Water. ( 257 



from oils, from animal or vegetable matters, is 

 still the same when rendered pure by chemical 

 means. It is undoubtedly of consequence to 

 consider by what means water may be altered 

 from its purity ; but it is not difficult to obtain 

 it in that state in which the most rigid experi- 

 ment will find 110 admixture. Distil two-thirds 

 of inodorous, insipid, clear stream, rain, or snow 

 water, in glass vessels, to one-fourth, and the 

 water will be perfectly pure. This operation is 

 always necessary when water is to be used in 

 accurate experiments, for no natural water is 

 perfectly pure. Rain water is, indeed, much of 

 the same purity, when collected in clean vessels 

 at a distance from towns, and answers equally 

 well in experiments with distilled water ; it is, in 

 fact, water distilled by nature. Those springs 

 which flow from high and rocky grounds, where 

 we can discover no mineral or metallic sub- 

 stance, are the next in purity to rain water; 

 they are rain water collected on the hills, and 

 filtered through ground, in which there are no 

 soluble materials; but this is seldom the case, 

 for most soils contain lime, &c , but in so small 

 a quantity, that these spring waters are often a*s 

 useful as the purest. The water of rivers and 

 lakes is never so pure; indeed, it has often a 

 sensible taste and smell. 



But, besides the waters of rivers, which usually 

 derive their impurities from a mixture of animal 

 and vegetable substances, there are other waters, 



