had the fatal
result. First, the chemist tests with "group reagents." He knows that if
he puts into the glass containing the last brew certain bodies in
succession, some metals, if they are there, cannot be kept from rushing
into the arms of one, others will as passionately embrace another,
others still will unite with a third, while some will always repudiate
any alliance. There are in all cases signs of the union, when it takes
place, such as a blue or white or red colour, or a powder falling to the
bottom, or a fizzing of escaping gas.

In practice the analyst puts a little of the brew in a small glass
test-tube, pours in some distilled water, and carefully drops in some
hydrochloric acid. Now, if there is either silver, mercury, or lead, in
the brew, down goes a white powder; if none of these things is there, no
change follows.

Next he adds some sulphuretted hydrogen water, a sort of aerated water
smelling of rotten eggs. If tin, platinum, bismuth, cadmium, arsenic, or
one of several other metals, is in the brew, a coloured powder falls to
the bottom. Should nothing occur, he adds other things, until he has
tested for five groups of metals.

When he finds a poison belonging to a certain group, he has still to
ascertain which of five or six bodies it is.

For instance, after adding the first two test-liquors, if he sees a
yellow coloration or precipitate, he knows that he has either arsenic or
tin or cadmium. He then adds some strong ammonia, after boiling the
liquid till the smell of rotten eggs has disappeared. If the powder
dissolves, and the colour goes, he is quite sure he has found arsenic.

In this business-like way the murderer is convicted.

But now arises the necessity for making doubly sure, and another kind of
test altogether is employed. Life and death hanging on the result, the
test must be beyond all doubt. But arsenic is one of those
self-assertive things about whose presence there cannot be the most
infinitesimal doubt. Give a man a particle the size of a mustard-seed,
and let him swallow it. When he dies bury him, and let him lie under the
earth for a quarter of a century. Then gather the few remnants, give
them to a chemist, and he will return you a considerable portion of the
poison in the same state as that in which it was administered.

[Illustration: ARSENIC CRYSTALS.]

Probably the most famous special test for arsenic is Marsh's, the
invention of a Woolwich chemist, and equally famou