EMISSION VELOCITIES OF PHOTO-ELECTRONS. 
207 
would satisfy equally well the law that the energy (or the square of the velocity) 
was proportional to the frequency. The reason for this is that Ladenburg had only 
a short range of wave-lengths available for producing photo-electrons. One has to 
decide whether the experimental results lie on a straight line or on a short piece of a 
parabola some distance away from its vertex. Kunz concluded that his results 
supported Ladenburg’s law. They are, liowever, in good agreement with the energy 
law, except in the red end of the spectrum. 
3. Experiments with Surfaces Formed in vacuo.—^It was shown in an earlier 
experiment that the maximum emission velocities of photo-electrons from surfaces of 
Zn, Cd and Hg, all distilled m vacuo, were approximately the same. It was thought 
advisable to investigate the velocities of the photo-electrons from a surface of mercury 
prepared in vacuo by the method described below. 
The mercury, on which the light* fell, was contained in the shallow iron dish C (fig. I) 
which was 2 cm. in diameter and 1 mm. deep. (The scale of the reservoir D is ten 
times less than that of the rest of the apparatus.) 
opening the tap T and allowing the mercury to 
overflow at C. The apparatus was evacuated 
before this was done, and the vacuum was main¬ 
tained by the liquid air method. The maximum 
emission velocity was measured by the minimum 
potential between B and C necessary to prevent 
electrons leaving C, and passing through the 
gauze to A. The brass cylinder E and B were 
kept at a potential of —2 volts relative to A, in 
order to prevent the escape of electrons from A 
due to reflected light. The gauze B was sym¬ 
metrically situated with respect to A and C, and 
as a potential difference of about 2'15 volts 
between C and B was sufficient to stop all the 
electrons passing from C to B, this potential could 
be taken as a measure of the velocity without any 
correction for the “ streufeld.” 
Several new surfaces were formed, and the 
maximum emission velocity was found to be 
2T5 volts. The arrangement used by Klages 
was not suitable for the accurate measurement 
of the velocities; one can only say from his tables that the maximum emission 
velocity is somewhere between 2’0 and 2'6 volts. It might be urged that a surface 
The unresolved light from the mercury arc was used in this experiment. The shortest wave-length 
was therefore A. 1849. 
New surfaces could be formed by 
Fig. 1. 
