484 Mr. W. C. M. Lewis : Experimental Investigation 



Calculation of the Adsorption Coefficient : — 

 As before, the expression for the adsorption is 

 -p _ c do 



~ ~~W£dc~' 



For a solution of concentration 0*114 per cent, (anhydrous) 

 caffeine the right-hand side of this expression has the follow- 

 ing numerical value : — 



c = 0*114 per cent. = 0*00114 grm./c.c, 



•d 2 x 4*2 x 10 7 

 = 194 ergSj 



T = 288° abs., 



do _ 3 dynes/cm. 

 ST ~~ 0*00114 grm./c.c.* 



J^___ 9-4 v 10-3 



*• UTdc ~^ xW 



It will be noted that this value is approximately of the 

 same order as the similarly calculated values for previous 

 substances. The substitution of a non-electrolyte for a 

 simple inorganic salt does not seem therefore to have any 

 marked effect on the theoretical value for the adsorption 

 coefficient. It would seem, therefore, that in both cases we 

 are dealing with a factor of the same magnitude as regards 



its effect on the tension and the consequent value of -j- . It 



will be remembered, however, that the results obtained in 

 previous cases pointed to the existence of certain discrep- 

 ancies between observed and calculated values of F. It is of 

 particular interest, therefore, to attempt to measure Y directly 

 for caffeine, since in this case there is no possibility of sepa- 

 ration of ions. 



Determination of T for Caffeine in Aqueous Solution. 



The emulsion method * was employed in these deter- 

 mination. 



Solutions of caffeine approximately 0*125 per cent, were 

 prepared, and 500 c.c. shaken in steamed-out glass vessels 

 for three days with about 1-2 c.c. of oil. The resulting 

 emulsions were very uniform under the microscope, the 

 average diameter of the particles being 6 x 10~ 5 cm. It is 

 worthy £>f note that this is of the same order as the average 

 diameter of the particles in the previous cases. 

 * See previous paper. 



