CLASSIFICATION OP MINERALS. 133
tional
groups must be selected from the peculiarity of their
natural-historical or chemical properties. Thus the zeolites are
easily seen to form such a natural group. The precious stones or gems
also, notwithstanding their diverse chemical composition, must ever
appear a highly natural family, when regarded as individual objects.
Their great hardness, tenacity, high specific gravity without the
me-.tallic aspect, their brilliant lustre, transparent purity, and
vivid colors,—all mark them out as a peculiar group. Only the diamond,
which might naturally seem to take the chief place in this class,
differs so much, not only in elementary composition, but in physical
properties, that it must be assigned to a different place.
Round
these .species thus selected, the other less important minerals are
arranged in groups or families. It is evident? that no precise
definition of these families can be given, as the connection is one of
resemblance in many points, not of identity in any single character. In
other words, it is a classification rather according to types than from
definitions, as every true natural classification must be. The same
cause, however, leaves the extent of the families somewhat undefined,
and also permits considerable license in the arrangement of species.
But both circumstances are rather of advantage in the present state of
the science, as allowing more freedom in the grouping of species than
could be obtained in a more rigid system of classification.
In
collecting the families into orders, the guidance ot chemistry is
followed rather than of natural history, though the latter is also
taken into consideration. Chemical names are assigned to the orders,
but still regarded as names derived from the prevailing chemical characters, and not as definitions. Hence it must not be considered an error should two or three mineral species be found in an order
