On the Purification of Water by means of Iron. G83 
acts in the same way as lime in the well-known Clark process ; 
the carbonates of lime and magnesia held in solution, in conse- 
quence of the presence of uncombined carbonic acid, are thrown 
<lown, and the water by that means is very materially softened. 
In domestic filters of ordinary construction, and indeed in all 
but spongy iron filters, the impurities are separated by mecha- 
nical action, and they become concentrated within the substance 
■or on the surface of the filtering medium, where germs of all 
kinds find a favourable breeding and feeding ground, so that, 
after a short time, a filter, unless cleaned, becomes actually 
injurious to the water passed through it, although apparently 
still rendering it bright and clear, because the extremely 
minute organisms which are injurious to health pass freely 
through the filter with the water. 
Animal charcoal and, to a smaller extent, vegetable charcoal, 
possess indeed the property of purifying water, not as mecha- 
nical filters only, but as media in the pores of which chemical 
reactions take place ; but these substances also soon become 
inert, and have to be periodically revivified by washing in 
acids and by heating. 
Spongy iron, on the other hand, from the nature of its action, 
•does not require periodical renewal, because its substance being 
slowly dissolved, fresh surfaces are constantly presented to the 
action of the water. When used in the form of a filter, indeed, 
the gradual accumulation of impurities requires to be removed, 
an order to allow the necessary freedom of percolation ; but the 
impurities which do collect among the iron are perfectly harm- 
less — there is neither animal nor vegetable life. 
The amount of iron dissolved is very small, depending on the 
•quality of the water. It is less than one-tenth of a grain per 
gallon, or one-seven-hundred-thousandth part by weight of the 
water operated on. 
The process of purification is of a twofold character. First, 
the water must be acted on by the iron, and afterwards the iron 
i;aken up must be separated from the water, and the free oxygen 
which had been absorbed must be restored to the latter. It is 
well also, when practicable, that the gross mechanical impuri- 
ties should be first separated by ordinary mechanical filtration. 
In domestic filters (Fig. 1, p. 684) Professor Bischof has 
arranged the whole process in a most ingenious and perfect 
manner in a single vessel. 
First, if the water to be filtered contains much suspended 
matter, comes a preliminary mechanical filtration, a layer of 
sand being placed on the top of the perforated plate, C, then the 
contact with spongy iron, r, and the final abstraction of the 
iron, and oxidation by means of a mineral called pyrolusite, S, 
