Electric Discharge in rarefied Hydrogen and Oxygen. 295 



water-vapour (=7?o— pi), was always large enough to measure 

 with sufficient accuracy and was often as much as 10 per 

 cent of p, and sometimes much more. But it is safe to 

 assume that no errors are involved in the results by taking 

 Ap so large, provided that the current keeps steady or 

 changes but little, for then the distribution of force through- 

 out the discharge cannot be greatly affected. And the 

 consistency of the results given below is sufficient in itself 

 to justify this conclusion. 



The difference of potential, X, between the electrodes 

 during the discharge was found thus. Let C denote the 

 current, Z the voltage of the battery, and R the whole ohmic 

 resistance of the circuit exclusive of the gas, then 



X = Z - OR, 



and by taking the mean value of C we °et the mean value 

 ot X. 



This method of determining X was found to be far more 

 accurate and satisfactory than the use of an electrostatic 

 voltmeter, because both C and X were thus instantly deter- 

 mined bv a single observation, the ammmeter being dead- 

 beat. 



The quantity AQ of electricity which passed through 

 the gas during the discharge was obtained by multiplying 

 the mean value of the current by the time during which the 

 current lasted. 



The duration of the discharge varied greatly. It was 

 determined by the rate at which the current changed, 

 for the current was not allowed a variation of more than 

 5 per cent, on each side of its mean. It ranged from a few 

 seconds for the lowest pressures to 100 for the highest : the 

 lower the pressure, the more rapidly the current changes. 



Results of Experiments. 



Each of the following tables (p. 297) contains the results of 

 a series of experiments made with practically the same mean 

 pressure ; for the mean pressure lay in almost every case 

 within 3 or 4 per cent, of the pressure given at the head of 

 the table. 



To recapitulate : — 



p= half the sum of the pressures, measured in mms. 

 of mercury, before and after the passage of the 

 electricity, the water-vapour being completely 

 absorbed, = %(p*+pi). 

 Ap = the fall of pressure in mms. caused by the passage 

 of the electricity =p Q —p 1 . 



