250 Prof. Rijke on some Properties of the Induced Current. 



diameter. When the flame was produced by the combustion of 

 a gas, a Bunsen's lamp was used. In these experiments the 

 lower orifices of the lamp were closed, so that there was not a 

 mixture of air and gas. I have always used one of the large- 

 sized RuhmkorfFs coils. Two Bunsen's elements were ordi- 

 narily used, each consisting of plates which exposed 352 square 

 centimetres of surface. 



The following are the first results which I obtained : — 



Medium in which the 



inducing circuit was 



broken. 



Greatest 

 striking distance. 



Ratio between the 

 striking distances 

 obtained when the 

 circuit is broken in 

 the given medium 

 and in air. 



Atmospheric air ... 

 Hydrogen flame ... 



millims. 



{t\} 



M9-3-1 

 119-6 J 



J"56-6\ 

 \58-6J 



f 57-21 

 159-0/ 



Mean 58-1 



1 



2-395 

 7110 

 7173 



Thus the action of an alcohol flame was far inferior to those 

 which hydrogen and coal-gas flames exert. The last two appear 

 to produce almost the same effect. 



3. I am stating nothing new to most persons in mentioning 

 that the shape of the electrodes between which the discharge of 

 a conductor of a Leyden jar, or of an induced current, is made 

 to pass, exercises a considerable influence on the striking distance ; 

 and every one knows, further, that this distance increases consi- 

 derably when a tapering conductor is taken for the positive elec- 

 trode, and for the negative electrode a plate arranged perpendi- 

 cularly to the first electrode. It remained to inquire if induc- 

 tion sparks, which, by being made to strike between new elec- 

 trodes, were able to cross greater distances, could be further 

 elongated when the rupture of the inducing circuit, instead of 

 taking place in air, took place in a flame. The following results 

 prove that this question ought to be answered affirmatively, 

 excepting merely that the increase is not so considerable : — 



