presentative selection of the
above stones, each with a sharp edge, is kept for the purpose of
scratching and being scratched, and those usually set apart for tests in
the various groups, are as follows:--

1 Talc
2 Rock-salt, or Gypsum
3 Calcite
4 Fluorspar
5 Apatite
6 Felspar
7 Quartz
8 Topaz
9 Corundum
10 Diamond

The stone under examination may perhaps first be somewhat roughly
classified by its colour, cleavage, and general shape. One of these
standard stones is then gently rubbed across its surface and then others
of increasingly higher degrees, till no scratch is evident under a
magnifying glass. Thus if quartz ceases to scratch it, but a topaz will
do so, the degree of hardness must lie between 7 and 8. Then we reverse
the process: the stone is passed over the standard, and if both quartz
and topaz are scratched, then the stone is at least equal in hardness to
the topaz, and its classification becomes an easy matter.

Instead of stones, some experts use variously-tempered needles of
different qualities and compositions of iron and steel. For instance, a
finely-tempered ordinary steel needle will cut up to No. 6 stones; one
made of tool steel, up to 7; one of manganese steel, to 7-1/2; one made
of high-speed tool steel, to 8 and 8-1/2, and so on, according to
temper; so that from the scratch which can be made with the finger-nail
on mica, to the hardness of the diamond, which diamond alone will
scratch readily, the stones may be picked out, classified and tested,
with unerring accuracy.

It will thus be seen how impossible it is, even in this one of many
tests, for an expert to be deceived in the purchase of precious stones,
except through gross carelessness--a fault seldom, if ever, met with in
the trade. For example--a piece of rock-crystal, chemically coloured,
and cut to represent a ruby, might appear so like one as to deceive a
novice, but the mere application to its surface of a real ruby, which is
hardness 9, or a No. 9 needle, would reveal too deep or powdery a
scratch; also its possibility of being scratched by a topaz or a No. 8
needle, would alone prove it false, for the corundum group, being harder
than No. 8, could not be scratched by it. So would the expert go down
the scale, the tiny scratches becoming fainter as he descended, because
he would be approaching more nearly the hardness of the stone under
test, till he arrived at the felspar, No. 6, wh