may have come from them, but, if so, it
is singular that none appeared elsewhere. At 18 feet down was a mass
of travertine measuring nearly 3 feet across and from 6 to 12 inches
thick. It had formed around the lower part of a stalagmite 18 inches
long, and the bottom of the whole formation rested horizontally on
clay. This gave the excavators hope that an earth floor had been
reached, as the stalagmite was vertical and resembled in all respects
stalagmites in the cave. But it was soon found to be a foreign
inclusion, and the same confused mixture of clay and stone continued
below as above. Various fragments of stalactites and stalagmites were
found as part of the detritus. These, especially the vertical one,
seem to confirm the supposition that the roof reached out this far at
a period which is quite recent as compared with the age of the cave.

To a depth of 25 or 26 feet the task of excavating was as tedious and
difficult as digging up a much-traveled, rocky road, the earth being
dry enough to scour the shovels. Then the earth grew moist and within
2 feet was muddy. Cavities appeared, into some of which a switch could
be thrust 3 or 4 feet. Where such a cavity extended under a large
stone, stalactites were in process of formation. Soon the earth began
to work into a soft mud under the feet of the workmen, and at 32 feet
particles and small clods were noticed falling from the sides of the
shaft. A foot lower this breaking away became more decided. It may
have been due merely to the loose condition of the wet earth allowing
unsupported portions to fall from the freshly exposed surface, but
there was also the risk that the softer earth was sliding under the
weight of that above. The workmen, two of whom were experienced well
and cistern diggers, declared the risk too great and demanded to be
brought to the surface.

The depth reached by this shaft was at least 5 feet lower than at any
point inside, within 200 feet of the mouth of the cave. The material,
with the exception of the sand layer, was almost identical from top to
bottom, there being no apparent difference other than increase of
moisture in the lower part. The only explanation suggesting itself at
present is that the chasm is filled with large loose rocks up to a
point near the bottom of the shaft; that debris from the hillside
above has covered these more rapidly than it could settle in the
crevices and cavities among them; and that water which makes its way