4 Cady — On the Energy of the Cathode Rays. 



minute or so necessary to take a set of readings. This is the 

 chief cause of the irregularities in the numerical values given 

 below. 



Before the apparatus was put together, the thermo-element 

 was calibrated as follows. The radiation from the blackened 

 surface of a Leslie cube, filled with boiling water, passed 

 through a series of diaphragms and fell upon the thermo-ele- 

 ment. The constant of radiation for lampblack at 100° was 

 taken according to Kurlbaum- as 0-0176 gr.-cal/sec. From the 

 dimensions of the diaphragms could then be calculated, by use 

 of the Stefan-Boltzmann law, the amount of heat Q radiated 

 from the hot surface at temperature T to the thermo-element 

 at temperature t. Like the Leslie cube, the thermo-element 

 was also covered with a layer of lampblack. 



The galvanometer deflection s was observed after the thermo- 

 element had been exposed for one minute to radiation from 

 the cube. The amount of heat per scale division is then 

 e = Q /s , and the amount corresponding to the deflection s is 

 Q = s.e. 



During calibration the thermo-element was in air at atmos- 

 pheric pressure; when in use with the cathode rays, it was in 

 a partial vacuum, and moreover the layer of lampblack had 

 then been removed. This would tend to make the rise in 

 temperature of the junctions in the latter case greater than in 

 the former, assuming the quantity of heat supplied to be the 

 same : for the loss of heat by radiation and conduction to the 

 air must have been greater during calibration. The effect of 

 this error would be to make the observed energy of the cathode 

 rays too great ; but the error is smaller, the more the loss of 

 heat by conduction and radiation to the air can be neglected in 

 comparison to the loss by conduction through the metal of the 

 thermo-element itself. 



In measuring the cathode-current, the contact ac (fig. 1) was 

 closed, thereby connecting the thermo-element to earth through 

 the galvanometer. For the thermal circuit, ah was closed, and 

 at the same time either ce or ed, since it was necessary to con- 

 nect some point to earth. But when for exam pie cd was closed, 

 it was found that a small fraction of the cathode-current also 

 flowed to earth through the galvanometer, along the path 

 ShaGdc. Closing ce caused approximately the same deflection 

 in the opposite direction. Hence the mean of the two read- 

 ings could be taken as the deflection for the thermal current 

 alone. 



In order to eliminate the effect of heat radiated from the 

 cathode, the rays were deflected before each observation by 



* F. Kurlbaum, Wied. Ann , lxv, p. 759, 1898. 



