[fierheller] to determine MINIMUM SPARK POTENTIALS 



107 



Fi'oni the form of the current curves for the different voltages it is 

 easy to deduce the pressures corresponding to zero currents. These zero- 

 current pressures for the different potentials are collected and given in 

 column I of Table II, and in column III of the same table the spark 

 potentials as deduced from the experimental curve in Fig. Ill are also 

 given. From a comparison of the pressures in column I with those in 

 column III it will be seen that the numbers are practically the same, 

 which goes to show that the modification here outlined for determining 

 pjiarking potentials is equally reliable with that formerly followed. The 

 chief advantage possessed by this method over that formerly used is that 

 the spark potentials are obtained by following the variations in a perfectly 

 definite phenomenon, namely the strength of the current passing through 

 the discharge chamber, whereas with tlie older method, the information 

 sought for was obtained by a series of slow and tedious trials with 

 potentials extending over a range Avhose limits were gradually brought 

 more and more together. 



TABLE II 

 Hydrogen 



From the preceding description it is obvious that the gradual 

 diminution of the pressure in the discharge chamber when the current 

 was passing could only lead to the determination of spark potentials 

 which lie on that branch of the curve 5n Fig. Ill below the critical 

 pressure. 



The spark potentials corresponding to pressures above the critical 

 pressure may, however, be ascertained by the same method if a gas be 

 admitted instead of withdrawn from the discharge chamber when the 

 current is passing. Such a proeerlnre was adopted in the case of hydrogen 

 in the present enquiry for the potentials 350 and 400 volts. The various 



