Dr. W. M. Bayliss. 



[June 30, 



same dilution is only dissociated to 52 per cent. This is probably due to 

 colloidal association of molecules in the latter case. If so, it presents an 

 interesting case for investigation, since ionisation would be some function 

 of the surface of the aggregates. 



Owing to the fact that this dye is the disodium salt of a conjugated 

 disulphonic acid, it is of interest to compare its conductivity with that of 

 disodium sulphate. The molar conductivity of this latter at infinite 

 dilution, under the same conditions as Congo red above, was found to be 

 2600 x 10." 4 recip. ohms. Sodium chloride, with a corresponding value of 

 1200 x 10 -4 , and sodium naphthylamine-sulphonate with that of 1150 x 10 -4 , 

 may be compared. 



From the values given it will be seen that in moderately dilute solutions, 

 such as those whose osmotic pressure was given in my previous paper, 

 Congo red may be considered to be a strong electrolyte. The migration 

 velocity of the anion, although it has a molecular weight of 650, must be not 

 much less than that of SO/. Biltz and v. Vegesack* calculate this migration 

 rate as 57 in the same units in which S0 4 ' is 67. 



Congo red does not obey the Ostwald dilution law, the values of the 

 "constant" showing steady and considerable diminution with increasing 

 dilution. 



It was found impossible to obtain at 25° C. solutions more concentrated 

 than 1 molecule in 14 litres, or 4'975 per cent. At this dilution the salt was 

 dissociated to the extent of 45 per cent., whereas sodium sulphate is 

 dissociated to 74 per cent. 



The Factors of the Osmotic Pressure. 



The experimental facts which have to be accounted for may be realised by 

 taking the case of Congo red at a dilution of 120 litres. I find that the 

 osmotic pressure of such a solution is slightly lower than what it would be 

 if the salt were undissociated. But measurements of the electrical conduc- 

 tivity show that it is dissociated to the extent of at least 60 per cent, at this 

 dilution, so that, if the dissociation is into two Ha" and one bivalent anion, 

 the concentration of osmotically active elements within the membrane is 

 m/66 in ions and m/300 in non-dissociated salt, or m/54 in all. 



A simple explanation of the discrepancy has been advocated by Biltz and 

 v. Vegesack/f viz., that the Ha* ions to which the membrane is permeable do 

 not take any part in the production of osmotic pressure ; so that, since only 



* ' Zeit. f. physik. Chem.,' 1910, vol. 73, p. 481. 

 t ' Zeit. f. physik. Chem.,' 1910, vol. 73, p. 490. 



