11 



reconcile the conductivity and osmotic effects. Of course, when it is 

 possible to take hydration into account as it is hoped to do when 

 certain measurements have been completed, the equivalent conduc- 

 tivity of the ionic micelle will then be ascertained. 



On reviewing the results for all concentrations and temperatures, 

 it is found necessary in concentrated solutions to ascribe to the ionic 

 micelle a conductance which is equal to that of the potassium ion. 

 This is by no means identical with the well-known and striking fact 

 that the mobility of many mechanical suspensions as well as of colloids 

 in the electric field is as great as that of an ordinary ion, since in the 

 case of the ionic micelle of soap it is not merely the mobiUty but the 

 equivalent conductivity that is so great. 



There is a good deal of evidence for the conception that the 

 mobility of the ionic micelle increases with concentration owmg to 

 diminishing hydration due to increasing amount of neutral colloid 

 in the micelle. The most imjjortant considerations is, that other- 

 wise it would follow from the principle of mass action that within 

 a very narrow range of concentration the colloidal electrolyte would 

 pass completely into undissociated form and no longer conduct. 



The small but unmistakable difference between a potassium and 

 a sodium palmitate solution is not readily accounted for if the ionic 

 micelle is considered to be merely an aggregate of palmitate ions, and, 

 therefore, necessarily identical in the two cases. There are a number 

 of reasons for considering that the ionic micelle contains some of 

 the colloidal undissociated sodium and potassium palmitates respec- 

 tively, which would thus account for the difference in behaviour, 

 since thus the micelles are no longer identical. 



For these reasons the formula ascribed to the ionic micelle in a 

 soap such as sodium palmitate is — 



(NaP).r- (P')n -(HaO)™. 



According to this formula, the composition of the micelle must 

 alter continuously with change in concentration or temperature or 

 upon the addition of salts. Thus in very concentrated solution or in 

 presence of large amounts of another electrolyte such as sodium 

 hydroxide or chloride, the soap must be nearly all colloid of approxi- 

 mately the composition — 



(NaP).r- (P')n- (H20)m. 



Again this formula accounts for the real difference, namely, in 

 conductivity and osmotic behaviour, between solutions of potassium 

 and sodium soaps, as was pointed out above. 



With regard to the value of n, the number of negative charges on 

 the micelle, it must be at least 10, and probably is very much greater. 

 Thus the " molecular weight " of the ionic micelle must be at least 

 of the order of magnitude of 3,000, although the true molecular weight 

 of palmitate is only 255. In a similar way the enormous molecular 

 weights ascribed to various substances which occur only in the 

 colloidal form may well be derived from the aggregation of compara- 

 tively small molecules. 



