DOI Page / Citation link:
https://doi.org/10.11588/diglit.4945#0262
The Croesus Structure. 251
northernmost of which measures 2*40 by 1*50. To this use of larger and
therefore better slabs under the base-plinths is due the fact that in all but
four cases the pavement which supported columns has been removed wholly
by later builders, while the intercolumnar slabs, which are, as a rule, smaller,
averaging about • 70 by • 50, have been left in position. The positions of the D
columns can be roughly distinguished by these gaps. No special foundations
were put in under the columns, the usual three layers of paving-blocks being
considered sufficient to support the weight. As will be seen later, however,
these have not always fulfilled the expectation. The pavement ran under
the walls of the cella, the slabs being usually laid lengthwise under the
line of a wall; but even there the arrangement was not very symmetrical,
no effort being made to secure uniformity of dimension or parallelism
of sides. So, in square F. 8, occur slabs supporting a wall, which vary
from 2*50 to 2*30 in length and from -70 to '50 in breadth. The upper
surface of these slabs is level, but roughly worked, and a shallow sinking
•40 by '25 by -02 was seen in the centre of many of them. It appears
that all the pavement surface was left rough in the first place until the
superstructures had been bedded upon it. Then the exposed portions were
levelled and worked to a smooth face. In certain places the faint dividing
line between the rough and smooth indicates the position of a vanished
superstructure, as, e.g., the X. wall of the cella, or the lost half of the column
base-plinth in square G. 12.
The "through bonders" appeared within the cella; but with the exception
of a few small and isolated slabs on the N. side, no other cella pavement
remained in position. Probably the cella-ftooring was made of unusually fine
slabs which were too attractive to later builders to be left alone.
Three mortised incisions, as Wood observed, occur in the peristyle
pavement nearly on a line with the third columns on the N. and S. flanks
and between the N. antac (see p. 10 and Plan, Atlas I., squares D. 11 and
E. 11). They are T-shaped and about "o8 deep, the two on the N. having
the head of the T lying to E., that on the S. having it to W. These were
obviously sockets for metal uprights.
Slope and settlements.—It appears from the levels of surviving patches of
pavement that the architect made the whole floor to slope slightly from the cella
walls outwards to the edges of the platform and also to curve downwards at the
extreme angles. Both facts are certain ; but owing to the numerous settlements
of the platform, caused first by the weight of the Croesus Temple, and
afterwards by that of the superimposed platform and temple of Hellenistic
2 k 2
northernmost of which measures 2*40 by 1*50. To this use of larger and
therefore better slabs under the base-plinths is due the fact that in all but
four cases the pavement which supported columns has been removed wholly
by later builders, while the intercolumnar slabs, which are, as a rule, smaller,
averaging about • 70 by • 50, have been left in position. The positions of the D
columns can be roughly distinguished by these gaps. No special foundations
were put in under the columns, the usual three layers of paving-blocks being
considered sufficient to support the weight. As will be seen later, however,
these have not always fulfilled the expectation. The pavement ran under
the walls of the cella, the slabs being usually laid lengthwise under the
line of a wall; but even there the arrangement was not very symmetrical,
no effort being made to secure uniformity of dimension or parallelism
of sides. So, in square F. 8, occur slabs supporting a wall, which vary
from 2*50 to 2*30 in length and from -70 to '50 in breadth. The upper
surface of these slabs is level, but roughly worked, and a shallow sinking
•40 by '25 by -02 was seen in the centre of many of them. It appears
that all the pavement surface was left rough in the first place until the
superstructures had been bedded upon it. Then the exposed portions were
levelled and worked to a smooth face. In certain places the faint dividing
line between the rough and smooth indicates the position of a vanished
superstructure, as, e.g., the X. wall of the cella, or the lost half of the column
base-plinth in square G. 12.
The "through bonders" appeared within the cella; but with the exception
of a few small and isolated slabs on the N. side, no other cella pavement
remained in position. Probably the cella-ftooring was made of unusually fine
slabs which were too attractive to later builders to be left alone.
Three mortised incisions, as Wood observed, occur in the peristyle
pavement nearly on a line with the third columns on the N. and S. flanks
and between the N. antac (see p. 10 and Plan, Atlas I., squares D. 11 and
E. 11). They are T-shaped and about "o8 deep, the two on the N. having
the head of the T lying to E., that on the S. having it to W. These were
obviously sockets for metal uprights.
Slope and settlements.—It appears from the levels of surviving patches of
pavement that the architect made the whole floor to slope slightly from the cella
walls outwards to the edges of the platform and also to curve downwards at the
extreme angles. Both facts are certain ; but owing to the numerous settlements
of the platform, caused first by the weight of the Croesus Temple, and
afterwards by that of the superimposed platform and temple of Hellenistic
2 k 2