1854.] OF THE ROYAL INSTITUTION. 361 
very interesting experiments on this subject,* and Mr. Grove not by 
means of M. E. Becquerel’s plan, but by igniting by the voltaic 
battery two platinum wires placed close to each other in an ex- 
hausted receiver, and connecting them with a third voltaic battery, 
had obtained slight deflections of a delicate galvanometer. 
These effects are however far inferior to those shown by flame, 
and appear to depend more upon the state of the terminals than 
upon the state of the intervening gas:—thus, until the terminals 
attain a red heat, no transmission takes place, whatever be the 
degree of attenuation of the gas; while if the terminals have attained 
a red heat, the current is much more easily transmitted by rare than 
by dense gas. Thus alterations in the density of the gas do not 
appear to affect the trasmission, until a certain definite change has 
taken place in the state of the terminals. Reasoning from these 
effects, and bearing in mind the effect of rarefied gas on the dis- 
ruptive discharge and the quasi-radiation of matter in the phenomena 
elicited by Moser and others, Mr. Grove inclined to the opinion 
that the transmission across heated gas differed specifically from that 
across flame, the former being in some respects analogous to the 
disruptive, while the latter resembled the electrolytic discharge. 
Flame moreover has been observed to conduct better in one 
direction than another, and the question next arises will flame produce 
or generate a voltaic current? M. Hankel and M. Buff have 
published papers shewing, by the use of highly sensitive galvanome- 
ters, a current apparently produced by flame, which passes from the 
upper to the lower part of the flame. M. Buff attributes this 
current to thermo-electricity—the flame being a conductor and two 
metals in contact with different parts of it, the thermo-current 
passes from the hotter to the cooler metal, and hence the result. 
Mr. Grove in studying this subject, and without having then read 
the papers of Hankel and Buff, found the results so varying in ordi- 
nary flame that he could come to no satisfactory conclusion; he was 
then led to think, that as in the flame of the blow-pipe, the direction 
or line of combustion is more definite than in ordinary flame, he 
might get more definite results. He experimented with the latter 
flame, and immediately got very distinct evidence of a current not 
due to thermo-electricity, as it could be made to conquer both the 
effect of the thermo-flame- current noticed by Buff, and of any thermo- 
current excited in the junction of the wires exterior to the flame. 
This current which Mr, Grove termed the flame current proper, 
moves from the root towards the point of the blow-pipe flame—the 
best points for placing the collecting spirals or plates of platinum being 
for the one a little above the root or base of the blue cone, and for the 
other, in the full yellow flame a little beyond the apex of the blue cone. 
As the latter metal is much more heated than the former, the 
thermo-flame-current is opposed to, and though it by no means 
destroys, it tends to weaken the effect of the flame current proper ; 
* Annales de Chimie, Nov. 1853. 
