Ho ROYAL SOCIETY OF CANADA 



potential drop in the balance of the circuit which included the milliam- 

 meter, the connecting wires, and the spark gap. As the resistance of the 

 connecting wires and of the milli ammeter was small the drop in 

 potential corresponding to these would be negligible, and consequently, 

 the differences in the potential readings mentioned would represent the 

 potential required to maintain the current selected in the spark gap. 



A few results bearing on this point have been calculated from the 

 measurements of this investigation and are collected in Table IV. 



The first of these corresponds to a pressure of 9 mm. in the discharge 

 chamber and two others to pressures of 1.5 and 2 mms. respectively. 



As the critical pressure in hydrogen with a spark gap of 3.3 mm. 

 was between 3 and 4 mm. of mercury, the first of the pressures selected 

 was therefore in the region below and the other two in the region above 

 the critical pressure. 



Column I of Table IV contains the pressures. Column II the 

 corresponding sparking potentials, column III current readings deduced 

 from the Weston voltmeter readings, and Column IV the observed 

 potential differences which were found to maintain the currents between 

 the electrodes in the spark gap recorded in the table. 



From the numbers given it will be seen that in the region above 

 the critical pressure a slight fall occurred in the potential between the 

 electrodes when the current increased. In the region of pressures below 

 the critical one, however, the opposite occurred, and the larger current 

 intensities required the higher potential differences to maintain them. 



These results are in keeping with those of Mr. Brown, who found 

 that above the critical pressure the potential required to maintain a 

 current dropped below the sparking potential and tended towards a 

 limiting Value which appeared to be independent of the current, W'hile 

 below the critical pressure the maintaining potential rose more or less 

 rapidly above the sparking potential as the current increased in intensity. 



In conclusion it is my pleasure to thank Professor J. C. McLennan 

 for his kindly interest and valuable advice throughout the investigation. 



