82 PHYSICS: R. A. MILLIKAN 
The data on lithium shown in figure 2 seems to establish assertion 1, 
the difference between curves I and II taken with the mercury hne 
4339 and a Hilgar monochromator showing how the true shape (II) 
of the photo-current-potential curve was entirely falsified by a httle 
stray short wave-length fight (see I) until a filter of aescuHn which cut 
out all waves of shorter wave-length then 4339 was used. 
Assertions 2 and 3 are strikingly verified in figures 3 and 4, the latter 
of which is the plot of the intercepts shown in figure 3, against the 
frequency. These figures relate to measurements on sodium, in which 
the saturation currents were from 75 to 500 times the largest currents 
in figure 3. These latter correspond to a deflection of 80 mm. in 30 
seconds. 
Similar measurements on Hthium are shown in figures 5 and 6. The 
mean value of Planck's h thus photo-electrically determined should not 
be in error by more than 0.5 per cent. 
This value is 
h = 6.57 X 10-27 
(see figures 4 and 6). The value of e in- 
volved in this determination of h is the 
author's value^ 4.774 X lO-^^. 
The correctness of assertion 4 is shown 
in figure 6 by the coincidence of the inter- 
/(= .70 .65" .do^SS .SO ju, cept 1^0 = 57.0 X 10^^ (see dotted line) with 
^ the direct determination of vq with satura- 
FIG. / . 
tion currents shown on the lower right 
hand corner of figure 6. To obtain the intercept the Vqv line is dis- 
placed in the direction of positive potentials by the amount of the 
measured contact EMF. Also in the case of the sodium the shown 
in figure 4 corresponds to the wave length 0.68/*. The direct determi- 
nation of this long wave length limit for the sodium is shown in figure 
7. The agreement is perfect. 
Assertion 5 was tested for two different surfaces of lithium and one 
of sodium and the contact EMF computed by the equation in 5 agreed 
in each case with the directly observed contact EMF to within less than 
2%. The details of this test will be found in the papers in the Physical 
Revieiv. 
Assertion 6 has not been tested in this work but Schottky^ has re- 
cently measured the contact EMF's between white hot wires and cold 
cylinders and found results which agree, within the fairly wide limits of 
