28 PROF. 0. W. RICHARDSON ON THE EMISSION OF ELECTRONS 
we have ascertained the position of the true zero at that time as lying between the 
limits — 1 -544 and — 1 -639 volts. 
The next point is to ascertain what was the position at that time of the chemical 
characteristic curve in relation to the volt scale used. Relevant chemical data were 
taken at the same time as the photo-electric data given in Table VI. An examination 
of these data shows that with —1-00 volts the chemical current was then 34 per cent. 
of its saturation value at — 3 volts. It follows that for any C0C1 2 characteristic curve 
for this apparatus the true zero will lie between the limits of 0-54 and 0 -64 volts to 
the right of that voltage for which the current is equal to 34 per cent, of the saturation 
value. Applying this to the curve, points thus :—- x , in fig. 12, the two limiting values 
are given by the two broken vertical lines at — 1 *78 and —1-88 volts respectively. 
The inclined broken line represents the tangent to the characteristic at the point of 
inflection. It will be seen to intersect the horizontal line through the unit 3-volt satura¬ 
tion current value at the same point as the vertical line through the right-hand zero 
limit, to the accuracy within which the lines can be drawn. Thus the true zero lies 
between the intersection of the tangent at the point of inflection and the horizontal 
saturation current line and a point one-tenth of a volt to the left of this intersection. 
These results support the conclusions already drawn from less complete data in the 
experiments with the spherical anode. In that case the true zero was only located 
to within 0 -58 volt, but it was definitely placed to the left of the corresponding inter¬ 
section. It therefore seems reasonable to assume that either with spherical or cylin¬ 
drical anodes the true zero in this potential scale lies a fraction of a volt to the left of 
the point of intersection of the tangent at the point of inflection with the horizontal 
saturation line. Considering the nature of the experiments, the location of this zero 
