256 
PHYSICS: R. C. WILLIAMSON 
Proc. N. a. S. 
energy of the electron, e the electronic charge, h the quantum constant, 
and V the frequency of radiation, finds application in the field of X-rays, 
either in calculating the maximum frequency of radiation excited in soUds 
by electrons having energy corresponding to the voltage V, or, conversely, 
in computing the maximum energy of the secondary electrons liberated 
from solids by radiation of frequency v. 
Work in the metallic vapors indicates that in the case of the alkaline 
metals the principal series of doublets is radiated when ionization of the 
vapor by electronic collision occurs, and the limiting frequency of the series 
is given by the substitution of the ionizing potential in the above equation. 
Conversely, when it is desired to ionize a vapor by radiation, thus liberating 
electrons, it is natural to assume that frequencies must be employed which 
are equal to, or greater than, the above limiting frequency calculated from 
the ionizing potential. The results of Kunz and Williams in caesium 
vapor are at least in accordance with this view. In the case of potassium 
vapor, the observed ionizing potential is 4.1 volts® and the corresponding 
principal series limit is 2856 A.^ 
The experiments here described were planned to verify the ionization 
of metallic vapor by radiation of optical frequencies and to investigate the 
ionizing power of the radiation as a function of the frequency. The follow- 
ing considerations must be borne in mind in determining the method : 
(a) Thermelectrons are emitted quite freely by glass and metals in the 
presence of potassium vapor at temperatures as low as 150°. Also such 
surfaces in the presence of potassium vapor, or when covered with solid 
potassium, may give a marked photoelectric emission if there is any scat- 
tered light. The procedure must then be arranged so that currents con- 
sisting of positive ions, due to ionization, can be distinguished from these 
electronic currents. 
{b) Absorption of the radiation by the vapor should be Hmited, if 
possible, to the region between the electrodes. 
(c) Where accelerating fields are used, ionization by collision of elec- 
trons with the vapor may occur, and such an effect must be distinguished 
from ionization by radiation. 
Therefore, the following method was adopted. A jet of hot potassium 
vapor was directed into a cool vacuum chamber and condensed upon the 
walls, a part of the latter being cooled by means of liquid air. A carefully 
diaphragmed pencil of light traversed this jet a short distance from the 
nozzle and then entered a conical glass tube serving as a light trap. A 
mercury arc in quartz was used as the source of radiation, and absorbing 
screens could be interposed between the arc and the vacuum chamber. 
Electrodes were so arranged that the positive ions formed in the jet could 
be accelerated through a gauze to travel against a suitable retarding field 
(Vr) to an electrode connected with an electrometer. The energy gained 
