18(3 Prof. Gr. Wiedemann on the Laws of the 



equally to both tubes. As, however, in the tube I. the dis- 

 charge begins with greater initial velocity, more electricity 

 might, in consequence of the rarefaction of the air, follow the 

 first discharge than in the tube II. How far these different 

 conditions balance one another cannot be decided without 

 further investigation. If, further, the tension of the electri- 

 city is too feeble to break through the denser gas, all the elec- 

 tricity from the electrodes of the tube I. may eventually flow 

 back to those of the tube II., and eventually to the coil ; and 

 in that case no discharge would take place in the denser gas. 

 If, on the other hand, the gas in the tube II. is so much rare- 

 fied that it is insufficient to convey the electricity, the discharge 

 can only pass through the tube I. with the denser gas. 



If a tube be connected with a coyistant battery, different rela- 

 tions may occur. If the electromotive force at the poles of 

 the battery be too small to communicate to the electrodes the 

 tension necessary for the commencement of a discharge, no 

 current passes. If the electromotive force is sufficient, or 

 more than sufficient, there ensue discharges which may con- 

 tinue for shorter or longer periods, according to the quantity 

 of the electricity accumulated at the poles and to the velocity 

 of the conduction of the electricity from the battery. Their 

 further continuance depends upon the conductivity of the sub- 

 stances composing the battery. If this is small, the tension 

 on the electrodes diminishes so much at the discharge that 

 the battery cannot offer sufficient tension for the continuance 

 of the discharge ; consequently it requires a longer time 

 after each discharge to reproduce the tension at the poles ; 

 there ensue single intermittent discharges. If the conducti- 

 vity of the battery is comparatively great, and if the tension 

 is renewed so quickly that, in the time between the first and 

 the next discharge, the gas rarefied by the discharge at the 

 electrodes has not yet returned to its former condition, the 

 first discharge leads on to the second, and so on in a con- 

 tinuous current, in which the substance of the electrodes 

 eventually participates. But this current differs again from 

 the ordinary galvanic current in that it is essentially dependent 

 upon the impulse which the portion of the gas lying nearest 

 to the electrodes receives from the electricity conveyed to it. 

 The discharge of the latter ensues as when the machine is 

 used as in the experiments mentioned at the beginning of this 

 paper. Thus the " resistance to the discharge," or more pro- 

 perly the quantity of electricity simultaneously discharged, is, 



