and the Formation of Liquid Films. 105 



ct 4-2(7. The electrical thickness, t e , was calculated by Reinold 

 and Riicker on the assumption that the conductance ^was 

 proportional to ct e . Consequently, 



ct + 2<r = cte' i 



and if r= --, the ratio of the electrical to the actual thickness, 



<r=ict(r-l). 



In the experiments c was '0164 gm. per c.c; thus, £]being 

 measured in micromillimetres, 



cr = -0082 x 10~ 7 f (*•— 1) gm. per sq. cm., 

 = '0082 t(r— 1) milligrams per sq. metre. 



The following table shows the values of the excess at 

 different thicknesses, calculated from this equation. The 

 experimental numbers are taken from tables xiv., xv., and 

 xvi. of their paper, and arranged in order of decreasing 

 thickness of film. 





Eatio of 







Optical 



Electrical to 



Surface Excess 



Eatio 



thickness. 



Optical 



(mgni. per 



recalculated. 



pp. 



thickness. 



sq. metre). 





641 



1-66 



3-46 



146 



484 



169 



273 



1-61 



388 



160 



1-91 



1-75 



332 



1-81 



2-21 



1-87 



320 



178 



2-05 



1-91 



314 



207 



2-75 



1-93 



314 



190 



232 



1-93 



307 



184 



2-12 



1-95 



301 



T84 



2-08 



1-97 



296 



1-98 



2-38 



1-99 



97 



4-47 



275 



4-02 



97 



i 



i 



4-19 



2-53 

 Mean 24 



4-02 



27-7 



5'S 



1-09 





These results show that the surface excess remains prac- 

 tically constant down to a thickness of 97 /jl/j,. The first three 

 numbers, it is true, show fairly large differences among them- 

 selves, but this is due to the experimental error being here 

 a good deal magnified in the calculation ; and an examination 

 of the fourth column, which contains the theoretical ratios 

 calculated using the mean value, 2*4 mgm. per sq. m., of the 

 excess, shows that the differences between them and the expe- 

 rimental ratios nowhere exceed the possible errors of these 



