68 



Mr. W. Makower on the Molecular 



reduction makes a comparison of the results obtained with 

 the three plugs less troublesome. 



If Graham's law of diffusion were strictly applicable to 

 these experiments, the value of A\/M should be constant 

 whatever gas is employed. From an inspection of the results, 

 however, this is seen not to be the case ; the value of X VM 

 decreases as X decreases. If, however, the values of \ \/M 

 as ordinates be plotted against the values of X as abscissa?, 

 the points so obtained corresponding to the values of oxygen, 

 carbon dioxide, and sulphur dioxide are found to lie on a 

 straight line not parallel to the axis of X (as should be the 

 case if Graham's law r were true), but inclined at a small angle 

 to it (fig. 3), By linear extrapolation the value of Xy^M 



Fig-. 3. 



KM 



IB 





for the emanation is found to be 1'23 for plug I. ; the value 

 of -1362 X \/M for plug II. 1'21, and the value of -312X \/M 

 for plug III 1'205. These values are given in brackets in 

 the tables. By dividing the values of X^/^H for the emana- 

 tion by the corresponding values of X and squaring, the value 

 of the molecular weight of the emanation can be calculated. 

 The experiments with plugs I., II., and III. lead respectively 

 to the values 85'5, 97, and 99. On the assumption that the 

 emanation is a monatomic gas, it probably occupies the 

 vacancy in the periodic table in the fluorine group between 

 Mo (96*0) and Ru (102;. As this is the first vacancy in the 

 periodic table, it is impossible to assign a place in the periodic 

 table to the emanation on the assumption that it is poly- 

 atomic ; the present experiments, therefore, confirm the view 

 that the emanation is monatomic, 



