﻿228 Mr. J. A. Brown on the Potential required 



in value for the zinc, nearly six times larger than the increase 

 of the other two. 



The holes in the plates were quite large enough to ensure 

 an immediate adjustment of pressure in all these gaps, which 

 were joined in parallel. The fact that the gauge showed an 

 increase in both cases, shows that we are not dealing with an 

 absorption of gas in the one case and an emission in the other, 

 as one might a priori be led to believe. That this cannot be 

 an effect in the gas is thus definitely shown from the fact that 

 the two metals had an opposite effect on the sparking potential, 

 and that pumping out and admitting fresh gas (not shown in 

 Table) did not influence the process of return to its original 

 value. Many sets of this kind were taken, with several 

 different spark-gaps, the last ones being always about one 

 centimetre in length, and all with consistent results ; but it 

 should be noted that at higher pressures, above the critical 

 pressure, this effect was much less noticeable. Two plates of 

 absolutely pure silver were obtained from Messrs. Johnson, 

 Matthey & Co., of London, and were found to behave exactly 

 as the zinc, i. e. running the current raised the sparking 

 potential. 



It was not absolutely certain, however, that the apparatus 

 was entirely air-tight, and it was feared that some oxygen or 

 other gas might be having a temporary chemical effect on the 

 electrodes. Great care was accordingly taken to obviate any 

 possibility of leak, and the apparatus was left standing with 

 only hydrogen in contact with the plates for about a month. 

 On resuming work again, it was found that this effect had 

 entirely disappeared ; nor after two months' constant use has 

 it made its appearance again. It should be said that absolutely 

 no air has been allowed to come in contact with the plates 

 since that time. Hydrogen has been the only gas having 

 access to them. It seems probable that occluded gas has 

 something to do with this effect, which is undoubtedly on the 

 electrodes and not in the gas of the sparking- chamber. 



Whatever its cause or seat, it has certainly disappeared 

 with the assurance of the presence of no gas but hydrogen. 

 Since then, the only difference between the results obtained 

 from the aluminium and the zinc electrodes has been the slight 

 and consistent one already mentioned, due probably to a small 

 difference in the distance apart of the electrodes. 



C. Effect of Reversing the Current. 



Another proof that the effect noted above was the result of 

 something closely connected with the plates, and not a phe- 



