28° Messrs. Richardson, Nicol, and Parnell on the 
Now, if w is the mean velocity, the number of collisions per 
second is proportional to N?w. But Ap « 6,4,=C?6e—7”8 and 
w « 6, so that the ionization would vary as 
em A 
Ge *P. 
This formula agrees well with the results of Wilson, who 
found for g. at 133 mms. the value 36,000, whereas the number 
given in this paper is 36,500. The weak point in this 
argument lies in the assumption that the total solubility of 
the hydrogen falls off with increasing temperature, since we 
have seen that it is exceedingly probable that at the temper- 
atures at which the experiments in the present paper were 
made the solubility rises very rapidly with temperature. 
Reference to Wilson’s paper* will, however, show that such an 
assumption is necessary to explain the Jag of the electric leak 
behind changes of temperature. We shall not dwell further 
on this somewhat intricate subject as we hope to make 
further experiments bearing on the matter. 
So far as we have been able to discover, the first suggestion 
that hydrogen undergoes dissociation in passing through hot 
metals was made by Ramsay 7, to explain some facts in 
connexion with the diffusion of hydrogen through palladium. 
The same opinion was also put forward somewhat later by 
Hoitsemat. Winkelmann (loc. cit.) appears to have been 
the first to apply this idea to the case of platinum. 
The principal results of this investigation may be summed 
up as follows :— 
The rate of diffusion of hydrogen through the walls of a 
hot platinum tube which is maintained at zero pressure (or 
exposed to the air) on one side is very approximately pro- 
portional to the square root of the pressure of the hydrogen 
on the other side over a range of pressure from 1 to 760 mms. 
This result can be explained by supposing that the hydrogen 
is dissociated and that the dissociated atoms pass freely 
through the platinum. Such an assumption probably involves 
the presence of a certain amount (which may be vanishingly 
small) of both molecules and free atoms both inside and 
outside the platinum. 
There is no clear evidence of the existence, either of any 
appreciable fraction of dissociated gas, even at the lowest 
pressures, outside the platinum or of any appreciable fraction 
of the gas passing through in the combined state. 
The results cannot be explained on any theory which makes 
the_number of diffusing particles proportional to the number 
* Phil. Trans. A. vol. ccil. p. 266. 
+ Phil. Mag. vol. xxxviil. p. 206 (1894). 
{ Zettschr. f. Physik. Chem. xvii. p. 40 (1895). 
