1893.] 



The Electrolysis of Steam. 



107 



insulate a small positive charge perfectly, while it instantly loses a 

 negative one ; (2) that an uncharged electrode immersed in this gas 

 acquires a positive charge. 



If the spark length is increased until the discharge passes as a 

 spark, then the electrode leaked slowly, whether charged positively or 

 negatively ; the leak in this case is, however, .very small compared to 

 that which exists when the discharge passes as an arc. 



Hydrogen. 



When similar experiments are tried in hydrogen, the results are 

 quite different. 



When the arc discharge passes through hydrogen the electrode E 

 always leaks when it is charged positively ; it does not merely lose its 

 positive charge, but acquires a negative one, its potential falling to 

 U, where U is a quantity that depends on the size of the arc, and on 

 its proximity to the electrode ; in my experiments 5 or 6 volts was a 

 common value for U. If the electrode E is initially uncharged, it 

 acquires a negative charge, the potential falling to U, while if it is 

 initially charged. negatively it leaks if the negative potential is greater 

 than U, until the negative potential falls to U, when no further leak 

 occurs ; if the negative potential is less than U, the negative charge on 

 the electrode increases until the potential becomes equal to U, when 

 it remains steady. 



It is much more difficult to get a good arc in hydrogen than in 

 oxygen, and, as it is essential to the success of the preceding experi- 

 ments that the discharge should pass as a well-developed arc, the 

 experiments with hydrogen are a little more troublesome than those 

 with oxygen. 



These experiments show that the oxygen in or near the arc dis- 

 charges a negatively electrified body, but not a positively electrified 

 one, while the hydrogen in or near the arc discharges a positively 

 electrified body, but not a negatively electrified one. And also that 

 an unchanged electrode becomes positively electrified in the oxygen, 

 negatively electrified in the hydrogen. 



I next endeavoured to see if this charging up of the electrodes is 

 due to an electrification developed by the contact of the gas in the 

 arc with the electrode, or whether this gas behaved as if it possessed 

 an independent charge of electricity. 



If the electrification is due to the contact of the gas with the elec- 

 trode, then it ought to disappear when the electrode is covered with a 

 layer of a non-conductor ; if, however, the gas in the arc behaves as 

 if it were charged, then, even though the electrode is covered with a 

 non-conductor, the electrostatic induction due to the charge on the 

 gas ought to produce a deflection of the electrometer in the same 

 direction as if the electrode were uncovered. 



