598 Scientific Proceedings, Royal Dublin Society. 



As doubt exists with regard to the purity of the material employed by 

 some of the earlier workers, much attention has been given to the preparation 

 of a pure material for density determination. The choice of dehydrating 

 agents in removing the last traces of water was limited to substances which 

 did not cause any condensation of the acetone to more complex compounds. 

 Eecovered acetone obtained from the manufacture of cordite, and containing 

 only small traces of aldehydic or ketonic compounds other than acetone, was 

 converted into the bisulphite compound, from which the more soluble im- 

 purities were removed by washing with small quantities of water. Decom- 

 position of this addition-compound by means of sodium carbonate solution 

 followed by a fractional distillation, gave acetone containing some water and a 

 small amount of carbon dioxide in solution. The acidity was removed by 

 shaking with dilute aqueous potassium hydroxide. The acetone obtained 

 after another fractional distillation was allowed to stand over anhydrous 

 calcium chloride for several days and then re-distilled, the fraction 55-57° C. 

 being collected. This portion was again dehydrated with anhydrous calcium 

 chloride and fractionated, using an eight-section Young "evaporator" still- 

 head, and collecting between 56-56-5° C. This operation was repeated three 

 times, in the last fractionation the portion boiling at 56 - 3° C/760 mm. being 



20° 

 collected. The pure acetone had a specific gravity of 0-79091 which, cor- 

 rected for buoyancy, was - 79123. Submission to a further purification 

 process did not alter this boiling point or the density. The acetone was 

 tested and shown to be free from all traces of carbon dioxide, volatile or fixed 

 acids, and alkalinity. There were no compounds of an alcoholic nature present. 

 The absence of readily oxidisable impurity was shown by the addition of a 

 small quantity of dilute potassium permanganate solution when the charac- 

 teristic red colour obtained persisted for several days. 



Pure n-butyl alcohol was obtained from the crude fermentation alcohol by 

 dehydration over anhydrous potassium carbonate followed by repeated frac- 

 tionation. The fraction boiling between 117-118° C. was heated under reflux 

 for several hours with pure freshly-burnt calcium oxide. The dried alcohol 

 was distilled, the fraction boiling at 117'6° C. being collected. The pure 

 n-butyl alcohol boiled at 117'6°C/763 mm., and had a specific gravity of 



90° 

 0-80953 —j3 which, when corrected to vacuum, was 0-80974. 



Determination of densities of mixtures. 



The densities of the pure liquids and of the mixtures were obtained by 

 means of Perkin's modification of a Sprengel tube in which a bulb had been 

 blown above the mark to permit of expansion of tbe liquid (Trans. Cheni. Soc, 



