104 ROYAL SOCIETY OF CANADA 



gated with air), was somewhat higher than the number obtained by Carr 

 under similar conditions. This discrepancy, however, was considered to 

 be due to differences in the air in the two sets of measurements. In 

 Carr's measurements care was taken to have the air both clean and dry 

 but in the present experiments no particular steps were taken to clean 

 and dry the air in which the sparks took place. After taking the meas- 

 urements in air the apparatus was thoroughly washed out with pure and 

 dry hydrogen and a second: set of spark potentials taken for different 

 pressures in this gas. The .results are given in the Table in Fig. IV and 

 a curve illustrating them is shewn in the same figure. This curve as 

 well as the results which 'it represents, is practically identical with that 

 deduced by Paschen's Law from Carr's numbers, and in view of this 

 agreement it was concluded that all Lhe conditions of the experiment 

 were suitable for making a test of the modification in the method of 

 procedure referred to above. 



The manner in which the modification was applied is illustrated 

 by diagram II in Fig. II. 



^ 



As the diagram shews, one terminal of the battery of storage cells 

 was joined to earth and the other to the junction of a divided circuit. 

 One of the branches of the divided circuit led to the terminal of an 

 electrostatic voltmeter which accordingly gave the voltage of the cells. 

 The other branch of the circuit included a high reading Weston voltmeter 

 possessing a resistance of 65,000 ohms, a Weston milliammeter, and the 

 spark gap. 



In making the measurements the voltage from the battery was first 

 applied to the circuit and then the pressure in the discharge chamber 

 was reduced until discharge occurred. When this happened the passage 

 of the current was indicated by both the Weston milliammeter and the 

 Weston voltmeter and the reading on the electrostatic voltmeter gave the 

 potentials at the terminals A and B of the discharge chamber circuit. 



Care was always taken by altering the number of cells in the 

 battery, as required, to see that the electrostatic voltmeter readings 

 remained constant over the whole range of pressures. 



