220 REPORTS ON INVESTIGATIONS AND PROJECTS. 



conclusively that the same result is obtained whether the initial pressure is 

 below or above the final one. When the slight difference in concentration is 

 corrected for, the osmotic pressures are identical. 



(2) Owing principally to scarcity of approved manometers and lack of 

 bath accommodations, the cells, in former years, have been opened, as a rule, 

 after having maintained an apparently maximum pressure from 24 to "^2 

 hours. During the past year it has often been practicable to keep cells in the 

 bath for much longer periods. The conclusion reached is that, if a cell de- 

 velops and maintains a given pressure for 24 hours, no further increase in 

 pressure is to be expected. Concurrently the durability of the membrane has 

 been tested. In one instance a cell containing a 0.5 normal solution was leff 

 in the bath for 60 days after it had developed maximum pressure. On the 

 sixty-first day the pressure was the same as on the second, namely, 12.522 

 atmospheres, and all the intermediate fluctuations were smaller than the 

 corresponding variations in atmospheric pressure. There is no known reason 

 for believing the membrane in this cell to have been superior to any of the 

 others employed in obtaining the pressures cited above. 



An account of the completed investigations of the past two years has been 

 published in the American Chemical Journal in a series of five papers under 

 the general title of "The relation of osmotic pressure to temperature." The 

 separate titles are given below : 



I. The manufacture of the cells employed in the measurements (vol. xlv, p. 91). 

 II. The manometers (vol. xlv, p. 237). 



III. The regulation of temperature (vol. xlv, p. 383). 



IV. The membranes (vol. xlv, p. 517). 

 V. The measurements (vol. xlv, p. 554). 



While the ratio of osmotic to gas pressure is constant for each concentra- 

 tion up to 25°, showing that the law of Gay-Lussac for gases applies also to 

 osmotic pressure, that ratio is in every case greater than unity and varies 

 from one concentration to another. In other words, there is no direct ex- 

 perimental evidence in the conduct of cane-sugar solutions at temperatures 

 not exceeding 25° that the law of Boyle likewise applies to osmotic pressure. 

 The high ratios in question are not, however, to be accepted as evidence ad- 

 verse to the law, for they may be due to some form of molecular association, 

 e. g., hydrates. It is to be expected, if such aggregations exist, that at some 

 higher temperature they will begin to break down, and that the solutions will 

 give, therefore, smaller ratios of osmotic to gas pressure than have been ob- 

 tained at 25° and below. Hence it became of importance to measure the 

 osmotic pressure of cane-sugar solutions at temperatures above 25°, and 

 work along this line has been in progress during the past year. The baths 

 hitherto employed are not adapted to work at high temperatures, and others 

 have been designed and tested which enable us to measure osmotic pressure 

 at temperatures up to the boiling-point of water. A number of determina- 

 tions of pressure at 30°, 35,°, 40°, and 50° have already been made, and it 

 has been found that, even at 30°, the ratio of osmotic to gas pressure is, for 



