THE 



AMERICAN JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



Art. XLYIII. — A Determination of the Molecular Weight 

 of Radium Emanation by the Comparison of its Rate of 

 Diffusion loith that of Mercury Vapor • by P. B. Perkins. 



[Contributions from the Sloane Physical Laboratory of Yale University.] 



On account of the very small quantity of radium emanation 

 which has hitherto been available for experimental purposes, a 

 direct determination of its density has not been possible. Sev- 

 eral investigators have, therefore, attempted to determine the 

 density by comparing the rate of diffusion of the emanation 

 with that of a known gas or vapor and applying Graham's 

 Law, according to which the molecular weights of two gases 

 should be inversely proportional to the squares of their coeffi- 

 cients of diffusion. Such experiments have led to widely dif- 

 ferent results and the values obtained have been difficult to 

 reconcile with the value which would be expected on the dis- 

 integration theory of radio-activity. This investigation was 

 undertaken to clear up, if possible, these discrepancies. 



History. 



Shortly after the discovery of the emanation, an approximate 

 estimate of its molecular weight was made by Rutherford and 

 Miss Brooks,"- by determining its rate of interdiffusion in air 

 contained in a long metal cylinder. Values were obtained for 

 the interdiffusion constant intermediate between those for carbon 

 dioxide and ether, the molecular weights of which are 44 and 

 74 respectively. They, therefore, concluded that very probably 

 the molecular weight of the emanation was between 40 and 100. 



Results obtained by Curie and Dannef, who measured the 

 interdiffusion constant for emanation passing through capillary 



* Transact, of the Roy. Soc. of Canada, 2d series, 1901-2. 

 f Comptes Rendus, cxxxvi, p. 1314, 1.903. 



Am. Jour. Sci. — Fourth Series, Vol. XXV, No. 150. — June, 1908. 

 31 



