METHOD OF DETERMINING SOLUBILITY 



The air for the solution was first passed into a large bottle 

 having a layer of strong potash solution, then through four 

 bottles of the same sort of solution of concentration 1 to 2. It 

 then passed through the meter and through four gas washing 

 bottles with water which were wholly immersed in the bath. 

 The purpose of these last bottles was to saturate the air with 

 water vapor at the temperature of the bath. The use of air for 

 stirring in such cases is not new 2 , but so far as could be ascer- 

 tained there has been no attempt to compensate for the unavoid- 

 able loss of solvent by evaporation. The air then passed 

 through the pipet and into the solution. The bottles, tube for 

 the solution, the pipet and all connections, save only the tip of 

 the capillary exit tube from the solution vessel, were wholly 

 immersed. 



Before undertaking actual determinations of solubility the 

 degree of compensation for evaporation was tested by many ex- 

 periments extending over periods from one to five hours, at tem- 

 peratures from 25° to 65°, the rate of the air current being 

 about 1.5 cubic foot per hour. This rate is sufficient for the 

 agitation of the solution. The tube containing the water and 

 pipet was weighed at the beginning and end of each experiment. 

 In some experiments there was a small loss, in others a slight 

 gain in weight, the difference never amounting to more than .035 

 gram per cubic foot of air. Since it was shown that the gain or 

 loss was closely proportional to the volume of air, and, there- 

 fore, to the time, it is evident that it could not exert any ap- 

 preciable influence on the results in such work. A few blank 

 tests have been made at 80°. At this temperature there was 

 an apparent loss of about 0.2 gram of solvent per cubic foot of 

 air. This loss is probably due to secondary causes that it may 

 not be possible to eliminate at such a high aqueous tension. Even 

 if confirmed by later experiment it seems hardly probable that 

 it could cause a degree of supersaturation that could materially 

 affect the results. However, there are other considerations that 

 make it seem probable that this method is best suited for mod- 

 erate temperatures, and that at or near the boiling point of the 

 solvent some one of the special methods for such temperatures 

 may more conveniently be used. 



In carrying out an actual determination the substance under 

 investigation was dissolved in about 60cc. of water in sufficient 



2 Pawlewski, Berichte d Deutsch Chem. Gessel., p. 1040. 

 3 



