46 



ALCOHOL 



alcohol present; or, in other words, with the amount of water. And since tho 

 strength of spirits is determined by ascertaining their density, it becomes highly im- 

 portant to determine the precise ratio of this increase. This increase in density with 

 the amount of water, or diminution with the quantity of alcohol, is, however, not 

 directly proportional, in consequence of the contraction of volume which mixtures of 

 alcohol and water suffer, 



It therefore became necessary to determine the density of mixtures of known 

 composition, prepared artificially. This has boon done with great care by Mr, 

 Drinkwater ; * and the following Table by him is recommended as one of the most 

 accurate : 



Table of the Quantity of Alcohol, BT WEIGHT, contained in Mixtures of Alcohol and 

 Water of the following Specific Gravities: 



The preceding Table, though very accurate as far as it goes, is not sufficiently 

 extensive for practical purposes, only going, in fact, from 6 to 10 per cent, of alcohol ; 

 the Table of Tralle's (page 50) extends to 50 per cent, of absolute alcohol. 



Moreover, Drinkwater's Table has the (practical) disadvantage (though scientifically 

 more correct and useful) of stating the per-centage by weight ; whereas in Tralle's 

 Table it is given by volume. And since liquors are vended by measure, and not by 

 weight, the centesimal amount by volume is usually preferred. But as the bulk of 

 liquids generally, and particularly that of alcohol, is increased by heat, it is necessary 

 that the statement of the density in a certain volume should have reference to some 

 normal temperature. In tho construction of Tralle's Table tho temperature of thg 



Memoirs of the Chemical Society, vol. iil. p. 454. 



