Weights of Radium and Thorium ffimanations. 



63 



which it is mixed, and still more upon the number of mole- 

 cules of >uch gas present. To test which of these hypotheses 

 is correct, experiments were made to see whether the rate of 

 diffusion of carbon dioxide through the plugs varied with its 

 pressure. The results of these experiments arc oiven in 

 Table II. 



Table 11.— U0 9 . 



Plug I. 



Plug II. 



Pressure in 

 centimetres mercury. 



X. 



Pressure in 

 centimetres mercury. 



X. 



760 



1 



•234 



76-0 



1-58 



too 



•765 



8-8 



3-28 



9-4 



•687 



7-0 



2-76 







5-9 



3-74 



From the above table it appears that a variation of X with 

 the pressure, similar to that observed with the radium ema- 

 nation, occurs with carbon dioxide. We are therefore led 

 to the conclusion that the variation in the density of the 

 emanation is only apparent, and that the pores of the plugs 

 used were comparable in size with the mean free path of gas 

 molecules. Subsequent experiments were all carried out at 

 atmospheric pressure, the upper surface of the porous plug- 

 being in contact with air. In this way, all experiments were 

 performed under similar conditions, and could therefore be 

 compared. 



A further point i- brought out by a study of Table II. It 

 will be noticed that the values of X for hydrogen and carbon 

 dioxide (at atmospheric pressure) are not in the inverse ratio 

 of the square roots of the molecular weights of these gases. 

 We cannot, therefore, assume this law to calculate the mole- 

 cular weight of the emanation from a comparison of its rate 

 of diffusion with that of hydrogen, a fact which gives an ex- 

 planation of the very high value for the molecular weight 

 obtained above on this assumption. The cause of tin; failure 

 of (rraham's law as applied to these observations is not certain 

 but the most probable explanation i> that when a gas heavier 

 than aii- i- diffusing upwards through the plug and air down- 

 wards into the diffusion-vessel, a layer accumulates on the 



