124 ROYAL SOCIETY OF CANADA 



it is possible that all the others would have exhibited similar changes, 

 had the observations been continued long enough. With these lead 

 cylinders, which were freshly cleaned the day before readings were 

 commenced, the highest maximum value of the negative charge acquired 

 was about 64 millivolts (a scale deflection of about 33 millimetres). 

 At the end of the period of sixty-seven hours, it was found that the 

 zero had drifted 6.5 mms. to the negative side. The readings given in 

 Table IX, however, were all corrected for this change in zero, and also 

 for a gradual change in the sensibdlity of the electrometer due to a 

 dropping in the potential of the storage cell used to charge the needle, 

 from 170 volts to 144 volts. 



In one or two cases where a drop from the initial maximum charge 

 occurred, it was observed that, for the same combination, on different 

 days the drop was not always in evidence. (See Table I, compare Nos. 

 28, 29, and Nos. 33, 33). After the cylinders had been cleaned, tin 

 No. 2 rose in about four hours to a maximum of 171 millivolts, and 

 then dropped to 38. The next day, in the same time, it rose steadily 

 to 82 millivolts. It is, of course, possible, that had the observations 

 been continued long enough a drop might have been observed here too, 

 and it should be mentioned that the former set of observations was 

 taken between 3 p.m. and 7.30 p.m., while the latter were taken 

 between 9.30 a.m. and 2 p.m. McKeon's explanation of these results 

 as being due to daily variations in an external cause is possible, 

 but it would seem that a more probable one is found in a change taking 

 place in the surface of the metals, perhaps arising from modifications 

 in the state of tTie atmospheric air. 



(c) Variation in the magnitude of the charge with the intensity 

 of penetrating rays. 



McKeon attributed the variations which he observed to changes in 

 the amount of ionization of the enclosed air. That this can not be the 

 cause is shewn by the fact that, while the presence of the radium affects 

 very greatly the rate at which the charge is acquired, it modifies but 

 little the maximum value. This was observed by McLennan and Bur- 

 ton, as well as by the writer. 



McLennan and Burton also found that, when the cylinders were 

 placed in a tank filled with water, the layer of water being 13 cms. thick, 

 the maximum value of the charge was unchanged. By this means the 

 natural radiation was lessened,^ and 3'et the charge remained the same. 



'Univ. of Toronto Studies, Phys. Science Series, No. 2; also, Phys. Rev., 

 Vol. XVI, No. 3, p. 184, 1903. 



