DIFFUSION OF SALTS 195 



some days the bottle was removed, and the quantity of each 

 salt which had diffused out of the bottle was ascertained. 

 The comparative diffusibility of the two salts, under perfectly 

 equal conditions, was thus ascertained. By taking the 

 diffusibility of one salt as a standard throughout each series 

 of trials in which a variety of salts was employed, the com- 

 parative diffusibility . of these salts was ascertained. The 

 ' diffusion coefficient ' for each salt was calculated by a formula 

 given in Marignac's paper. The numbers given in Table XXXI 

 are comparative only, the standard salt in each series being 

 taken as unity. The comparison holds good only between 

 the numbers contained in the same column. 



TABLE XXXI 



KELATIVE DIFFUSION COEFFICIENTS OF SALTS (MAEIGNAC) 



Although the plan of diffusing two salts together, when it 

 is desired to ascertain their relative difFusibility, has many 

 advantages, the results obtained are not quite those which 

 would be found if the salts were diffused separately. In an 

 experiment with a mixture, the rate of diffusion of the more 

 diffusible salt is generally very nearly what it would be if 

 diffused alone, while the rate of diffusion of the less diffusible 



o 2, 



