A. E. Kennelly — Alternating Currents. 



455 



disk made six little separate voltameters each with its own pair 

 of insulated leads. By uniting more or less of these leads out- 

 side the dish, the voltameter might be made to comprise any 

 or all of the couples of electrodes, thus enabling a considerable 

 range of electrode area to be at command as shown by the pre- 

 ceding table. 



Over this sextuple voltameter there rested a glass bell Y cms in 

 diameter, terminating after 10 cms of elevation in a vertical glass 

 tube of uniform bore 37*5 cms long and about 0*6 cms internal 

 diameter, closed at the top by a rubber tube and clamp. This 

 tube was graduated for volume by comparison with a burette 

 and found to have a capacity per linear centimeter from several 

 measurements of 0*2684 ± 0*002 c. c. In using the instrument, 



the tube and bell were filled 

 with solution and gas electro- 

 lytically generated until the 

 latter stood at a certain, mark 

 near the top. All being 

 ready, the measured current 

 was then passed through the 

 apparatus for a noted inter- 

 val, at the end of which the 

 length of tube occupied by 

 gas was measured downward 

 from the fiducial mark. 



In the condenser method 

 the connections were as 

 shown in the diagram : where 

 A is the adjustable con- 

 denser, V the voltameter 

 with commutators c c for 

 connecting the separate 

 couples of plates in multiple, JB a Cardew voltmeter, and 1 1 

 the leads to the dynamo. 



Under these conditions we know that the absolute quantity 

 of current passing through the voltameter assuming that the 

 periodic variation of electromotive force follows the sine law : 



Q = n nekt XlCr 1 * (1) 



where Q is the absolute quantity of electricity that traverses 

 the voltameter in time tj e is the mean alternating potential 

 difference as indicated by the Cardew voltmeter B ; k is the 

 capacity of the condenser A in microfarads ; and n is the num- 

 ber of alternations per second. This formula is thus indepen- 

 dent of the resistance of the voltameter, which is quite negli- 

 gible under these conditions. If Y be the volume of gas at 

 zero Centigrade theoretically decomposed by a continuous cur- 



