ON COLLOID CHEMISTRY AND ITS INDUSTRIAL APPLICATIONS. 29 
2. Electrical Endosmose with Pure Liquids, Coehn’s Rule. 
The early experiments on electrical endosmose taken in conjunc- 
tion with those on the migration of suspended particles (cataphoresis) 
made it appear that solids were charged negatively in contact with 
water. When turpentine was used instead of water, however, 
Quincke found that conditions were reversed and that such solids as 
quartz, shellac, silk, clay, asbestos, porcelain, ivory shavings, animal 
membranes, &c., were electropositive, with the single exception of 
sulphur. Sulphur weakly electronegative against turpentine, against 
water was the most strongly electronegative of all the solids which 
Quincke studied. 
Coehn” has considered the question raised by these differences, 
and has proposed the following empirical rule : 
When two non-miscible substances, one of which is a pure liquid, 
are in contact, the -substance with the higher dielectric constant is 
positive against the substance with the lower. 
Quincke’s data appear intelligible in the light of Coehn’s rule. 
Water and turpentine have dielectric constants of 81 and 5, respec- 
tively. Furthermore, the constant in the case of water is one of the 
largest known and we should expect nearly every substance to be 
electronegative against water. In contact with turpentine, however, 
many substances ought to be electropositive, exactly as Quincke 
found was the case experimentally. Coehn’s rule appears to hold 
fairly well for pure liquids (although there are some weak points in 
his own evidence)", but when applied to aqueous solutions, 
especially those containing dissolved electrolytes, the rule fails 
utterly. 
3. Electrical Endosmose with Solutions. Effect of Acids, Bases 
and Salts. 
In all the early experiments, as we have seen in the preceding 
- sections, water flowed to the cathode. Not only pure water showed 
this unidirectional tendency, but aqueous solutions appeared to do so 
as well, the only difference being that the rate of flow was less for 
solutions than for the pure solvent. Consequently, after the 
acceptance of the physical theory of the electrical double layer, the 
belief became general that all solids were charged negatively by 
contact either with water or with aqueous solutions. 
Nevertheless, instances!® had been recorded where aqueous 
solutions flowed to the anode rather than towards the cathode. 
Hittorf* found, for instance, that cadmium chloride solutions flowed 
to the anode through animal membrane but went to the cathode 
through earthenware, and although Perrin is usually credited with 
the first definite statement that “reversals” to the anode were possible 
and could be produced at will by suitable choice of electrolyte in 
2 Wied. Ann., 64, 227 (1898), Zeit. Elektrochemie, 16, 586 (1910). 
13 Cf, Briggs, Jowr Phys. Chem., 21, 204 (1917). f : 
14 Porrett, Thomson's Annals of Philosophy, 8, 74 (1816); Daniell, Phil 
Trans., 129, 97 (1839) ; Gernez, Comptes rendus, 89, 303, 348 (1879). 
16 Wiedemann, Hlektricitat, 2, 153 (1883); Gore, Proc. Roy. Soe. 31, 258 (1880) 
16 Zit. Phys. Chem. 39, 613 (1902) ; 43, 239 (1903). 
