I li
H il
WUilll
AND PORTO SANTO. 81
empty) are in the same direction as the stream. So abundantly
ha ve the gases escaped in the parts near the surface, that the cavities
are considerable, probably from the cells left by the gases having
been so close and frequent as to communicate with each other, and
from their slight and imperfect partitions being subsequently
obliterated by decomposition. The sides of these ridges present
larger cells; continued projecting ledges, twists, and folds; and
look in every respect like igneous streams, suddenly petrified.
In the cavernous part (represented in the drawing) this scoriaceous
basalt, particularly open to the attack of destructive elements, from
its numerous cells, is in such an advanced state of decomposition,
that it looks like a dark brown indurated mud, in which more
obdurate masses are here and there imbedded. But the shallower
vaulted cavities, which we remark in the lowerpart of these ridges,
are probably owing to the heated stream of basalt having passed
over some patches of water left by the retiring tide, just before it
reached the sea, which, being immediately converted into vapour,
raised these vaults above the parts where they were developed and
liberated. This scoriaceous basalt, where it is not in an advanced
state of decomposition, is full as hard as the compact, but of a less
specific gravity q.
Bearing in mind, that the streams of scoriaceous basalt presenting
these characters, have flowed directly into the sea, which still
washes over them, and that the columnar basalt crowns the tops
of the cliffs only, (which are from fifty to one hundred feet above
the level of the sea) it is evident, that sudden congelation, instead
of producing such an arrangement of particles, as M. Faujas supposed,
entirely prevents itr. We shall presently find too, that the
i Compact basalt, 2.9. Scoriaceous basalt, 2.6: by Guyton’s glass Areometer,
and in distilled water. Red tufa, 1.75. Yellow tufa, 1.94.
r In the natural section close to the Pontinha on the east, the columns are more
symmetrical than to the west j and the basalt, where it has been broken away by the
workmen, presenting transverse cracks, and a hackly appearance. Plate 3, C.