70 THE ROYAL SOCIETY OF CANADA 



ever, the conductivity be small, the heat produced by the electiical 

 energy will raise the temperature of the wire until it glows brightly, 

 and the photo-electric current will, in consequence, be large. Let us 

 now see what effect reducing the pressure should have. 



According to the kinetic theory,* the heat conductivity of a gas 

 at constant temperature is independent of the pressure until the mean 

 free path of the gas molecules becomes comparable with the dimen- 

 sions of the containing vessel. Further, the loss by thermal radiation 

 will also be independent of pressure and the convection effect will 

 gradually become more and more negligible as the pressure is re- 

 duced, its influence depending, however, on the shape and size of the 

 containing vessel, f 



This holds, of course, only when the gas is kept at a constant 

 temperature, and the ideal course would be to enclose the glow-lamp 

 in a thermostat. Owing, however, to the nature of the experimental 

 method, this was impracticable. Let us then investigate briefly the 

 nature of the photo-electric current — pressure curve to be expected 

 under the actual conditions of the experiment. 



First, we see that, as time goes on, and the containing vessel 

 becomes heated, the temperature gradient will decrease, the amount 

 of heat conducted away will drop off, and the platinum wh-e will glow 

 more brightly. This effect, of itself, at constant pressure, would soon 

 end in a condition of equilibrium between the containing vessel and 

 the outer atmosphere. 



But, as noted above, the transfer of heat by convection will de- 

 crease to zero as the pressure decreases, and will depend, moreover, 

 on the dimensions of the containing vessel. There will thus be a 

 gradual rise in the temperature of the wire, quite apart from the time- 

 effect mentioned. Since the photo-electric current will depend on 

 the visible radiation only, and since this latter will obviously form 

 a larger and larger fraction of the total energy emanating from the 

 wire as the temperature rises and the wire passes from dull red to in- 

 candescence, we see that the resultant curve between photoelectric current 

 and pressure will exaggerate the increase of temperature with decrease 

 of pressure. Thus, instead of the theoretical straight line, parallel to 

 the pressure axis, and indicative of constant conductivity we must look 

 for a curve tending more and more abruptly from the axis of pressure. 



II. Apparatus. 



For the investigation, a photo-electric cell of the Elster and Geitel 

 type was employed. It is shown in Fig. 1 and consisted of a surface of 



*The Kinetic Theory of Gases, O. E. Meyer, p. 287. 

 fKundt and Warburfî, Pofrjr. Ann. Bd. 156. 



