NEW MODE OF CHEMICAL ANALYSIS. 
41 
This process, of which the results appear to me to be very 
exact, seems susceptible of very numerous applications in 
analytic chemistry. I would point out particularly the fol- 
lowing: 
1. In treating by the method pointed out, a known weight 
of a metal, slightly, or not at all acted upon by hydrochloric 
acid, we ascertain at once the composition of the chloride 
formed, or of the oxide remaining in solution in the excess of 
the hydrochloric acid. In certain cases, it is very difficult 
to determine this composition by the means hitherto em- 
ployed. In fact, it often happens that we cannot obtain the 
chloride in a constant state of composition, either from its 
volatility, or on account of a commencement of a decomposi- 
tion produced by the heat. It is the same of certain oxides 
which we cannot obtain perfectly pure. I would quote, for 
example, the perchlorides of gold and platinum, the compo- 
sition of which, I think, it will be very easy to verify by aid 
of the process pointed out. 
2. Likewise in certain cases where it is not possible to take 
the exact weight of the substance to be peroxidized, whilst 
the quantity of oxide produced may be ascertained with pre- 
cision. All the products of the oxidation of phosphorus, 
below phosphoric acid, may be analyzed in this manner, 
by ascertaining the quantity of phosphoric acid produced 
during the reaction, and the oxygen required to produce the 
change.* The body which serves to determine the quantity 
of oxygen, weighs fourteen and one-half times as much as the 
oxygen, which lessens very much the chances of error. 
3. We may determine the relative proportion of two oxides 
of iron, by mixing them with an excess of the peroxide of 
manganese, and treating the mixture by hydrochloric acid. 
* To analyze phosphatic acid, M. Dulong measured the quantity of 
chlorine absorbed by an indeterminate quantity of this acid during' its 
change to phosphoric acid, and then weighed the phosphoric acid formed. 
This process presents considerable analogy with that which I have 
described. 
vol. v. — NO. [. 6 
