1894.] 



on Electric Discharge through Gases. 



243 



with the oxygen, forming water, in which hydrogen is the electro- 

 positive element ; thus, in this case, it is the positively charged 

 hydrogen which is in the state best fitted for pairing. The conse- 

 quence is, the positive charge would be most readily removed from 

 the gas and the negative left — exactly the opposite to that which 

 occurred when the electrodes were bright. This reversal, as I stated 

 before, is verified by experiment. 



I have hitherto only spoken of the phenomena which accompany 

 the passage of electricity from the electrode to the gas, or from the 

 gas to the electrode. 



I shall now pass on to consider the properties of the discharge 

 when it is entirely confined to the gas. 



"We may produce a discharge which, during the whole of its 

 course, shall be confined 

 to the gas in the way 

 represented in the dia- 

 gram. 



The two poles of a 

 Wimshurst machine are 

 connected to the insides 

 of two jars A and B, 

 while the outsides of 

 these jars are connected 

 together by a metal wire 

 wound so as to form a 

 coil. The electricity 

 from the Wimshurst 

 machine charges up the 

 jars, the difference of 

 potential between the 

 poles increases until a 

 spark passes. The pas- 

 sage of the spark puts the 

 insides of the two jars in 

 connection, and the jars are discharged. The discharge of the jar, as 

 was proved from the theory of electro-magnet action by Lord Kelvin 

 more than forty years ago, and shortly afterwards confirmed by the 

 experiments of Feddersen, is an oscillatory one, producing currents 

 surging backwards and forwards through the wires with extraordinary 

 rapidity. The subject of these oscillatory currents is one which in 

 this year is tinged with melancholy. The year had hardly commenced 

 when we lost Hertz, whose splendid work on these electrical oscilla- 

 tions is known to you all. The Managers of this Institution have 

 marked their sense of the importance of this work by devoting a 

 special lecture this session to this work alone, and they have 

 entrusted that lecture to a most distinguished worker in the same 

 field as Hertz. It would therefore be presumptuous on my part to 

 refer in any detail to Hertz's work, but no physicist, and least of all 



