I^W Dt Colquhoun on the Assail/ 



mineral in union with carbonic acid, have been dissolved, Biit 

 the clay ingredient is scarcely changed in any respect, and re- 

 mains mixed with the carbonaceous matter, in the form of a 

 black-coloured powder. Let this insoluble portion be sepa- 

 rated by a common paper filter, and then let it be thoroughly 

 washed by repeated affusions of water. * 



The liquid which passes through the filter contains all the 

 oxide of iron which existed in the ore in a state of combina^ 

 tion with carbonic acid ; it may contain also lime, magnesia, 



* As some of my readers may probably have had very little practice in 

 chemical manipulations, I trust that a few observations on the management 

 oijilters will Hot be deemed superfluous. After a precipitate has been 

 collected upon a filter and dried, it cannot be again completely separated, 

 because a certain quantity of it adheres so obstinately to the fibre of the 

 paper, that it cannot be removed without bringing away with it a portion 

 of the filter. H^nce it would be impossible iii this manner to determine 

 the exact amount of a precipitate. The expedient usually resorted to for 

 obviating this difficulty, is to weigh the filter beforehand : the increase of 

 weight which it acquires in the process is held to indicate the weight of 

 precipitate. But this method is liable to objection; for the weight of the 

 filter is apt to be altered by the liquids filtrated through it. Besides this, 

 unsized paper has a strong attraction for moisture, and the quantity of hy-^ 

 grometric water which it contains is constantly changing, according to the 

 degree of humidity of the atmosphere : it is necessary, therefore, to take 

 some peculiar precautions, in order to have the filter in exactly the same 

 state of dryness both at the commencement and at the termination of the 

 experiment. A much more accurate mode of filtering is to employ two fil- 

 ters. They ought to be in all particulars, such as thickness, fineness, size, 

 &c. as nearly ahke as possible ; and they must be carefully counterpoised 

 against one another on a balance, small portions being cut off from the 

 Jieavier of the two, until they both possess exactly the same weight. One 

 of these is then to be inserted within the other ; and the precipitate is to 

 be poured on this double filter, and washed in the tisual manner. It is ob- 

 vious, that whatever alteration one of these sustains in the course of the pro- 

 cess, will in all probability be sustained by the other; and also, that, at 

 .the termination of the desiccation, the two filters will be in precisely the 

 same state in r^ard to hygrometric humidity. The difference of weight 

 between the inner and the outer filter will tlierefore indicate very accu- 

 rately the weight of the precipitate. A very high degree of precision may 

 be attained by this mode of manipulating. 



When the precipitate requires to be heated to redness, its weight in the 

 filter ought to be ascertained in the first instance ; a determinate quantity 

 of it is then to be ignited, and from the loss of weight which this sustains, 

 it is easy to calculate the loss which would have been sustained had the 

 whole been ignited. # 



