248 ON THE PURIFICATION OF DRINKING WATER. 
tion goes on ; and most houses are supplied with tanks, cis- 
terns, or water-butts, which may be regarded as deposit 
reservoirs on a small scale.* 
The purification of water by subsidation and decantation 
is the simplest of all modes of depuration ; but unfortunately 
it is a very slow one. M. Leupold states that the water of 
the Garonne, taken when the river is swollen, does not 
recover its natural limpidity by ten days of perfect repose. 
The coarser impurities very quickly subside ; but the finer 
matters are deposited very slowly. 
During the time that deposition is going on, the water is 
exposed to the atmosphere, and in consequence suffers some 
chemical change. The bi-carbonate of lime, which it holds 
in solution, undergoes decomposition, half of its carbonic 
acid is evolved, and chalk or simple carbonate of lime de- 
posited. 
Ca 0, 2 C0 2 = C0 2 + Ca 0,C0 2 
Bi-carbonate Carbonic Chalk, 
of lime. acid. 
In this way the atmosphere aids in softening those waters 
which owe the whole or part of their hardness to bi-carbonate 
of lime. The atmosphere, however, is a source of contami- 
nation, as well as of purification. This must be very appa- 
rent when we take into consideration the immense quantity 
and variety of foreign bodies, inorganic as well as organic, 
contained in it. Ehrenberg tells us that, exclusive of in- 
organic substances, he has detected no less than 320 species 
of organic forms (Polygastrica, Pytolitharia, Polythalamia, 
and soft vegetable parts) in the dust of the winds ; and it is 
obvious, therefore, that waters which are chemically very 
pure, would become contaminated by a prolonged retention 
in the deposit reservoirs. 
* Sir William Clay says, that there cannot be less than from 
30,000 to 40,000 cisterns in the district supplied by the Grand Junc- 
tion Water Company. 
