226 
MR. A. Ll. HUGHES: EMISSION VELOCITIES OF PHOTO-ELECTRONS. 
(2) It has been shown that the energy of the fastest electrons emitted when mono¬ 
chromatic light falls on the surfaces is proportional to the frequency of the light. 
The results are expressed in the form V= A;n —Vq, where V is the velocity measured 
in volts and n the frequency. Ladenburg’s law, that the velocity is proportional to 
the frequency, has been shown to be incorrect. 
(3) The values of h and Vq have been found directly for the elements Ca, Mg, Cd, 
Zn, Pb, Sb, Bi, and As, and the values of Vq for Se and O 2 indirectly. 
( 4 ) The values of h and for elements of the same valency change regularly with 
the atomic volume. 
( 5 ) The product of V^, into e, the charge on an electron, has been identified with 
the work required to separate an electron from the molecule. 
(6) It has been shown that the velocity distribution curves usually obtained in 
photo-electric experiments do not by any means represent the actual distribution of 
velocities of the electrons. 
( 7 ) The maximum emission velocity of photo-electrons from a continually forming 
mercury surface is identical with the velocity from a surface of mercury prepared by 
distillation. 
The spectrograph used in this research was obtained by means of a Government 
Grant througli the Royal Society. 
In conclusion, I wish to express my best thanks to Prof. Sir. J. J. Thomson for his 
interest and encouragement during the course of this investigation. 
\_Xote added Jidy 11, 1912. —-The preceding paper will be rendered more complete 
by a short reference to two papers which appeared after it was written, and which 
bear upon some of the points discussed therein. In Table V. the maxmium of the 
selective effect for Mg is given as A 1730 by Lindemann’s formula. Pohl and 
Pringsheim (‘ Yerh. d. D. Phys. Ges.,’ p. 546, 1912) have obtained the interesting 
result that a freshly distilled surface of Mg does not show any selective effect, but 
that, in less than an hour after distillation, a well-marked selective efiect appears 
with its maximum at A 2500. It was mentioned in connection with Table Y. of this 
paper that the emission velocities for Mg are quite regular, although this metal is 
different from the other elements (except perhaps Ca) in that it has a selective effect 
with its maximum within the range of wave-lengths used in this investigation. The 
emission velocities appear, therefore, to be unaftected by the presence of a selective effect. 
From a consideration of the spectrum of the source of light used by Lenard in 
his recent experiments on the ionisation of air by ultra-violet light, Lyman (‘ Phys. 
Zeits.,’ XIII., p. 583, 1912) concludes that the ionisation of air by light does not 
take place unless the light contains wave-lengths less than about A 1300. This 
confirms the experimental result which I obtained in an earlier research, and which I 
have used in the present paper to determine the ionising potential of oxygen.] 
5 SER1912 
