Mr Barlow, On the Osmotic Pressure of Aqueous Solutions. 235 



This experiment thus shows that when the influence of the 

 impurities dissolved from the air is provided against, the experi- 

 mental values for the osmotic pressure approach more closely to 

 the theoretical ones. 



It was after the last experiment that a preliminary experiment 

 was made in order to examine the deterioration of the outer water 

 by changes in its electrical resistance. 



By the telephone method, the resistance of the water before 

 the experiment was 12875 ohms at 18 0, 4 (the mean of five read- 

 ings). After the experiment, after having been in contact with 

 the cell for nearly three weeks, the water was again examined under 

 the same conditions : its resistance was found to be 2014 ohms at 

 19°'3 ; or one-sixth of its former value. 



Arrangements were then made to carry out such an examina- 

 tion of the water with greater care. 



Owing to other work, this examination of the resistance of the 

 outer water at intervals had to be put off for a considerable time, 

 during which cell e was washing in distilled water. 



The telephone method could not be used with sufficient accuracy 

 for the high resistances offered by the water used at first. A com- 

 mutator of the kind described and used by Fitzpatrick* and 

 Whetham-f- was set up. It was driven by a water-motor. 



Water redistilled over alkaline permanganate and then over 

 acid potassium sulphate was used. It had an absolute conductivity 

 of 3-3 x 10- 6 . 



Cell e was standing in the glass bottle used for the outer 

 vessel from July 1904 to January 1905. The distilled water was 

 frequently changed. During the last week before the experiment 

 it was standing in and was filled with the redistilled water. The 

 cell was therefore as clean as it well could be. The electrodes 

 were of platinum, covered with platinum black and then heated 

 to redness. They were between the outer vessel and the cell, and 

 about half-way down the latter. The outer vessel was closed to 

 the air by means of a cork covered with molten wax and was 

 standing in a water-bath fitted with a thermostat. The tempera- 

 ture of the bath was fairly steady, and only occasionally was the 

 temperature raised above that of the room in order to get the 

 temperature coefficient for purposes of correction for small changes 

 in the temperature of the water-bath. These corrections seldom 

 affected the observed values above the second place of decimals. 



I should say that the constant of the electrolytic cell in which 

 the conductivity of the water was obtained, was found by using a 

 standard solution of potassium chloride. Its diameter was about 

 the distance between the cell e and the outer vessel. The actual 

 resistance of the water in the electrolytic cell was about 42200 

 ohms ; that of the water, after arranging for experiment in the 



» B. A. Report, 1886, p. 328. t Phil Trans. A, Vol. cxciv. p. 330 (1900). 



