Vol.. 8, 1922 
PHYSICS: L. L. NETTLETON 
353 
' Allen, Frank, 1920, Persistence of vision and the primary color sensations. Amer. 
J. Physiol. Optics, 1, 94-134. 
8 Steindler, 1906, Der Farbenempfindlichkeit des normalen und farbenblinden Auges. 
Sitz. Weiner Akad., 105, 11a, 115-6. 
** One of these instruments belongs to this Laboratory, while the other was loaned to 
us by the kindness of Professor G. H, Parker, Harvard University. 
CHARACTERISTICS OF A SHORT WAVE OSCILLATOR AT VERY 
LOW PRESSURES 
By L. L. Nejttlkton 
Department Physics, University of Wisconsin 
Communicated October 11, 1922 
Various investigators have experimented with three element vacuum 
tubes oscillating at a wave-length of the order of one meter. All of them 
have used commerical tubes, having a cylindrical arrangement of filament, 
grid, and plate. 
Barkhausen and Kurz^ discovered a type of oscillation which was ap- 
parently due to the motion of electrons in the tube itself, and was indepen- 
dent of the inductance and capacity of the external circuit. The computed 
time for an electron to travel across the tube, from filament to plate and 
back, under the potentials applied, agreed fairly well with the period of the 
waves as measured by means of Lecher wires coupled to the tube. 
Whiddington^ experimented with a similar type of oscillation, but he 
observed much lower frequencies and attributed the observed effects to the 
motion of ions instead of electrons. An explanation was given which in- 
volved a discontinuous emission from the filament. 
Gill and MorrelP have experimented extensively with the Barkhausen 
type of oscillation, using commercial tubes made by the Marconi company, 
and have given an explanation of the effect which involves a natural mode 
of oscillation of the electrical system connected to the tube. 
The work of Barkhausen and Kurz and that of Gill and Morrell was done 
with so-called "hard" tubes, which in the process of manufacture have 
been thoroughly heated to remove occluded gas and then sealed off from 
the vacuum pump. These men have assumed that they were dealing 
with a purely electronic phenomenon, and that no gaseous ionization was 
concerned. 
The essential departure in the experimental arrangement in the present 
investigation was the use of a tube which was left permanently connected 
to the vacuum pumps, and was made with a ground joint to make the in- 
