60 Mr W. Makower on the Molecular 



i. e. when £ = 0, we obtain from equation (1), by integrating, 

 the equation 



whence 



The value of X thus defined is taken as a measure of the 

 rate of diffusion of the emanation through the plug. 



3. Hates of Diffusion of Gases of known Molecular Weight. 



To determine the molecular weight of the emanation, its 

 rate of diffusion was compared with that of hydrogen, oxygen, 

 carbon dioxide, and sulphur dioxide through the same plug. 

 A pure sample of the gas to be experimented with was sucked 

 into the diffusion- vessel, and after a known time (t minutes) 

 some of it was withdrawn and analysed to find out how much 

 had escaped through the porous plug. When no statement 

 to the contrary is made, the gas was diffusing out into the 

 atmosphere. 



In deducing the value of X from these observations, a slight 

 modification had to be introduced into the method of calcu- 

 lation given above ; for, since the rates of diffusion of the 

 gases on the two sides of the plug were different, the volume 

 of the gas (at constant pressure) contained in the diffusion- 

 vessel varied during an experiment. It was important to 

 maintain the pressure equal on the two sides of the plug to 

 prevent any gas escaping by e fusion, and this was clone by 

 occasionally adjusting the level of the reservoir R. 



As above, equation (1) 



dy _ X 



dt ~~ - V y ' 



y being in this case measured volumetrically. 



Here V is not constant, but varies during an experiment. 

 JSince the variation was small, it was taken as a linear function 

 of the time, so that 



V = V (l+a*)*, 



where V y is the volume at the beginning (2 = 0), and a is a 

 constant to be determined by observing V 2 , V , and /. 



* This assumption is true except for the case of hydrogen ; but as 

 this gas was employed only in preliminary experiments the same assumption 

 was taken to be sufficiently nearly correct. 



