1911.] 



The Properties of Colloidal Systems. 



253 



very dilute solution of the dye, and required boiling for two hours with a 

 reflux condenser, presumably of glass, in order to regenerate it. I think it 

 quite possible that sufficient alkali was dissolved from the glass to account 

 for the result. 



There is, also, apart from the action of carbon dioxide, a slow aggregation 

 in solutions of Congo red, and especially of benzo-purpurin. This change is 

 reversed by heating. For example, a dilution of the latter dye of 94 litres, 

 which had remained for some time in the osmometer with frequent 

 changes of water free from carbon dioxide, had an osmotic pressure of only 

 68 mm. Hg, or 35 per cent, of the " molecular." After heating to 86° C. the 

 osmotic pressure rose to 175 mm. Hg, or 88 per cent, of the " molecular." 



The Temperature Coefficient of the Osmotic Pressure of Congo Bed. 

 Some incidental observations on the effect of temperature seem to show, 

 as far as they go, that the osmotic pressure of this dye is proportional to 

 the absolute temperature. A solution which had an osmotic pressure of 

 123 mm. Hg at 28-5° C. showed one of 138 mm. Hg at 62° C. Now 



1. No hydrolytic dissociation is to be detected in solutions of Congo red, 

 and merely a trace, if any, in sodium caseinogenate. 



2. On the other hand, electrolytic dissociation occurs to a large degree. 

 Carefully purified Congo red in dilutions of 28 litres is 50 per cent, ionised ; 

 in 500 litres, 80 per cent. Although considerable, however, it is not so great 

 as that of sodium salts of other organic acids of small molecular weight at 

 corresponding dilutions, probably owing to colloidal aggregation in the case 

 of the solutions of the dye salt. 



3. The osmotic pressure found experimentally, both by direct measurement 

 and by vapour pressure, is, throughout a wide range of concentration, 

 uniformly between 95 and 100 per cent, of what it would be if no dissociation 

 existed. Since it should be from one and a half to three times this value, 

 according to the concentration, it is plain that there are some abnormal 

 conditions present. 



4. The sodium ion being kept within the membrane merely by electrostatic 

 forces, it might be supposed that it is inactive in the production of osmotic 

 pressure. The agreement of vapour pressure with direct determinations is 

 sufficient to show that this is not the case. Moreover, Chicago blue, con- 

 sisting of a single large non-diffusible anion, like Congo red, but with four 



123 x 



273 + 62 

 273 + 28-5 



= 137. 



Summary of Conclusions. 



