Stokes' Law, that the frictional resistance to motion of a definite 

 amount of substance depends directly upon its degree of subdivision. 

 Thus, if say ten simple palmitate ions unite to form one single particle 

 carrying ten electrical charges, the resistance to movement under the 

 action of the electric current is considerably diminished and therefore 

 one should expect greatly enhanced conductivity. This is, indeed, 

 observed to a very appreciable extent in some cases, but it is partly 

 off-set by the action of this enormous aggregation of electrical charges 

 in condensing upon the particle large amounts of water and, indeed, 

 any other constituent available in the solution such, for example, as 

 undissociated colloidal soap. This also obviated the difficulties 

 advanced by Bayliss^* in proposing his tentative assumption of a mere 

 aggregation of the anions of Congo -red. 



This heavy hydration is held by many authorities to be a plausible 

 explanation of the high viscosity frequently exhibited by substances 

 of this class as well as the effect of varyuig conditions upon this 

 viscosity, and it also accounts for the effects of temperature, for 

 example, the high temperature coefficient of the conductivity. 



The recognition of this class of colloid in spite of the assistance of 

 isolated data for particular cases, was long delayed owing to the 

 supposed irreconcilability of these properties and to their being 

 ascribed to the presence of impurities and hydrolysis, &c. It will 

 be seen that the results have now been experimentally established 

 beyond reasonable doubt. 



It is worth while noting that a colloidal electrolyte differs wholly 

 in its behaviour from the class of substances represented by the 

 dextrines which may be termed " semi-colloids." The semi-colloids 

 are non-electrolytes exhibiting various degrees of osmotic activity, 

 ranging from that of a typical crystalloid such as dextrose down to 

 that of a colloid such as starch. 



It is also worth while mentioning that cataphoresis has often 

 been confused with conductivity. Most ions and charged colloidal 

 particles and even coarse suspensions exhibit a velocity of cataphoresis 

 of the same order of magnitude, but it is only in the case of the ions 

 and a few selected cases of colloids that this has been regarded as 

 being identical with conductivity. In dealing ^\dth colloids, in no 

 case was a conductivity predicated greater than might be expected 

 for a very slow-moving complex ion, whereas the theory of the ionic 

 micelle predicts enhanced conductivity, and in the case of soap 

 experiment shows that ionic micelle has to be recognised as being 

 several times more mobile than the ion 5 from which it is derived. 



The writer incidentally considers that it has not yet been proven 

 that there is any difference in kind between cataphoresis and ionic 

 migration, except that in an ion the mmiber of electrical charges is 

 e([ual to the number of equivalents of substance in the ion. If so, 

 electrical endosmosis would be a result of solvation. Quantitative 

 data are being sought in order to test this point. 



II. — Constitution in Alcohol. 



In alcohol soaps exhibit a wholly different and much simpler 

 behaviour. The soap here exists in the form of a simple unpolymerised 



A 2 



