loride of sodium; the second--which is,
perhaps, never wanting in eruptive cones that are often found lined with
it inside--is seldom generated in the fumaroles of the lava, and
therefore it is not easy to define the moment of its appearance.
Sometimes one collects micaceous peroxide of iron on the lava, but it is
often transported there from the mouths of eruption, as happened on this
occasion.

Trustworthy writers are of opinion that all the oxides are derived from
the decomposition of the chlorides, but I think I have clearly
demonstrated that, with regard to copper and lead, the opposite
statement may be affirmed; for the oxides are changed into chlorides,
and hydrochloric acid liberated. Oxide of copper forms sublimates at the
beginning, at the same time as the sea-salt; and if the fumarole be
anhydrous or, as Deville would say, _dry_, this oxide does not change
into either a chloride or a sulphate; but if the fumarole gives watery
vapour, after a little hydrochloric acid is formed, which changes the
oxide into a chloride, and if whilst this is going on oxide of lead be
developed, it is changed into the chloride of lead, so frequently found
in combination with chloride of copper. Then the sublimations change
from white to red or yellow, and specimens when carried away gradually
turn light blue, but when heated on platinum over a spirit lamp they
resume their yellow tint. Sometimes the yellow colour remains longer,
and in time changes to green; this also happens on the fumarole itself,
the green commencing at the zones furthest removed from the centre,
where the temperature is highest. When these sublimations are greenish,
they become far less soluble than at first. The yellow, so common at a
certain period on the fumaroles of the tranquil lavas, never attracted
attention before I first examined it, doubtless, because it was
considered chloride of iron, and yet in small eruptions this is only
found close to the discharging mouths, and never in the sublimations of
the fumaroles of the lava; but, on the other hand, it is the most
copious and common product on the lavas of the great eruptions. This
probably also accounts for the fact that lead, which is so obvious in
the fumaroles of the lavas, had never previously been observed. In 1855,
I noticed the crystallized chloride of lead in a fumarole in the Fossa
della Vetrana, and this induced me always to look for it on the
fumaroles of the later lavas; and I ascertained