OSMOTIC PRESSURE AND VISCOSITY 235 



ence in viscosity (Fig. 62) is more striking than the difference in 

 osmotic pressure (Fig. 69). This is possibly connected with the 

 fact that the lowering in viscosity due to heating to 45C. was 

 measured immediately after the temperature had reached 20C. 

 again, while the osmotic pressure of the same solutions was 

 measured after the solutions had been standing for 22 hours at 

 20C. During this time a considerable formation of sub- 

 microscopic particles of solid jelly had probably occurred in the 

 solutions previously heated to 45C. 



It was expected that when we put a collodion bag filled with a 

 1 per cent solution of gelatin of e.g., pH 3.5, which had been kept 

 for 1 hour at 45 and cooled to 20 into a beaker containing a 1 

 per cent solution of the identical gelatin chloride solution of pH 

 3.5, but which had not been heated to 45C. before being brought 

 to 20C., water would diffuse from the latter into the former 

 solution. This experiment was carried out with a positive result. 



These experiments support the idea expressed in the preceding 

 chapter that protein solutions are true solutions which may or 

 may not contain solid particles of protein capable of swelling. 

 In the case of gelatin solutions the formation of submicroscopic 

 particles of solid jelly from isolated molecules or ions is a reversi- 

 ble process, and this leads in this case to a reciprocal variation of 

 osmotic pressure and viscosity of such solutions. 



This probably explains a phenomenon which has puzzled the 

 writer for a long time, namely that the osmotic pressures of 

 gelatin solutions of the same pH and concentration of originally 

 isoelectric gelatin showed occasionally variations for which he 

 could not account. It now becomes probable that this was due 

 to a factor which was not taken into consideration, namely, 

 that on standing at room temperature a gradual transformation 

 of isolated molecules or ions into larger aggregates takes place, 

 which must diminish the osmotic pressure but increase the 

 viscosity. This source of variation was eliminated in the viscos- 

 ity experiments in which the gelatin solution was always heated 

 first to 45C. and then as soon as this temperature was reached 

 the solution was cooled to the temperature desired for the viscos- 

 ity measurements. It is probable that the same uniformity of 

 treatment is also required for the osmotic pressure experiments. 



This reciprocal relation between osmotic pressure and viscosity 



