10 Cady — On the Energy of the Cathode Rays. 



as possible, but small discrepancies were always taken into 

 account. It was necessary to use great care in insulating the 

 entire circuit from earth. 



The bolometer was found to be much more sensitive than 

 the thermo-element, although its effective surface was smaller. 

 The source of error from gas-evolution was also less serious, 

 owing to the rapidity with which observations could be made. 



The process of etching and washing such small bolometers is 

 somewhat complicated when a metallic frame is used, as the 

 latter has to be protected by a coating of wax, which is after- 

 wards dissolved away. Thus it is difficult to avoid straining 

 the delicate platinum strips. This may become a very serious 

 source of error, as it is impossible to know the extent of the 

 injury. Moreover, a brownish deposit was nearly always 

 formed upon the strips, probably from some action of the acid 

 used to dissolve the silver, upon impurities in the wax. This 

 may in some way have caused a further source of error. 



Four different bolometers with brass frames were used. The 

 results are briefly given below, though in at least two cases 

 great irregularities occur. It was not until a frame of slate had 

 been made that reliable observations were obtained (see below). 



Bolometer I. Eleven strips, ll - 5x0'66 mm . Resistance w at 

 room temperature = 2621 ohms. 



The observations were all made at high pressures ; the ratio 

 i VjQ is therefore chiefly dependent upon the pressure and not 

 upon the cathode-current. 



V <?-io 5 Watt. iV/Q. 



2590 213 2-21 



2610 140 2-28 



4090 541 1-69 



4210 376 1-53 



Bolometer 11. Nine strips, 11*5 Xl'0 mm . w = 15*74 ohms. 



An influence machine was used instead of the battery as 

 source of discharge. As the range of values of i is not great, 

 onlv the average values are here given : i = I'10-IO -7 amp. ; 

 pres. = 0-020 mm ; i Vj Q = 0-81. 



Bolometer III. Seven strips, ll'5xl'# mm . w — 8-75 ohms. 



A battery of 30 cells was so arranged that it could be con- 

 nected at will between P (fig. 2) and the earth. This raised 

 the bolometer and of course the entire bridge to the potential 

 of the battery. The Faraday cylinder was kept in contact with 

 P. Readings were taken with the bolometer alternately at 

 and +60 volts. 



From the table it is evident that as in the case of the thermo- 

 element, charging the bolometer to a certain potential increases 

 the ratio i V/Q to a greater extent, the stronger the cathode- 

 current, that is, the greater the conductivity of the gas. 



Pressure. 



i - 10 7 Amp. 



0-116 



18-0 



0-113 



12-3 



0-064 



22-3 



0-055 



13-7 



