with oxygen at 3 to 2 of an atmosphere pressure, with an 
instrument of small dispersion, and also with a grating of 15 feet 
radius. I studied the upper half of a gas blow-pipe flame sup- 
plied with oxygen under a pressure of as much as 22 atmospheres, 
using merely a quartz prism, and by such means only could the 
purpose of the investigation have been satisfactorily fulfilled. 
There is, however, one further remark to be made, namely, not 
only was the method of burning the gas different to that employed 
by Eder and Valenta, but the composition of the gas itself may 
have been very different. 
If, for instance, the gas contained cyanogen or cyanides it would 
certainly yield the cyanogen lines or bands upon combustion, for 
very minute traces suffice to produce some part of that spectrum. 
I entertained no doubt whatever that the cyanogen bands entered 
into the spectrum of the oxy-coal-gas flame examined by me. 
Moreover I have proved the existence of cyanogen compounds 
in the gas in quite sufficient quantity to account for it in the 
flame. 
The supply pipe of an ordinary Bunsen burner was made to 
deliver gas at the rate of five cubic feet per hour, and for a period 
of one hour, into a series of bulbs containing a solution of ferrous 
sulphate, made strongly alkaline with caustic potash solution. 
