Abstract
The method used was to measure the number of electrons from a straight tungsten filament which were able to arrive at a co-axial cylindrical electrode against various retarding potentials. In order to eliminate certain disturbing factors, particularly photo-electric effects, this electrode was made in the form of a very fine grid and those electrons passing between the grid wires were collected upon an outside electrode and there measured. A rather complicated intermittent heating current arrangement allowed emission from the filament only when its surface was at uniform potential, and insured that the retarding potential had exactly the desired value. A current regulator kept the heating current constant to 1/30 percent. (1) Electrons from tungsten. Measurements of the variation of electron current with voltage were made at eight different temperatures ranging from 1440°K to 2475°K. Correction was made for the contact potential difference between filament and grid. At each temperature it was found that, except in the range of voltage where the current was limited by the space charge phenomenon, the current varied with voltage in just the manner calculated upon the assumption that the electrons leave the filament with velocity components distributed according to Maxwell's law for an electron atmosphere in temperature equilibrium with the hot filament. At 2475°K the assumed Maxwell distribution was verified up to a retarding potential so great that only one electron out of 1010 emitted electrons was able to reach the collector. It is believed that the present results are more reliable and extensive than any hitherto obtained, and that they are conclusive for electron emission from tungsten in a high vacuum. (2) Electrons from oxide coated platinum. Subsequent measurements by Dr. C. Davisson have shown that the electrons emitted from Wehnelt cathodes also have velocity components distributed according to Maxwell's law.

This publication has 9 references indexed in Scilit: