Polish Archaeology in the Mediterranean (13.2001(2002))

ALEXANDRIA

EGYPT

rather small in size. Their western ends had
either been destroyed by the portico
stylobate or were extended further west,
well beyond the confines of the present
archaeological trench. The function and
internal arrangement of the rooms is fairly
difficult to determine and the position of
potential doorways is hardly clear.
Much to our surprise, rooms 8-14,
which we had previously taken for the
western wing of the villa, are in fact
separated from the rest of the house by
a double wall standing some 2-2.40 m
high in places (Fig. 5). Although the upper
parts of the structure show clear evidence
of reworking and rebuilding, both walls
were initially constructed in a similar
isodomic technique, using large limestone
blocks some 50-53 cm wide and 40-50 cm
high. The course of the other wall,
although not preserved along the entire
length of the wing, leaves no doubt that it
had belonged to a separate, architecturally
independent structure extending to the
west, apparently yet another house.
Room no. 10 (4.10 m long), adjacent to
a small lavatory (no. 8), had a floor made of
assorted broken marble tiles. In the next,
small room no. 12 (c. 2.35 m wide),
separated only by a very thin screen wall
(c. 23 cm thick), the floor was made of
tamped earth and clay. In the western part
of the room, below the floor, a small
section of a sewage channel was cleared; it
was constructed of small stone blocks, and
ran towards a street located further to the
south. Floor subsidence in room no. 10
permits its course to be traced. A similar
tamped-earth floor was also encountered in
the neighboring room no. 13. In its
southeastern corner we came across a large

well. Its rectangular casing was made of
regular, vertically mounted slabs. The
bottom of the well was not reached,
explorations having to be halted until
necessary safety precautions could be
taken. The eastern part of the room
comprising the well was closed off with
a thin screen wall (c. 25 cm thick) built of
dressed stone (cf. Fig. 3). This is where
a large accumulation of broken vessels com-
prising mostly Palestinian (early form of
Gazan LR4) and Egyptian (Mareotic AE 3)
amphorae accompanied by some cooking
pots was found. All the datable sherds were
from the 3rd century. The finds were
carefully processed: as a result, several
complete forms could be restored. The
largest room, no. 14 (c. 3.35 m wide),
located further north, revealed the same
type of floor.
The fill featuring the western run of
rooms yielded large quantities of
architectural decoration fragments
featuring very good workmanship. These
pieces, which included limestone capitals
(Fig. 6), cornices, fluted colonnettes, etc.,
were found loose in the fill among
scattered demolition debris. Such evidence
of sumptuous interior decoration had been
recorded previously, during investigations
of the eastern wing.9) Large fragments of
tessellated mosaics were found wedged
among the broken ashlar and decorative
elements. Their stratigraphical position
indicated that they may have fallen from
rooms located on the second floor.
In all of the newly explored area,
a destruction layer varying in thickness
from 0.60 to 1.20 m was discovered
directly superimposed on the Early Roman
ruins. It consists mostly of large stone

9) G. Majcherek, PAM XI, op. cit., 7-38, fig. 6; for architectural decoration in general, cf. P. Pensabene, Elementi
architetonici di Alessandria e di altri siti egiziani (Rome 1993).

39