•206 
MR. A. Ll. hughes ON THE 
1‘86 volts, and for zinc 1’12 volts. The source of light used by Ladenburg was a 
mercury arc. His metals were polished with emery and oil, and were exposed to the 
atmosphere for some time before the apparatus for measuring the velocity could be 
exhausted. 
Millikan and Winchester"^ in their investigations on the photo-electric effect 
found enormous differences in the emission velocities from different metals. The 
values range from 1'34 volts for silver down to 0 volt for lead. Their apparatus was 
much more complicated than the type of apparatus usually used for measuring the 
velocities, and the interpretation of their results is correspondingly more open to 
doubt. As in Ladenburg’s experiments, the newly polished surfaces were unavoid¬ 
ably in contact with the atmosphere for some time before the apparatus was 
exhausted. These experiments and others show that the state of the surface is of 
great importance in photo-electric investigations. 
Such results led to a search for methods of preparing surfaces which would show 
greater regularity in their photo-electric behaviour. A promising way of obtaining a 
clean uncontaminated surface seemed to be to sputter away the surface-layer of a 
metal in vacuo, v. Baeyer and GehrtsI' found that the emission velocities from 
such surfaces were much increased, and concluded that the velocities were the true 
velocities with which the electrons left the metal. Later experiments;}; showed that 
this view was incorrect, and that the electric discharge polarised the surface of the 
metal in such a way that the electrons were accelerated on passing through the 
surface film. 
New surfaces of liquids can easily be prepared in vacuo. It seems very probable 
that such surfaces are (at all events initially) free from gaseous films which may 
retard or accelerate the electrons passing through. Klages§ used surfaces of 
mercury formed.Li vacuo, and Kunz|| used similar surfaces of sodium-potassium alloy 
and of caesium. Surfaces of metals distilled in vacuo were used by the author,^ 
and far more consistent results were obtained with them than vrith the surfaces 
treated as electrodes. 
The law connecting the maximum emission velocity of the photo-electrons with the 
wave-length has been investigated by Ladenburg'11 for Cu, Zn and Pt, and by Kunz"^* 
for Na-K aUoy and Cs. Ladenburg concluded that the velocity of the electron was 
proportional to the frequency. JoFFEtt pointed out that the experimental results 
* Millikan and Winchester, ‘Phil. Mag.,’ XIV., p. 188, 1907. 
t V. Baeyer and Gehrts, ‘Verb. d. D. Phys. Ges.,’ p. 870, 1910. 
I Hughes, ‘Proc. Camb. Phil. Soc.,’ XVI., p. 167, 1911; v. Baeyer and Tool, ‘Verb. d. D. Phys. 
Ges.,’ p. 569, 1911. 
§ Klages, ‘Ann. d. Phys.,’ XXXI., p. 343, 1910. 
II Kunz, ‘ Phys. Rev.,’ XXXIIL, p. 208, 1911. 
H Ladenburg, loc. cit. 
** Kunz, loc. cit. 
tt JoefL ‘Ann. d. Phys.,’ XXIV., p. 939, 1907. 
