﻿962 Mr. J. Kendall on the Solubility of 



therefore used at these temperatures on account of its greater 

 convenience. A rubber cork, through which passed a short 

 silica tube, being fitted to each arm of the tube, pure air 

 was drawn through the solution during the whole period of 

 nitration. 



Immediately after filtration, and before cooling, the amount 

 of liquid filtered over was ascertained by direct weighing. 

 It was found by experiment that the difference in weight 

 obtained by weighing when hot, and by weighing after 

 closing the mouth of the vessel and allowing to cool, was 

 between 1 and 2 grammes. This, with a filtrate of 2 litres, 

 introduces an error of less than *1 per cent., which is negli- 

 gible. The saving of time is very great, as the solution 

 requires at least two hours to cool to the ordinary tempe- 

 rature, and another half-hour to be brought to the boiling- 

 point again. The actual evaporating down of the solution 

 takes two to three hours, being conducted rapidly at first 

 and more slowly as the volume decreases. When 50 to 

 100 c.c. remain, the vessel is allowed to cool, and the amount 

 of the residue found to within *1 gramme by weighing. 



The silica vessels remained practically constant in weight 

 throughout the whole series of experiments. 



The calculation of results is very simple, since all the 

 solutions are so dilute that their densities may be taken as 

 equal to that of water at the same temperature. Thus, if 



N 

 2000 Grammes of solution be added to 10 c.c. r-r acid, and 

 ° 10 



evaporated down to 50 grammes, and it is found that 1*53 c.c. 



N 

 -^-r baryta solution are required to neutralize 25 c.c. of the 



residue, we have 



10(-l)-2000^.q -01 X 1'53 

 60 b ~~~ 25 ~ ' 



where x is the normality of the solution, and a and b the 

 number of cubic centimetres occupied by one gramme of water 

 at temperature of filtration and at the ordinary temperature 

 respectively. 



The results thus obtained in normalities are transformed 

 in the tables given below into grammes per litre, and the 

 figures of Kohlrausch are appended for comparison. 



All experiments were performed at least in duplicate, in 

 some cases they were repeated many times. This was 



