ii4 PHYSICAL CHEMISTRY 



40 mv was observed. With a wooden membrane and KC1 the 

 emf was 7 mv, and the dilute side positive. 



The emf produced by porcelain membranes in Hocker's ex- 

 periments was very small, due to the fact that he waited twenty- 

 four hours for equilibrium to occur before making the read- 

 ing. He found that time to be required owing to the thickness 

 of the porcelain. Very irregular results were obtained if the 

 reading were taken soon after filling the apparatus. The author 

 observed that the bowl of a clay pipe showed negative osmose 

 and was thin enough to come to an equilibrium more quickly, 

 provided the previous solution had been thoroughly soaked out 

 of it. Briinings (op. cit.) found Nernst's formula to be approxi- 

 mated with a burned clay membrane separating n/100 and n/1000 

 NaCl solutions, the observed emf being 50 mv and the calculated 

 59 mv. Briinings obtained rapid equilibrium by soaking the 

 membrane in the more concentrated solution, in one case boiling 

 it in this solution sixteen days with a reflux condenser. Briinings 

 tried membranes of burned clay, wood, bone, carbon and marble 

 with NaQ and KC1, and in every case the dilute side was posi- 

 tive. Since the CI ion is faster than the Na or K ion these mem- 

 branes must be more permeable to cations than to anions. 

 Probably this is due to repulsion of the anions by the negative 

 charge of the membrane. 



The effect of the electric charge of the membrane on its rela- 

 tive permeability to anions and cations is illustrated by an experi- 

 ment of Mines' (1911 b). In the following concentration cell 

 the emf is zero: 



Zn, ZnS0 4 | n/8 NaCl | n/80 NaCl | n/8 NaCl | ZnSO„ Zn. 



If, however, a gelatine membrane is placed between two of 

 the NaCl solutions the dilute side becomes positive with an emf 

 of about 60 mv. The gelatine, therefore, is more permeable to 

 cations than to anions, and it is well known that natural gelatine 

 is electronegative. Gelatine is made positive by the ions of 

 polyvalent metals, and hence Mines treated the gelatine with 

 a salt of Gadolinium to test its effect on the emf. The same 

 gelatine membrane used above, after treatment with GdCl 3 and 

 being returned to the concentration cell, caused an emf of 8 mv, 

 but the dilute side was negative, showing that the membrane had 



