Section 61. Apparatus and Procedure. 



155 



into a pipette filled with purified air, and this solution was run into the 

 bomb which was kept covered with a perforated watch-glass. (Some of 

 the same solution had usually been left standing in the bomb over night.) 

 The liquid was shaken around the sides of the bomb and poured out by 

 slightly raising the watch-glass. This rinsing process was twice again re- 

 peated. Then a quantity of solution sufficient to nearly fill the bomb at the 

 highest temperature (156 or 218) to which the experiment was to be 

 carried was forced into the pipette (one of suitable capacity, 113 or 104 

 c.cm., with an appropriate mark on the stem being used), and was run 

 into the bomb, whereupon the watch-glass was removed and immediately 

 replaced by the cover, which had been previously rinsed with solution and 

 placed bottom downwards on washed filtered paper, which removed most 

 of the adhering solution. Experience showed that portions of the same 

 1/30 normal (or even 1/100 normal) ammonium hydroxide solution, al- 

 though extremely sensitive to contamination by carbon dioxide, could be 

 successively introduced into the bomb in this way without showing a 

 greater difference in conductance than 0.1 per cent. In the sodium 

 hydroxide experiments, the air pressure in the bomb was reduced after 

 closing it to 2 to 4 cm. mercury, but with the solutions of the other sub- 

 stances (which were carried only to 156) the air was not exhausted. At 

 each temperature after the bomb had nearly attained the temperature of 

 the bath it was removed, shaken, and at once returned to the bath. After 

 the conductance seemed constant, a series of three readings was made at 

 5-minute intervals and the mean of these was taken; except that in the 

 case of sodium hydroxide at 218, where a slow progressive decrease of 

 conductance was observed, the maximum value was taken.* In the case 

 of this substance at 218, moreover, in order to reduce the contamination, 

 the solution was transferred from the 18 directly to the 218 bath, and 

 the values at 100 and 156 were determined by separate experiments. 

 In almost all experiments the conductance was remeasured at 18 after the 

 heatings as a check upon the contamination. 



*That the progressive change was not very large with even a 4-milli-normal sodium 

 hydroxide solution at 218 is shown by the following conductance values obtained 

 with such a solution at the times after the bomb was introduced into the 218 bath 

 that are shown in the first column of the table. Shaking the bomb caused very 

 little change in the conductance. 



