624 SCIENCE PROGRESS 



The first thing that strikes one on looking at the foregoing 

 tables is perhaps the fact that copper ferrocyanide membranes 

 are capable of withstanding pressures up to ioo atmospheres 

 and over. The results obtained in both the investigations 

 described are in fact a striking commentary on the efficiency of 

 copper ferrocyanide as a semi-permeable membrane, and some- 

 what discount the recent criticism of an American chemist who 

 speaks of " so-called semi-permeable membranes." 



The pressures observed by Messrs. Morse and Frazer, by 

 Lord Berkeley and Mr. Hartley, far exceed any recorded in the 

 earlier investigations of osmotic pressure. Pfeffer, in his 

 classical researches, records a pressure of 436"8 cm. of mercury 

 for a 3'3 per cent, solution of potassium nitrate, and one of 

 307 '5 cm. for a 6 per cent, solution of cane sugar ; but even these 

 maximum pressures are greater than any observed during the 

 thirty years that have elapsed since Pfeffer's work was pub- 

 lished. The fact that very high pressures are now attainable 

 is doubtless due in the main to the care that has been bestowed 

 on the deposition and electrolytic consolidation of the membrane. 

 It is possible also that the highest pressures recorded in Table II. 

 have been attainable only because of the low temperature at 

 which the experiments were carried out. Indications are not 

 wanting that the membranes become less efficient if kept or used 

 at higher temperatures. 



The question naturally arises whether the determinations 

 of osmotic pressure obtained by the two methods, and recorded 

 in Tables I. and II. respectively, are concordant. As already 

 indicated, it is scarcely possible to answer this question satisfac- 

 torily, since the ranges of concentration in the two cases overlap 

 to a small extent only. At the same time it may be interesting 

 to compare the 0*533 normal solution in Table I. with the 0*527 

 solution in Table II. The value 1474 atmospheres is the osmotic 

 pressure of a 0-533 normal solution at 24 , and if it is supposed 

 that osmotic pressure is proportional to the absolute temperature, 

 the osmotic pressure of the solution at o° would be about 13*6 

 atmospheres, a value smaller than the pressure recorded in 

 Table II. for a less concentrated solution. Interpolation also for 

 other points on the common range of concentration indicates 

 that the pressures observed by Lord Berkeley and Mr. Hartley 

 are in general rather higher than those observed by the 

 American workers. Too much weight, however, must not be 



