295 
from Hot Platinum. 
These figures are at any rate interesting as indicating the 
order of agreement between the different observations. As a 
matter of fact the probable error of any observation is much 
greater than its deviation from the mean, owing to sources of error 
which affect all the observations, so that one has to be very careful 
in drawing conclusions from them. 
<J> . 
The signification of the constant b = ^ which occurs in the 
exponential factor is equally important, since is the work done 
by an ion in overcoming the discontinuity in the potential at the 
surface layer. We obtain b from the equation 
i c i. e 
l0 ge^7- hlOgejp 
& e 
where C, C' and 0, 6' are corresponding currents and absolute 
temperatures. Substituting the values of C and G' for 0 = 1571, 
O’ = 1378 respectively we get the average value of b from 1378 to 
1571 absolute as 4‘93 x 10 4 . 
Since R is equal to 
1-204; 
x 10- 16 
we have <E> = 4*93 x ^ ^ x 10 _12 = eS(f>, where e is the charge on 
an ion and is the discontinuity in the potential at the surface 
of the metal. From this we obtain 8cj) = 1*365 x 10 -2 electro- 
static units 
= 4T volts. 
This is therefore the discontinuity in the potential at a platinum- 
vacuum surface and it is of the right order to give the contact 
E.M.F. as the difference of its value for different metals. The 
author intends to make further experiments on other conductors, 
notably iron and carbon, with the hope of confirming this part of 
the theory. 
The preceding theory only claims to represent the main 
features of the radiation from a hot platinum wire. In the first 
place it is evident that the number of ions emitted per unit area 
cannot be regarded in the strict sense as an exact function of the 
temperature. This number will evidently be altered by anything 
which changes the state of the surface, so that we should expect 
the current to depend on the previous treatment of the wire. 
Variations under this head do certainly seem to occur, but the 
amount of disturbance produced by them remains to be in- 
vestigated. 
In conclusion I wish to thank Professor Thomson for numerous 
suggestions during the course of this work. 
