ox THE DISOHARCtE OF ELECTRICITY TIIROT^OII EX- 

 HAUSTED TUBES WITHOUT ELUCTRODES.* 



Bv J. J. Thomson, F. R. S. 



The following experiments, of which a short account was read before 

 tlie Cambridge Philosophical Society last February, were originally 

 undertaken to investigate the phenomena attending the discharge of 

 electricity through gases when the conditions are sinii)litied by confin- 

 ing the discharge throughout the whole of its course to the gas, instead 

 of, as in ordinary discharge-tubes, making it pass from metiillic or glass 

 electrodes into the gas, and then out again from the gas into the elec- 

 trodes. 



In order to get a closed discharge of this kind we must produce a 

 finite electromotive force round a closed circuit, and since we can not 

 do this by the forces arising from a distribution of electricity at rest, 

 we must make use of the electromotive forces produced by induction. 

 To break down the electric strength of the gas such forces must be very 

 intense while they last, though they need not last for more than a short 

 time. Forces satisfying these conditions occur in the neighborhood of 

 a wire througli which a Leyden Jar is discharged. During the short 

 time during which the oscillations of the Jar are maintained enormous 

 currents ])ass through the wire, and as with a moderate-sized Jar these 

 currents change their direction millions of times in a second, the elec- 

 tromotive force in the neighborhood of the wire is exceedingly large. 

 To make these forces available for producing an electrodeless discharge, 

 all we have to do is to make the wire conn(;cting the coatings of the Jar 

 the jtrimary of an induction-coil of which the discharge-tube itsdf forms 

 the secondary. The arrangements which I have employed for this pur- 

 ])ose are represented in the accomi)anying diagram. 



In {a) A is the inside coating of a L('yd<Mi jar: this is connected to 

 E, one of the poles of a Wimshnrst electrical nmcliine, or an induc- 

 tion-coil, the other pole F of the machine being connected to B, the outer 

 coating of the Jar. A C D is a wii-e connected to the inner coating of 

 the Jar, a few turns C (which we shall call the primary coil) are made in 

 this wire; these turns are scpiare if the discharge-tube is sqnare, ciicu- 

 lar if the discharge-tube is a spherical bull). The wire at I) is attached 

 to an air-break, the other side of which is connected with the outer 



From tho L. E. D., Phil. Mag., October and November, 1891; vol. xxxii, pp. 321- 

 330, and 115-464. 



229 



