the acid period, and first hydrochloric acid is produced, which, in
small lava streams, never conveys chloride of iron, and rarely attacks
the scoriae to form that salt, but expends its force in changing the
sublimations already there. For this reason chloride of iron, though
completely absent in the lavas of 1871, was abundantly found in those of
the 26th April, 1872. Sulphurous acid follows hydrochloric at a later
period, and sulphuretted hydrogen occasionally succeeds.

Having examined the gases of fumaroles by means of a graduated tube, and
the pyrogallate of potash, I always found that it contained less oxygen
than the surrounding atmosphere.

For several years I wished to see whether the fumaroles of the lavas had
a period of evolution of carbonic acid, as sometimes happens with
fumaroles near the craters, but I have always obtained negative
results. I often found that the atmosphere on the lavas contained an
excess of carbonic acid, but as these lavas had burnt many trees, and it
was probable that carbonic acid springs had formed under the lava, I
never considered it safe to form any conclusion on the subject.

IV.

BOMBS, LAPILLI AND ASHES.

The bombs ejected from the craters are like those carried down by the
lavas, but of smaller size, and they seldomer contain a nucleus similar
to those found in the latter. With the bombs properly so called, many
pieces of incandescent lava were thrown up, and in their fall went
beyond the base of the cone. A quantity of small scoriae varying in size
accompanied these projectiles, and those fragments, which we call
_lapilli_, fell at a greater distance. With the lapilli, and sometimes
without them, the smoke carried a very minute dust or sand, which is
generally called ashes. These ashes, when washed with water, lose
soluble constituents which they have collected in the smoke--such as
chloride of sodium and other chlorides and often free acids. The
insoluble part originates in the detritus of lava, and with the
microscope we can detect abundant fragments of those crystals which most
frequently occur in the lava of the same eruption.

The lavas of 1871, which were eminently leucitic, and almost entirely
deprived of pyroxene, resembled the ashes, which appeared to be
fragments of crystals of leucite, more or less enveloped in the paste of
the lava, so that having triturated the scoriae of the lava, and looked
at the powder through the microscope, it was apparently