Vol. 6, 1920 
PHYSICS: L. B. LOEB 
439 
35.7 cycles per second at a pressure of 753 mm. Curves II and III, 
figure 3, show the curves obtained in pure H2 at a pressure of 730 mm. 
and using frequencies of 712 and 14.5 cycles, respectively. 
These curves are practically saturation curves, their intercept with the 
voltage axis lying very close to the origin and being nearly independent 
of the frequency of alternation. Their ordinates are, however, but half 
of those for the corresponding saturation curves. This is to be expected 
since the time during which the electrometer deflection was measured 
10 
20 
50 
60 
70 
80 
SO 
/CO 
40 50 
^ ]/ in vo/i's 
FIG. 4 
Pure N2. Press = 753 mm. I— Saturation; II— N =714; III— N =35.7 
was the same for the saturation curves as for the curves taken with al- 
ternating potentials, while with the alternating potentials the accelerating 
field was on but half the time. The ordinates of the curves taken with the 
lower frequencies are in general slightly greater than those with the higher 
frequencies. This slight difference is probably due to the fact that the 
contacts at the commutator were slightly better at lower frequencies of 
alternation. The difference observed in the case of Nitrogen was due to 
a change in the intensity of the light from the mercury arc whose operating 
potential changed by two volts between the two determinations. Such 
a change easily accounts for the difference in the ordinates observed. 
It is obvious that the carriers must be entirely electronic for the curves 
cut the voltage axis at values of X^o so close to 0§ that the values of the 
