1911.] 



The Properties of Colloidal Systems. 



233 



one anion is formed and the membrane is impermeable to it, the undissociated 

 and dissociated fractions of the salt are equivalent in osmotic activity. On 

 this view it is therefore immaterial whether the salt is or is not dissociated. 



Unfortunately there are many reasons, both theoretical and experimental, 

 which show that such an explanation is inadmissible. 



It was pointed out by Laqueur and Sackur* that the Ua' ion in the case 

 of the sodium salt of caseinogen is kept within the membrane of a dialyser 

 by electrostatic forces alone. A similar state of affairs exists in the case of 

 Congo red. The Na' ions tend to pass through the membrane in obedience 

 to osmotic force. But they cannot travel beyond the distance at which the 

 electrostatic attraction of the opposite ions, which are unable to pass the 

 membrane, is equal and opposite to the osmotic pressure of the ]STa' ions. 

 The osmotic pressure produced by the non-diffusible elements, the anions 

 and the non-dissociated salt, shows itself in virtue of the mechanical 

 constraint exerted by the membrane, which allows water to pass freely 

 while holding back the bodies in question. In a similar way, the ISTa' ions 

 are prevented from escaping owing to the constraint of the membrane on the 

 opposite ions, so that the membrane must have to bear the pressure of both 

 ions.f To put it in another way, the pull of the anions on the cations could 

 not be effective unless the constraint of the membrane gave the former some 

 point d'appui, so to speak. 



We see, then, that what, at first sight, seems to be a reasonable view, 

 viz. that the fact that the electrostatic forces balance the osmotic pressure 

 of the diffusible ions prevents the manifestation of this pressure in an 

 osmometer, will not stand examination. 



Experimental evidence of various kinds, moreover, shows that all ions 

 within the membrane are osmotically active. If direct measurements of 

 osmotic pressure with membranes of parchment-paper did not give the full 

 osmotic pressure of the whole of the bodies in solution inside the membrane, 

 there would be a large discrepancy between values obtained in this way and 

 those obtained m other ways — by vapour-pressure determinations, for 

 example. The simple method of BargerJ is sufficiently accurate for the 

 present purpose. I found indeed, no difficulty in distinguishing between 

 dilutions of cane-sugar of 45 and 50 litres. Two experiments were made 

 with Congo red in dilutions of 30 and 50 litres. In the first case, it was 

 found that the dye solution had a vapour pressure very slightly higher than 



* ' Beitr. z. chem. Phys. und Path. Hofmeister,' 1903, vol. 3, p. 203. 

 t See the forthcoming] monograph on ' Colloids,' by "W. B. Hardy, for a more detailed 

 discussion of this question. 



J ' Trans. Chem. Soe.,' 1904,'vol. 86, p. 286. 



T 2 



