198 



THE EAKL OF BEEKELEY ON SOME 



an error of 0'16 per cent. This large difference in the two percentage errors is 

 remarkable, and I have not yet been able to account for it it is manifested in most 

 of the salts hitherto worked with. A fact which may possibly throw light on the 

 subject is noticeable in the above table, and is one which most of the salts also show, 



namely, that the solubilities obtained when 

 starting with an unsaturated solution, tend to 

 be slightly less than those obtained when 

 starting with a supersaturated one, and this 

 although the corresponding densities are prac- 

 tically identical. I hope to investigate the 

 matter while determining the electric conduc- 

 tivities of these solutions. 



Tap for closing 

 pyknometer. 



Thermometer. 



Platinum wire for_ 

 pulling off the filter. 



I.R-Stoppex FU 



Solution.- 

 Filter. 



Sa.lt.- 



Pyknometer. 



Side tube 

 graduated 

 incjn. 



Fig. 5. 



w 



The Densities and Solubilities at the 

 Boiling-point. 



Attempts were made to determine these in 

 a Beckmann apparatus, but without success 

 the difficulty of keeping a constant temperature 

 being too great so recourse was had to a method 

 first suggested by, I believe, BUCHANAN. 



In the apparatus shown in fig. 5, the outer 

 glass tube A contains water, and the inner 

 tube B the salt and solution ; by boiling the 

 water vigorously and closing the side tube C, 

 steam, passing through the tube D, is forced 

 to bubble rapidly through the solution (D is 

 graduated in centimetres so that the level of 

 the solution may be estimated while the 

 pyknometer is in the solution). The steam, if 

 passed rapidly enough through the solution, 

 stirs it thoroughly, and the temperature rises 

 up to the boiling-point of the saturated 

 solution and remains constant at this point as 

 long as there is enough undissolved salt left. 

 The constancy of the temperature therefore 

 indicates that saturation is attained. 



Determination of the Density. 



When it is seen that the steam is passing freely through the solution, an india- 

 rubber plug, through which the thermometer and pyknometer pass, and which is also 



