' of the Argillaceous Iron-Ore. 2!T5 



and oxide of manganese, when these ingredients happen to be 

 present in the ore in the form of carbonates. The solution 

 has always more or less of a greenish tint, in consequence of the 

 oxide of iron being chiefly in the state of black oxide. In the 

 farther progress of the analysis, the first step to be taken is to 

 convert the whole of this oxide into peroxide ; without which, 

 it would not be possible to make an accurate determination 

 of its quantity, because the black oxide of iron cannot be col- 

 lected in the form of a precipitate without its being partially 

 converted, in the course of the operation, into the red oxide. 

 This peroxidation may be conveniently effected by heating the 

 solution nearly to a temperature of ebullition, and then gradu- 

 ally adding nitric acid, in small quantities, until the deep 

 brown colour produced by the first portions of acid has disap» 

 peared, and is replaced by a clear reddish yellow. This \stiA 

 proof that the metal is now entirely peroxidated. 



Let this solution be diluted with water, and mixed with li- 

 quid caustic ammonia as long as a red-coloured precipitate 

 continues to fall ; and in doing this, some care should be taken 

 to avoid adding an excess of ammonia. The precipitate thus 

 produced consists solely of the peroxide of iron ; for caustic 

 ammonia does not precipitate lime from its solutions under any 

 circumstances, nor does it readily precipitate magnesia or oxide 

 of manganese, when their solutions contain a considerable ex- 

 cess of acid, even although it be added until the mixture ac- 

 quires an alkaline reaction. The precipitate ought to be col- 

 lected upon a filter, and thoroughly edulcorated; then h 

 should be carefully dried, first in the open air between folds 

 of common blotting paper, and afterwards in a moderate tem- 

 perature, so as not to char the filter, between 200° and 250° ; 

 in fine, to complete the expulsion of the whole moisture, let 

 the precipitate be exposed to a low red heat in a platinum cru- 

 cible for about fifteen minutes. 1 00 parts of it, when thus freed 

 from its water by ignition, represent 90 parts of protoxide of 

 iron, and 145 parts of carbonate of protoxide of iron.* This 



* The ammonia is apt to attract carbonic acid from the atmosphere, 

 and thus to acquire the property of decomposing solutions of hme. In 

 consequence of this, it may be proper in all those cases in which it is of 

 importance to determine the precise quantity of oxide of iron with extreme 



