186 PROCEEDINGS OP THE AMERICAN ACADEMY 



ride 0.0002 grm., in the mean, for 40 cm.^ of amyl alcohol, which is 

 approximately the quantity used in that case. It will be seen, there- 

 fore, that there exists for both salts an exact coincidence between the 

 mean total excess found and the difference between the figures which 

 indicate the solubility of the salts for the two methods of determina- 

 tion ; and, taking this fact in conjunction with the results of the 

 evaporation of amyl alcohol in blank, — the mean residue being 0.0004 

 grm. for 40 cm.^ and 0.0005 grm. for 50 cm.^, — it seems to be brought 

 out pretty clearly that the former set of figures represents more exactly 

 the solubility of the salts, though the difference between the two series 

 is not great. Resting, then, upon the former determinations, the solu- 

 bility of sodium chloi'ide may be taken as 0.0041 grm. in every 100 cm.^ 

 of anhydrous amyl alcohol, or one part in 30,000 parts by weiglit ; 

 and the solubility of potassium chloride, a little greater, is 0.0051 grm. 

 to 100 cm.'^ of amyl alcohol, or one part in 24,000 by weight. 



The conditions under which the salts are acted upon are such as 

 should insure the complete saturation of the solvent, and in this con- 

 nection it is interesting to note that for the quantities of material 

 employed the discrepancy between comparable figures never exceeds 

 0.0005 grm. 



In experiments (10), (11), and (14), (15), the alcohol was decanted 

 and filtered at once while hot; in (12), (13), and (16), (17), it was 

 cooled to 30° C. before decanting ; so it appears that the solubility of 

 the salts is not influenced by changes of temperature within the range 

 from 30° C. to 132° C. 



Used simply to wash the precipitate, amyl alcohol cannot, of course, 

 exert an effect at all comparable with that manifested in the experi- 

 ments which have been described, but to know just what this action 

 may be is important. Experiments (18) to (22) were undertaken, 

 therefore, to elucidate this point. 



Weighed amounts of the test solutions were evaporated nearly to 

 saturation in small glass beakers, amyl alcohol added, and, as in 

 the previous experiments, the whole heated until the salt had de- 

 posited and the residual alcohol had boiled quietly for some minutes 

 at its ordinary boiling point, the liquid decanted, filtered under gentle 

 pressure by means of a weighed perforated crucible and felt of asbestos, 

 the filtrate measured, the residue dislodged with the aid of a rubbing- 

 rod and transferred to the crucible and washed with anhydrous amyl 

 alcohol, the washings being collected and measured. The crucible 

 and contents were dried over a free flame turned low so that the heat 

 should not reach the melting point of the chlorides. 



