Atkinson, Thomas [Mitarb.]
Excavations at Phylakopi in Melos — London, 1904

Seite: 10
Lizenz: Creative Commons - Namensnennung - Weitergabe unter gleichen Bedingungen Nutzung / Bestellung
1 cm


§ 2.—The South-Wcstern Corner.

As soon as the ioner face of the south-western angle of tlie Great Wall
was determined, it was seen that only a very narrow strip of the inner town
had survived the erosion of the sea at that point. The waves, driven on
shore by the prevailing north-westerly winds (of whose persistent violence
in this locality we had subseqnently all too much experience), have at this
point eaten out a deep bay, carrying away all the western wall of the city
except some ten metres of its southern end, and also all but that strip of
constrnctions, which, nestling under the southern wall, widens from five
metres at the west, to some twentjr metres farther east, as the cliff curves
definitely away to northward (Fig. 5).

In this horn-like remnaut of the south-west quarter of the city, Mr.
Mackenzie had made soundings in 1800 with the results already described.

Fro. 5.—Tue Eroded S.W. Corner of tue Town prom within.

After the outer face of the Fortification had been cleared in the following
year, Mr. Cecil Smith turned Iiis attention to such Chambers as had not
been completely cleared within.

The passage communicating from the town with the postern at the foot
of the intra-mural staircase (infra, p. 33, Figs. 16, 17) was first op'ened out
(D 5). It had ere this become clear that this stairway was a means of access
from within to the crown of the rainpavt, not an exit to the open country;
but the outward profile of the enclosing bastion was uever clearly determined
owing to its ruinons condition.1 The whole fortification makes a marked
outward sally at this point, and fiossibly rose into a low but massive tower.

When the space in the extreme south-west angle (A B 5) came to be
wholly dug out to rock, house walls of a more primitive epoch and better con-
struction than the unsquared structures, which appear nearest the surface

1 Mv. Mackenzie dissents here.

See p. 257 infra.
loading ...