Cathode Fall from J lot Carbon. 195 



to 300 volts. At the same time a Braun electroscope was 

 kept connected across the fixed electrodes as a check on the 

 total potential-difference. 



The discharge was maintained by 500 small secondary cells 

 (Messrs. W. G. Rye & Co.'s large capacity type). A variable 

 resistance consisting of cadmium electrodes dipping into a 

 solution of cadmium iodide in amyl alcohol in a U-tube was 

 included in the circuit: and by its means the current could be 

 regulated with great precision with the aid of a fine adjustment. 

 A telephone was also in series to make sure of a steady 

 discharge. The current was measured by shunting an 

 Ayrton & Mather D'Arsonval galvanometer across a vari- 

 able and known resistance in the main circuit. In the 

 following tables and curves the deflexions obtained with the 

 different shunts have all been reduced to the ecpaivalent 

 deflexions of the one particular shunt (17 cms. of platinoid 

 wire) which came in for the largest share of use. With it a 



e> 



deflexion of 28*75 cms. corresponded to a current of 10 — : * 

 ampere, which is therefore the factor for converting all the 

 deflexions given below into amperes. 



The anode A consisted of a short carbon filament bent into 

 the form of a simple hoop. The cathode was made of a 

 carbon filament of approximately circular section 1 mm. in 

 diameter. Its total length was 6 '5 cms., bent into the shape 

 and dimensions represented in fig. 1, K. From these dimen- 

 sions its total surface may be estimated at about 2*05 sq. cms. 

 It was ascertained, by making the movable electrodes pass 

 through the middle hoop, that the electric field was symme- 

 trical on the two faces of the cathode. 



The heating of this electrode was accomplished by means 

 of a current supplied from a battery of portable storage-cells 

 which were placed on a platform supported by paraffin blocks, 

 so that the whole system was completely insulated. The 

 current could be varied from nothing up to b" amperes, and 

 accurately adjusted to keep the temperature constant by 

 means of a very convenient form of rheostat similar to that 

 described by Mr. Se;irle in the Phil. Mag. July, 1903. It 

 was noticed that if a certain current raised the filament to 

 any particular high temperature before or after the discharge, 

 then, while the discharge was passing, it was necessary to 

 increase the current through the electrode in order to maintain 

 the same temperature. /'. e. to prevent its resistance increasing. 

 This seems to indicate that the surrounding gas on becoming 

 ionized, and so a conductor of electricity, becomes at the 

 same time a better conductor or " convector " of heat, and so 

 help- to cool down the glowing electrode. Or, in other 



2 



