(1) In the absence of space charge. This effect, discovered by Schottky, which depends on the probability fluctuations of electron emission from a filament, has been measured over a considerable range of conditions in resonant circuits of which the natural frequency was varied from 8 to nearly 6000 p.p.s. The effect is much larger in the lower range of frequencies than the theory predicts. With a tungsten filament, the ratio of observed to theoretical effect $\frac{e^{\prime }}{e}$ is about.7 for frequencies above 200, but increases rapidly to 50 at 10 cycles per sec. With an oxide coated filament, the ratio increases from 1 at 5000 cycles to 100 at 100 cycles. This is interpreted to mean that the emission of electrons is not strictly chaotic but is influenced by irregular temporal changes in the cathode emissivity. In a high frequency circuit these changes become imperceptible and the emission is effectively random. (2) When current is limited by space charge the Schottky effect decreases because of the interaction of the electrons, and other disturbances may act upon the space charge so as to completely mask the remanent Schottky effect. The magnitude of the disturbances in amplifying vacuum tubes can therefore not be predicted from measurements on the true Schottky effect.

• Received 8 January 1925

DOI:https://doi.org/10.1103/PhysRev.26.71