WATER. 2(J5 



cement about the joinings of the worm is also to be avoided, as 

 the oxide of lead is readily dissolved by distilled water. The 

 first portions of the distilled water should be rejected, as they 

 often contain ammonia, and the distillation should not be carried 

 to dryness. 



Water employed for economical purposes is generally sub- 

 mitted to a more simple process, that of filtration, by which 

 it is rendered clear and transparent by the removal of matter 

 mechanically suspended in it. Such foreign matter may often 

 be removed in a considerable degree by subsidence, on which 

 account it is desirable that the water should stand at rest for a 

 time, before being filtered. The filtration of liquids generally 

 is effected on the small scale, by allowing them to flow through 

 unsized or filter paper, and that of water, on the large scale, 

 by passing it through beds of sand. The sand preferred for 

 that purpose is not fine, but gravelly, and crushed cinders or 

 furnace clinkers may be substituted for it. Its function, as that 

 also of the paper in the chemist's filter, is to act as a support for 

 the fiper particles of mud or precipitate which are first deposited 

 on its surface, and form the bed that really filters the water. 

 When the mud accumulates so as to impede the action of the 

 sand filter, the surface of the sand is scraped, and an inch or two 

 of it removed. Upward filtration through a bed of sand is some- 

 times practised, but it has the disadvantage that the filter cannot 

 be cleaned in the manner just indicated. Filtering under high 

 pressure and with great rapidity has lately been practised in a 

 very compact apparatus, consisting of a box, not above three 

 feet square, filled with sand. This filter which becomes speedily 

 choked with the mud it detains, is cleansed by suddenly revers- 

 ing the direction in which the water is passing through the box, 

 which occasions a shock that has the effect of loosening the 

 sand, and allowing the water to bring away the mud. The 

 action of such a filter, lately erected at the H6 tel-Dieu of Paris, 

 has been favourably reported on by M. Arago*. 



Matter actually dissolved in water is not affected by filtration. 

 No repetition of the process would withdraw the salt from 

 sea-water and make it fresh. Hence the impregnation of peaty 

 matter, which river water generally contains, and to the great- 

 est extent in summer, when the water is concentrated by evapo- 



* An. cle Ch. et de PU. t. 65, p. 428. 



