IV.J

ANCIENT CISTERNS.

39

me to one, under ground, which was plainly once a cis-
tern : its width is 13 feet 8 inches, the present height
to the spring of the arch 10 feet, and its length 36 feet.
In the arch I observe an aperture, as is usual in build-
ings of this kind. I also notice near the entrance an
earthen pipe, and, near the farther extremity, the mouth
of a small aqueduct which is eighteen inches wide and
almost as high. The walls are covered with a very
hard cement : where they have lost this covering, we
see the regular brickwork. I have no doubt, from the
appearance of the ground outside about this cistern,
that it formed one of several, which must have been
necessary to ensure a supply of water to so consider-
able a city, through the long drought of a Grecian
summer.

Pococke, in speaking of an arched building to the
north-east of the metdkhi, says: "it had some niches
which seem to have been designed for statues; it ap-
pears as a rough building, though probably it has been
cased.'" I quote him, since I wish to describe all that
exists, and the niches escaped my notice. He also
mentions the large cisterns "cased in fine brick" which
exist to the west of the metdkhi: but the finest and
most perfectly preserved cisterns that I saw are only a
few paces to the north-east of the house1. Their cement
is nearly every where preserved, but sufficient has fallen
off to shew that the walls are built of irregular small
stones, faced with regular brick-work on which the
plaster was laid. This was the commonest mode of
building cisterns: I have seen many such : some at
Priapos on the Hellespont, which, from all its remains,
would seem to have flourished for several ages under
the Roman Emperors.

Olivier searched in vain for marbles, inscriptions or
bas-reliefs at this Palaedkastron. I was equally unfor-

1 A sketch of one of the three arches, which form this reservoir, is given at
the head of Chapter V, below p. 61.