Force of the Electric Discharge through Gate*. 439 



then taken out and tested for its magnetism by an ordinary 

 magnetometer with lamp-and-scale arrangement. 



From these experiments it was found that the maximum 

 current in the discharge, as measured by its magnetizing 

 effect on the bundle of wire in the spiral, was not at all pro- 

 portional to the average current measured by the galvano- 

 meter. The experiments showed that while the maximum 

 current increased with the pressure of the gas in the discharges- 

 tube, the average current, and with it the magnetic deflexion, 

 decreased. Different kinds of wire, from soft iron to hard 

 steel, were used for testing the maximum current; but all 

 showed the same result. 



In order to increase considerably the maximum current in 

 the discharge without changing the pressure of the gas, 

 leyden-jars were connected across the electrodes as shown in 



%. 13. 



Fisr. 13. 



do coil 



do coil 



conducting sheet of me tot. I . 



The inside coatings of the jars were connected together and 

 to one terminal of the discharge-tube and the outer coatings 

 to the other terminal. 



The effect of this arrangement was to increase very much 

 the intensity of the discharge and its magnetizing power, but 

 always to decrease the deflexion, whether caused by magnetic 

 force or by convection- currents : this effect was increased by 

 making a spark-gap in the discharge- circuit at a in the 

 diagram (fig. 13). 



On examining this discbarge in a rotating mirror I found 

 that it was, so far as the mirror showed, instantaneous ; and 

 as the duration of the discharge, from these and previous 

 experiments, seemed to be very closely connected with its 



