13 



dehydration through lowering of the reactivity (rapour pressure) 

 of the water upon addition of the salt and its ions ; for as 

 McBain and Salmon have indicated, the heavily hydrated micelle 

 must be extraordinarily responsive to such slight changes in the 

 availability of the water. The second factor is formation of ionic 

 micelle through the influence of the added potassium ion. The 

 third, like the second, raises the viscosity, and consists in that 

 unexplained change which, upon further addition of salt, tends to 

 produce a jelly perhaps ten thousand times as viscous in the original 

 soap. This last effect may well be due to the formation of neutral 

 undissociated colloid and its subsequent linking up as explained in 

 Section VI below. It is interesting to note that in accordance with 

 all these views such high viscosities are more quickly obtained by 

 adding further quantities of the soap itself, rather than by adding 

 equivalent amounts of salt or alkali. 



However, it must be concluded that the foregoing is merely a 

 programme or working hypothesis for further experiment (a colloidal 

 explanation used as a term of reproach), and quantitative work is 

 called for in order to test whether or not the first two effects are due 

 solely to the value of the total osmotic pressure and to the actual 

 concentration of potassium ion respectively. Colloidal chemistry at 

 present abounds in such problems and " explanations " which need 

 to be replaced by precise and quantitative conceptions and measure- 

 ments, otherwise such explanations do more harm than good. 



In the case of concentrated soap solutions or those to which large 

 amounts of electrolyte have been added, the temperature coefficient 

 of viscosity becomes extremely great, so that the viscosity at 20° 

 may be several hundred or thousand fold greater than at the boiling 

 point.. Of course, the addition of these electroljrtes tends to induce 

 gelatinisation or, in the case of higher sodium soaps, salting out. 

 Since salts effect the viscosity to different extents, depending upon 

 the nature of the salt, the effect is ascribed to the anion. Obviously this 

 requires further and quantitative elucidation. 



The effect of position in the homologeous series is very marked. 

 Even in rather dilute solution the sodium behenate (C22) is highly 

 viscous, whereas in the case of the laurate (C12) only the most 

 concentrated solution containing large amounts of electrolyte can 

 really be termed viscous. A normal solution of potassium laurate 

 at 20° and at 90° exhibits 8-4 and 2-8 times the viscosity of water, 

 whereas a normal solution of potassium oleate is 1573 and 3-80 times 

 as viscous as water at the same temperatures. 



Potassium or sodium oleate (Cig) at room temperature is very 

 much more viscous than say potassium mj^istate (Ci4). It is 

 probably much Like the stearate (Cig) except that the effect of the 

 double bonds is to render it Uquid even at the freezing point. The 

 effects of additions of an electroljd^e is much more pronounced in 

 the case of the oleate than the laurate, and again in concentrated 

 soap solution as compared with dilute soap solution. In other words, 

 it is evident that the tjrpical effects of added electrolytes are to be 

 attributed to their effect upon^the ionic micelle and its development, 

 concentration and composition. 



