[ 939 ] 
XXXVII. On Spectra of Electric Light, as modified hy the Nature of the Electrodes and 
the Media of Discharge. By the Rev. T. E. Eobinson, D.D., F.R.S., &c. 
Eeceived June 19*, — Bead June 19, 1862. 
The important discovery of Wheatstone, that the spectra of electric sparks contain 
brilhant lines whose character depends on the nature of the electrodes, after being almost 
neglected for several years, has lately become an object of great interest ; and much 
additional light has been thrown on it by several physicists, of whom Masson, Angsteom, 
o 
PLiicKEE, and Kiechhoef are the most conspicuous. Angsteom has announced as a 
general law that the lines in question are produced by the electric current igniting the 
medium in which the discharge takes place, or molecules of the electrodes which are 
tom off by its passage, that each of these actions produces its own spectrum, and that 
those spectra are simply superposed without any modification. The gases with which 
he worked were at the ordinary pressure. Pluckee, on the other hand, used the well- 
known Geissler tubes, which contain minute proportions of highly rarefied gas or vapour. 
He attaches very little importance to the lines due to the electrodes {metallic lines), which 
he thinks are confined to those portions of the spark near the electrodes ; and he main- 
tains that in the centre of an exhausted tube of some length only gaseous lines are seen. 
He was embarrassed in several instances by the decomposition or absorption of the 
gaseous media ; and there must always be some doubt as to the precise nature of these 
media, as the tubes are hermetically sealed. On the other hand, Kiechhoef seems 
to attach most importance to the metallic lines, whose infiuence he has exhibited to a 
wonderful extent by a spectrum- apparatus probably unrivalled. All hold the doctrine 
of an essential connexion between the character of tlie spectral lines and the chemical 
nature of the substances which are present in the track of the discharge ; and the last, 
in conjunction with Bunsen, has based on this principle the new system of spectral 
analysis, which is rapidly becoming popular, and has applied it to explain the dark 
lines of the solar spectrum in a way which, if not absolutely certain, is singularly elegant. 
Yet it is impossible to overlook the fact that, in all this, much is assumed, not proved. 
Has it been established that these lines depend so absolutely on chemical character that 
none of them can be common to two or more different bodies \ Has it been ascertained 
that, while the chemical nature of the bodies present remains unchanged, the lines never 
vary if the circumstances of mass, density, &c. are changed And what evidence haA-e 
we that spectra are superposed, so that we observe the full sum of the spectra which the 
* The continuation of the paper, from p. 974 to the end, was not received complete till September 1 ; but 
the conclusions therein contained are embodied in the abstract presented to the Society on June 19, 1862, 
MDCCCLXII. 6 N 
