In _dry assays_ the metal is almost always separated and weighed as
metal; in _wet_ gravimetric assays the metal is more usually weighed in
the form of a definite compound of known composition. The general
methods of working resemble those of ordinary chemical analysis, and
their successful working is greatly helped by a knowledge of, at any
rate, those compounds of the metal which enable it to be separated, and
of those which are the most convenient forms in which it can be weighed.
But the work of the assayer differs from that of the analyst, inasmuch
as the bulk of his estimations are made upon material of practically the
same kind, varying only in richness; consequently in assaying, it is
possible (and necessary) to work on such a definite plan as will involve
the least amount of labour in weighing and calculating.

The assayer connected with mining has generally two classes of material
to deal with: those comparatively rich and those comparatively poor. For
example, silver in bullion and in ores; copper precipitates or regulus,
and copper ores and slags; and "black tin" and tin ores. He is only
occasionally called on to assay the intermediate products. It is
indispensable that he should have an approximate knowledge of the
substance to be determined. With new ores this information is best got
by a qualitative testing. Knowing that only certain bodies are present,
it is evident that the number of separations can be reduced, and that
simple methods can be devised for arriving at the results sought for.
The best method is that which involves the least number of separations.
The reactions must be sharp and complete, and yet not be liable to error
under varying conditions.

To bring the richer and poorer materials under the same conditions for
the assay, a small weight, say 1 gram of the richer, and a larger weight
(5 or 10 grams) of the poorer, substance is weighed up. A method is then