Perkins — Molecular Weight of Radium Emanation. 473 



(7) Three similar experiments with mercury at 275° C. k = 



•0407 ± -00032. 



(8) A series for mercury vapor at 250° C. under constant supply. 



Average amount diffusing per min. = -00456 ± -000031. 



Series (8) was used to test the constancy of the porous 

 plug by comparison with (3). 



Final Calculations for the Molecular Weight of the Ema- 

 nation. — A slight increase in the porosity of the plug appears 

 to have taken place as the experiments proceeded. By taking 

 the mean of the different sets of determinations for a certain 

 diffusion constant, any small error due to change in porosity 

 should be approximately eliminated. 



Average (k) at 250° C. for emanation = -0341. 



" (k) . " " mercury vapor = -0370. 



Applying Graham's Law, the molecular weight of the ema- 

 nation equals 



•0370 2 



200 = = 235. 



•0341" 



Average (k) at 275° C. for emanation = -0376. 



" (k) " " mercury vapor = -0407. 



•0407 2 

 Molecular weight of the emanation = 200 — - - = 234. 

 ° -0376 a 



Conclusion. — These experiments leave little doubt that the 

 molecular weight of radium emanation is greater than that 

 of mercury vapor (200). Although the actual value obtained 

 (235) is larger than the atomic weight of radium (226*5), yet 

 the uncertainties of the diffusion method render it likely that 

 this is due to experimental errors. It is, however, evident that 

 the emanation has an atomic weight in the immediate neigh- 

 borhood of radium— a result to be expected, on the disintegra- 

 tion theorj 7 , but which previous diffusion experiments had not 

 confirmed. 



It is with pleasure that I acknowledge my indebtedness to 

 Prof. H. A. Bum stead for suggesting the investigation and 

 for valuable advice during its progress ; also to Prof. B. B. 

 Boltwood for kind assistance throusrhout the entire work. 



