had their origin in hot solutions
from below. The presence of mercury in a deposit then becomes suggestive
of hot-water conditions.

(7) Ores sometimes occur in inverted troughs indicating lodgment by
solutions from below, as, for instance, in the saddle-reef gold ores of
Nova Scotia and Australia, and in certain copper ores of the Jerome camp
of Arizona (p. 204.) This occurrence does not indicate whether the
solutions were hot or cold, magmatic or meteoric, but in connection with
other evidences has sometimes been regarded as significant of a magmatic
source beneath.

Perhaps no one of these lines of evidence is conclusive; but together
they make a strong case for the conclusion that the solutions which
deposited the ores of this class were hot, came from deep sources, and
were probably primary solutions given off by cooling magmas.

The conclusion that some ores are derived from igneous sources, based on
evidence of the kind above outlined, does not mean necessarily that the
ore is derived from the immediately adjacent part of the cooling magma.
In fact the evidence is decisive, in perhaps the majority of cases, that
the source of the mineral solutions was somewhat below; that these
solutions may have originated in the same melting-pot with the magma,
but that they came up independently and a little later,--perhaps along
the same channels, perhaps along others.

POSSIBLE INFLUENCE OF METEORIC WATERS IN DEPOSITION OF ORES OF THIS CLASS

It is hardly safe, with existing knowledge, to apply the above
conclusion to all ore deposits with igneous associations, or in any case
to eliminate entirely another agency,--namely, ground-waters of surface
or meteoric origin, which are now present and may be presumed often to
have been present in the rocks into which the ores were introduced. Such
waters may have been heated and started in vigorous circulation by the
introduction of igneous masses, and thereby may have been enabled to
effectively search out and segregate minutely disseminated ore particles
from wide areas. This has been suggested as a probability for the
Kennecott copper ores of Alaska (p. 200) and for the copper ores of Ely,
Nevada. In the Goldfield camp (p. 230) the ores are closely associated
with alunite in such a manner as to suggest a common origin. It has been
found difficult to explain the presence of the alunite except through
the agency of surface oxidizing waters acting on hydrogen sulphide
coming f