290 
DR. FRANK HORTON ON THE ORIGIN OF THE 
The fact that the negative emission from a Wehnelt cathode often takes some time 
before it reaches its full strength is considered by Fredenhagen to support his view 
of the action of these cathodes. An investigation of this effect was therefore made in 
the case of the Nernst filament material heated upon platinum. For this purpose a 
strip of platinum foil was welded on to the platinum leads of the discharge tube and 
was covered with the material of a Nernst filament in the manner which has already 
been described. The temperature of this strip was determined by means of a thermo¬ 
junction welded on to it, and the thermionic emission was tested under different 
potential differences and with different gas pressures in the apparatus. At a low 
pressure and with 40 volts applied potential difference the negative emission was, at 
first, very much smaller than that from platinum alone ; the measured current at 
1440 J C. being 10~ 8 amperes cm. 2 , and at 1500° C. 3'2xl0~ 8 amperes cm. 2 —values 
which remained fairly constant during two hours’ heating. 
The positive emission was also tested, and this was very much larger. At 1260° C. 
it began by being 2'3 x 10 -5 amperes cm. 2 and decreased to 17 x 10 -6 amperes cm. 2 after 
heating for 40 minutes, the platinum strip being charged to 4-40 volts throughout 
this period. It was found that leaving the strip charged positively for some time did 
not cause any increase in the negative emission when tested immediately afterwards, 
but that after the strip had been charged negatively for a few minutes, the positive 
emission was abnormally large. On Fredenhagen’s view of the activity of the oxide 
cathode, the passage of the feeble negative thermionic current from the platinum strip 
is supposed to start the electrolytic decomposition of the oxide, and the re-combination 
of the elements separated by electrolysis is supposed to give rise to a further electron 
emission, thus increasing the current through the oxide layer and causing the gradually 
increasing effect sometimes observed with these cathodes. The material of a Nernst 
filament certainly appears to suffer electrolysis when a current flows through it, so 
that electrolytic decomposition may be produced by a thermionic current flowing 
through the material on the platinum strip ; but although a strong positive thermionic 
current was measured at 1265° C., no negative emission could be detected at this 
temperature on reversing the direction of the electric field. This experiment shows 
that the activity of the oxide cathode does not arise in the manner suggested by 
Fredenhagen. 
On the day following that upon which the observations just recorded were made, 
the effect of a stronger electric field was investigated. A potential difference of 209 
volts was used between the electrodes and the temperature of the platinum strip was 
gradually raised to 1500° C. The negative thermionic current at this temperature 
was l‘9xl0~ 7 amperes cm. 2 . It began to increase, very slowly at first, but after 
about 10 minutes a sudden rapid increase to l'3x 10 -6 amperes cm. 2 took place, after 
which the slow increase again continued during a further 10 minutes’ heating which 
was given. The apparatus was then left until the following day when the tempera¬ 
ture was at once raised to 1500° C. The negative emission was 2'0x 10~ 6 amperes 
