EMISSION VELOCITIES OF PHOTO-ELECTRONS. 
221 
is proportional to the frequency of the light. The evidence is based mairdy on obser¬ 
vations made with the wave-lengths X2537, X2257, and X 1849. The velocity V, 
measured in volts, is related to the frequency n by the formula V = hi — Y^y This 
implies that the emission of photo-electrons should cease below a certain definite 
frequency, and this was found to be the case experimentally. We may tlierefore 
consider the energy law to be established over the whole range of wave-lengths which 
are eflfective in causing the emission of photo-electrons. 
(ii.) The great differences between the results of experiments on the photo-electric 
effect, which have been published from time to time, are no doubt very largely due to 
more or less tenacious "aseous films coverincf the surfaces. The method of distillation 
in vacuo was adopted to avoid the presence of these gaseous films, as it is impossible to 
get rid of them when once they are formed. If a metal distilled in vacuo soon became 
covered with a retarding film, then we should not expect the agreement which was 
obtained between the experiments on distilled mercury and those on the continuously 
forming mercury surface, where the surface is exposed to the light as soon as it is 
formed. 
Millikan"^ and WrightI have obtained abnormally high values for the emission 
velocities from metals which were kept in a very good vacuum for several months. 
During that period the metals were frequently illuminated by intense ultra-violet 
light. Millikan explains the results by a purification of the surfaces from gaseous 
films which are considered to have .^pnarked retarding effect on the electrons. They 
also found that when these abnormally high values were obtained the emission velocity 
was a maximum (13'5 volts for Al) for the wave-length X2166. It is possible that, 
in Millikan and Wright’s experiments, the long-continued exposure to ultra-violet 
light produces a surface polarisation, for when the metal emits electrons it is acting as 
a cathode. There certainly seems to be more opportunity for unknown changes to 
take place, perhaps producing surface polarisation, than in these experiments. Here 
the measurements were usually complete in about four hours after the distillation had 
taken place. 
(iii.) Do the electrons which emerge from a plate illuminated by ultra-violet light 
all possess the same velocity, or do their velocities range from a maximum down to 
zero ? The latter would almost certainly be the case if the electrons suffered various 
amounts of scattering by the molecules in the surface before they emerged. The 
discussion in section 4 of this paper shows that the so-called velocity distribution 
curve obtained by plotting the leak against various retarding potentials is mainly a 
function of the obliquity with which the electrons strike the surrounding boundary. 
This question cannot therefore be answered from the curves given in fig. 7. In some 
experiments of Ladenburg, to which this consideration does not apply to the same 
extent, it appeared that monochromatic light produced electrons all of the same velocity. 
* Millikan, ‘Phys. Rev.,’ XXX., p. 287, 1910. 
t Wright, ‘Phys. Rev.,’ XXXIIL, p. 43, 1911. 
