cal patterns, squares and lozenges. I think that in that
particular case the peculiarity was due to the lava having flowed over
curved surfaces. In coming in contact with the atmosphere it had cooled
more rapidly on the upper face than the under, and in contracting quickly
had split at regular intervals, thus forming the geometrical pattern.

It was undoubted that we were there in the former centre of inconceivable
volcanic activity. In other parts of a great dome of rock I came upon
strange holes in the rock--extremely common all over that region--which
might at first glance be mistaken for depressions formed by glacial
action, but which were not. They were merely moulds of highly ferruginous
rock, granular on its surface and not smoothed, as one would expect in
the walls of cavities made by the friction of revolving ice and rock. Nor
did I ever find at the bottom of any of those pits, worn-down, smooth
spherical or spheroid rocks, such as are usually found in pits of glacial
formation. Those pits had been formed by lava and molten iron flowing
around easily crumbled blocks of rock, or perhaps by large balls of
erupted mud which had dropped on molten lava, that had then solidified
round them, while the mud or soft rock had subsequently been dissolved by
rain, leaving the mould intact. The latter theory would seem to me the
more plausible, as many of those pits showed much indented, raised edges,
as if splashing had taken place when the rock now forming the mould was
in semi-liquid form. Only once or twice did I notice hollows with a
suggestion of spiral grooves in their walls; but I think that those had
been caused at a more recent date by water flowing in and describing a
spiral as it travelled downward in the interior of the vessels.

On the hill where the observatory was situated two circular volcanic
vents were to be seen. The hill, which had a slope on one side, had
evidently been split, as on reaching the top I found that an almost
vertical precipice was on the other side. Quantities of quartz and
crystals were to be found on that hill. All over that region
quaintly-shaped rocks were also to be found, some like small cubic or
rectangular boxes, others not unlike inkstands, others in hollowed
cylinders or spheres. Many--and those were the quaintest of all--were of
a rectangular shape, which when split disclosed a rectangular hollow
inside. These natural boxes were mostly of iron rock, laminated, which
had evidentl