280 
DR. FRANK HORTON ON THE ORIGIN OF THE 
(1) The emission from calcium is increased when traces of oxygen are present; 
(2) After much use a Wehnelt cathode becomes “exhausted.” The oxide gradually 
disappears and the metallic surface of the supporting strip is perceptibly 
altered, probably owing to the action of the liberated calcium or oxygen; 
(3) When lime is heated so that no current goes through it there is no large 
emission of electrons. 
The only one of these which appears to the writer to be conclusive is the 
last. The first result follows from the fact that the emission from lime is much 
. greater than that from calcium,* and it is not necessary to attribute the emission to 
the process of oxidation. The second result supports the view that electrolysis 
takes place, but does not support the theory further than that; but from the third 
result it appears that electrolysis is essential in order to obtain a large electronic 
emission from lime. 
The experiments described in the present paper were performed to test the 
“ chemical action ” theory of the working of an oxide cathode. Fredenhagen 
experimented with a mixture of lime and magnesia which is a better conductor than 
lime alone. The experiments about to be described were made with Nernst filaments 
which consist of a mixture of oxides having a much larger conductivity than the 
mixture used by Fredenhagen. It has been found by Owen! that these filaments 
give a large electron emission at high temperatures. The following points have been 
investigated :— 
(1) Does the electron emission from a Nernst filament depend upon whether the 
heating current is alternating or continuous ? One would expect that, if the 
conduction through these filaments is electrolytic, the re-combination of the con¬ 
stituents would be more complete with an alternating current than when a continuous 
current is employed ; so that, if the electron emission depends upon this re-combina¬ 
tion, the thermionic current at a given temperature should be greater in the 
former case. 
(2) Is the emission from a Nernst filament the same when it is heated by 
conducting an electric current and when it is heated by a metal wire passing through 
it, the wire being raised to incandescence by an electric current ? In the latter case 
practically the whole of the heating current is carried by the metal wire, and any 
electrolysis which occurs is due to the thermionic current; so that re-combination 
must be very much less under these circumstances than when the filament itself 
conducts the heating current. 
* F. Horton, ‘Phil. Trans.,’ A, vol. 207, p. 149, 1907. 
t G. Owen, ‘Phil. Mag.,’ vol. 8, p. 230, 1904. 
