200 Prof. Millikan and Mr. Winchester on Influence of 



Righi, seem to show that the positive potentials acquired by 

 metals under the influence of ultra-violet light all increase 

 with increasing temperature*. In Rights experiments the 

 potential of copper changed from 125 scale-divisions at 13° C. 

 to 271 scale-divisions at 80° C, and that for zinc from 111 

 scale-divisions at 16° G. to 319 scale-divisions at 84° C. These 

 experiments, however, were all made in air, and are therefore 

 open to the same criticism as that made upon experiments 

 on discharge rates which were made in air. 



The experiment was made by connecting the gauze and 

 tube to earth, the electrometer to the wheel, and then keeping 

 the latter connected to earth until an observation was to be 

 made, when the coil was started and light allowed to fall 

 upon the particular metal under examination until the needle 

 assumed a steady position. The time required for the needle 

 to attain this permanent deflexion was different, of course, 

 for the different metals. The average time in the experi- 

 ments recorded below w r as about 90 seconds. With the 

 largest deflexions used no change whatever in the reading of 

 the electrometer because of leak could be detected in five or 

 six minutes. Hence the readings represent directly the 

 maximum positive potentials assumed. The experiments 

 recorded extended over a period of more than a week. The 

 A T alues of the deflexions could be duplicated at any time with 

 an error which was never greater than one per cent, and 

 usually considerably less than that ; the gain in accuracy in 

 these observations over those of the preceding table being 

 due, first, to the fact that fluctuations in the intensity of the 

 source have no influence upon these readings [cf. § 7), while 

 these fluctuations constitute a source of error of unknown 

 magnitude in the readings of the preceding table ; and, 

 second, to the fact that the observer's error in holding on the 

 light for exactly equal intervals of time, namely ten seconds, 

 has here disappeared. 



As the table shows, the observations were not carried above 

 100° 0. because at temperatures not far above this point the 

 natural leak of the electrometer became so large as to render 

 accurate observations impossible. 



It will be noticed that the observations are extremely con- 

 sistent, and that the potential is completely independent of 

 temperature within the temperature limits employed ; the de- 

 partures from the mean values in the case of some of the 

 metals being not more than one-third of one per cent. This 

 direct proof that the velocity of the electron within the atom 

 is entirely independent of temperature, is especially interesting 

 * Righi, Journal de Physique, x. p. 538 (1891). 



