[MCLENNAN] DIFFUSION OF ACTINIUM EMANATION 17 
If, however, we assume that K, the coefficient of diffusion of the 
emanation, varies inversely as the square of the pressure of the gas 
into which it is diffusing, 7.e., that 

eye 
ee 
q TD A 
then we have P — "pie ! 
VAR 
) 
and with the condition of maximum deposit = — 0 applied 
ap 
we have p M ve] 
À, 
À 1 
1 
which is the law observed by Kennedy. 
There seems no sufficient reason, however, for assuming the inverse 
square law for the coefficient of diffusion of the emanation and it 
would seem, therefore, that we must look elsewhere for an explanation 
of the discrepancy. 
In the argument presented above we have tacitly assumed (1) 
that q, the rate at which the emanation leaves the salt, is the same 
for all pressures, and (2) that a direct proportionality exists between 
the concentration of the positively charged active deposit particles 
and the concentration of the emanation, for all pressures of the gas in 
which the emanation is diffusing. 
If the active deposit particles became positively charged directly 
in the process of their creation from the emanation then such an assump- 
tion as the latter might be warrantable. If, however, these deposit 
particles gain their charge from the ions produced in the gas by the 
radiation emitted by the emanation in the process of transformation 
(as would seem to be true from the results of Schmidt*) then a factor 
depending on the pressure of the gas should be introduced into the 
relation connecting the concentration of the emanaton and that of the 
positively charged active deposit particles at any point. For the 


* Phys. Zeit., 9 Jahr., No. 6, p. 184, 1908. 
