dust thus be avoided. It is obvious that evaporation is promoted by
the use of vessels which admit of the exposure of a broad surface to
the air.

Liquids which contain suspended matter (precipitates) should always
be cautiously heated, since the presence of the solid matter is
frequently the occasion of violent "bumping," with consequent risk to
apparatus and analysis.

PART II

VOLUMETRIC ANALYSIS

The processes of volumetric analysis are, in general, simpler than
those of gravimetric analysis and accordingly serve best as an
introduction to the practice of quantitative analysis. For their
execution there are required, first, an accurate balance with which
to weigh the material for analysis; second, graduated instruments in
which to measure the volume of the solutions employed; third, standard
solutions, that is, solutions the value of which is accurately known;
and fourth, indicators, which will furnish accurate evidence of the
point at which the desired reaction is completed. The nature of the
indicators employed will be explained in connection with the different
analyses.

The process whereby a !standard solution! is brought into reaction is
called !titration!, and the point at which the reaction is exactly
completed is called the !end-point!. The !indicator! should show the
!end-point! of the !titration!. The volume of the standard solution
used then furnishes the measure of the substance to be determined as
truly as if that substance had been separated and weighed.

The processes of volumetric analysis are easily classified, according
to their character, into:

I. NEUTRALIZATION METHODS; such, for example, as those of acidimetry
and alkalimetry.

II. OXIDATION PROCESSES; as exemplified in the determination of
ferrous iron by its oxidation with potassium bichromate.

III. PRECIPITATION METHODS; of which the titration for silver with
potassium thiocyanate solution is an illustration.

From a somewhat different standpoint the methods in each case may
be subdivided into (a) DIRECT METHODS, in which the substance to be
measured is directly determined by titration to an end-point with a
standard solution; and (b) INDIRECT METHODS, in which the substance
itself is not measured, but a quantity of reagent is added which is
known to be an excess with respect to a specific reaction, and the
unused excess determined by titration. Examples of the latter class
will be pointed out as they occur in the pr