PIERCE. — THE COOPER HEWITT MERCURY INTERRUPTER. 411 



1.78 ohms. A single observation that I made with the mercury inter- 

 rupter and with capacity and inductance about equal to those of Linde- 

 mann in this case gave .60 ohms. Extrapolations from the other values 

 obtained by Lindemann indicate that 

 for the same discharge potential, aud 

 corresponding capacities and induc- 

 tances, his method gives for the resist- 

 ance of the spark in air, values per- 

 haps three or four times as large as the 

 resistance of the mercury interrupter. 



On the other hand, it is seen from 

 Table VIII that the values obtained 

 by Battelli and Magri for the resist- 

 ance of a 2 mm. spark in air between 

 solid metallic terminals # are of about 

 the same magnitude as the values I 

 obtained for the resistance of the 

 particular mercury interrupter. The 

 discharge potentials in the two cases 

 were approximately the same. 



Lindemann's observations were made 

 with a single discharge, while those 

 of Battelli and Magri and those of the 

 author were made with a great number 

 of discharges following each other in 

 rapid succession. This latter arrange- 

 ment would be the condition under 

 which the resistance would be least. 

 I found, however, that a considerable 

 change in the rapidity of charging made 

 no appreciable change in the resistance 

 of the mercury interrupter. 



Figure XI. 



VII. The Vacuum of the Cooper-Hewitt Interrupter. 



In order to be able rapidly to change the vacuum in the mercury 

 interrupter and to measure the pressure appropriate for use with a 

 transformer capable of giving about 15,000 volts, the apparatus of 



* Battelli ami Magri found the same resistance whether they employed plati- 

 num-iridium or cadmium terminals. 



