420 
LEEDS CHEMISTS’ ASSOCTAION. 
this element; and the end of the reaction is generally known by the production of some 
distinct colour. Thus, protosalts of iron combine readily with just half as much more 0 
as they already contain, and are thereby converted into persalts of that metal. Now we 
have tests that will distinguish the smallest quantities of either of these salts, and we 
can therefore tell at once when the protosalt is all converted into persalt. Again, per¬ 
manganate of potash, a salt whose solution in water is of an intense violet-red colour, 
yields up a certain quantity of 0 to substances capable of combining with it, and becomes 
thereby decolorized. If now a solution of KO Mn 2 0 7 be added to a solution, say of 
FeGS0 3 , so long as any of this remains the colour is discharged, but immediately it is all 
oxidized this beautiful colour shows itself and so marks the end of the operation. The 
permanganate is thus at once reagent and indicator. 
Protosalts of iron and oxalic acid are the chief substances thus readily decomposed by 
permanganate, and as we can readily convert any iron solution we may have into the 
state of a protosalt, the amount of oxygen or permanganate required to convert this into 
a persalt is an accurate index to the amount of iron contained in our solution. The uses 
of oxalic acid in connection with permanganate are mostly of what is called an indirect 
character; thus lime and lead may be estimated by their means ; not that permanganate 
has any action on either of them, but they are both perfectly precipitated by oxalic 
acid. If, then, to a solution of lime or lead, we add oxalate of ammonia in excess, 
taking care to know exactly how much we have added, the amount of this excess, more 
than would combine with the lime, and which we can estimate by the permanganate, 
subtracted from the whole amount used, gives us the amount which has combined with the 
lime ; from this the quantity of lime is readily calculated. This may seem a com¬ 
plicated way of arriving at the result, but it possesses an element of great convenience, 
which is, that we need not wait to separate the precipitates entirely, but if we make our 
mixture of oxalates, etc. to a certain volume, the upper part will very soon have settled 
enough for us to take out one-fourth or one-fifth part clear, and multiplying the results 
obtained from this by four or five, as the case may be, we very readily get what we want 
to know. Bichromate of potash again is used as an oxidizing agent, in iron estimations 
more particularly. It will not, however, act as its own indicator as permanganate will, 
but the solution must be tested from time to time until the appropriate reagents show 
that there is no protosalt left. Iodine is also in many cases a powerful oxidizer, and the 
beautiful reaction between iodine and starch enabling us to detect the smallest traces 
of free iodine, many valuable processes are based on its use, both directly and indirectly. 
It is used directly in the estimation of sulphuretted hydrogen or sulphides in mineral 
waters or other substances, as these compounds, as also hyposulphites, are decomposed 
by it. If a solution of starch be mixed with any of these and standard I solution added, 
no colour will be produced till they are exactly decomposed, but when this is the case 
the intense blue is instantly developed. On account of this reaction between hyposul¬ 
phites and iodine, a standard solution of hyposulphite of soda is used for the estimation 
of free iodine and is very useful in indirect analysis by means of it. Indirect analyses 
are so called when we do not directly estimate the amount of the substance we require, 
but the amount of something else that that substance may under certain circumstances 
be made to produce. For instance, suppose it to be required to know the value, that is, 
the oxygen-yielding power, of a sample of MnQ 2 . We do not estimate this 0 directly, 
—we have no convenient means of doing so,—but if it be boiled with HC1 a certain 
equivalent of Cl is produced to the amount of 0 yielded by the sample. We have, how¬ 
ever, again, no ready means of estimating this Cl, but if the Cl be passed through a 
solution of KI an equivalent of I is set free for each equivalent of Cl used, and this 
iodine we can estimate with 'great nicety, as above noticed ; but the indirectness of the 
analysis hardly ends here, for it is most convenient to destroy the I thus produced by a 
known excess of standard hyposulphite solution and then estimate the excess of hyposul¬ 
phite again by standard iodine and starch. This may seem a roundabout and complicated 
method, but is not so in practice, and there is no class of volumetric analysis that yields 
such accurate results as these indirect ones with iodine. Besides Mn0 2 there are many 
substances that will when boiled with HC1 liberate chlorine, and which can therefore 
be estimated in this way. 
We come now to the last division, that of analysis by precipitation, the principle of 
which is, that the substance to be analysed shall form an insoluble precipitate with the 
reagent employed. The end of the operation may be determined in three ways : firstly, 
