DOI Seite / Zitierlink:
https://doi.org/10.11588/diglit.2984#0038
DISCOVERY OF CURVATURE OF HORIZONTAL LINES OF THE PARTHENON
25
weight. The surface so sunk was defined by a circle, a little larger than the lower diameter of the
column ; and the roughened portion by a plan of the column, with the flutes carefully traced on the
pavement. The plan served afterwards for fixing the lower drum, on which also the flutes were
finished, but for a small height only.1 This drum was probably rubbed upon the surface of the
pavement prepared for it, until it accurately fitted, and was left with its flutes coinciding with the
plan traced, as above-mentioned. It was then ready for the reception of the second drum. But
between these two the bed is differently formed. Exactly in the middle of each frustum is a
square sinking about .37 foot wide at the surface (rather less below), and .23 deep (see the repre-
sentation in the previous woodcut), into which was firmly fitted a piece of wood, either flush with, or
sunk a little below the surface of the drum. A cylindrical hole about. 16 foot in diameter was then
bored in the middle of each of these small blocks, and the two drums were then connected by
a round wooden pin, accurately fitting the hollows in the two inserted blocks of wood. A sort of
centre was thus formed by means of which one drum could be made to rotate upon the other.2
Around this centre (taking the case of one of the lower drums where the diameter is about 6.2 feet) we
find that the surface of the drum is smooth for a diameter of about .81 foot, after which it is sunk away
about .01 in a ring of .42 wide, then sunk .005 in a ring 1.5 in width, leaving a ring of about .75 of smooth
work in the same plane as the surface round about the centre; so that the weight of the column was borne
on the circular portion .81 in diameter about the centre, and a ring of about .75 wide at the circumference.
It was no doubt found to be more easy to secure a perfectly fine bed joint by this method, than if the
column had been allowed to bear over its full surface.3 The object of the wooden centre could not have
been to sustain any part of the weight of the block for the purpose of turning it cylindrically as in a lathe,
for which its strength would have been inadequate, but merely to keep it centrally over the drum below,
while moved to and fro through a small arc.
The stones were turned round with the help of ears (w-ra, ancones),1 which were at first left on the
cylindrical surfaces of the drums. The drums were probably also hoisted by ropes fastened round these
projections. The greater part of the operation of forming the beds was no doubt done on the ground ; but
the wooden centres were left between the drums when they were fixed, partly to secure to them the exact
positions in which they fitted most closely, when prepared below; and partly also to enable the operation
of rubbing to be renewed should any imperfection be found in the beds at the time of fixing the drum.
And until reduced by shrinkage, they would serve to prevent any lateral motion arising from accidents
or natural causes. When the columns were thus completed, the ancones were worked off, and the shafts
were left of a cylindrical or, more correctly, of a conical shape, until the last completion of the building;
in the state described in the Erechtheum inscription, ;uc<ntify<na,, dpd/3SaTa, unsmoothed and unfluted!1
The beds of the stones composing other parts of the building are somewhat similarly formed,
excepting that no parts of them are sunk. The middle of the bed is roughened by lines cross-hatched
and dotted, with a rim of smooth work left round the edges.
These stones seem to have had a full bearing over the whole surface; but a smooth junction only
towards the edges; which seems also to have been produced by rubbing them together till the edges,
at first left a little above the rest of the bed, were worn down to the general surface.
The vertical joints6 are formed like those of the drums of the columns, excepting that the apfj.it, or
1 See Plate XLVI., where is given an elevation of a similar drum.
- One of these pins from the Propylcea is preserved in the Museum of
the Institute of British Architects. It resembles olive wood; it is, however,
stated to be of the juniper cedar (Leake's Morea, ii. 8).
'■' These sinkings would have had the advantage of allowing the dust
produced in the rubbing the two blocks together, to escape readily into the
hollow parts.
4 Instances of these are shown, Plate XLVI. and vignette at the head
of Chap. I.
5 Besides the presence of the wooden centres, the measurements of the
scamilli impares given in Plates VII. and VIII. afford direct proof of the
stones having been rubbed one upon the other. Had the surfaces of the
beds been smoothed by the help of a common surface, they could not have
fitted in the perfect manner in which we find them, unless they were all
absolute planes. But among the dimensions given of the upper drums
(which as they could be measured with more certainty than the lower
are preferable for the present inquiry) it may be seen that in some cases
a slight difference exists between the sum of the measurements taken on
the north and south, and those taken east and west. The difference in
the second column from the northern angle is considerable ; but to avoid the
chance of this difference (.01) being thought a clerical error, I would refer
to the third from the south-east, which gives .004. This, though small,
is an appreciable quantity, and inconsistent with the two surfaces being
perfect planes.
0 Very great difficulty occurs in explaining how the vertical joints of the
architrave stones could be brought to the exquisite fineness which many of
them still retain. We may understand how the drums of the columns
would have been rubbed together by rotation, and how others of the
beds may have been formed by sliding one stone upon another: but how
the vertical joints of stones upwards of 14 feet in length and of con-
siderable depth could have been so perfectly formed is very wonderful ;
for they must have been too unwieldy to enable this object to be effected
by rubbing them together. Modern masons sometimes produce vertical
joints of considerable fineness by passing a thin saw blade between the two
contiguous surfaces, already carefully prepared, and afterwards wedging the
stones closely together. But some light has been thrown on this question,
H
25
weight. The surface so sunk was defined by a circle, a little larger than the lower diameter of the
column ; and the roughened portion by a plan of the column, with the flutes carefully traced on the
pavement. The plan served afterwards for fixing the lower drum, on which also the flutes were
finished, but for a small height only.1 This drum was probably rubbed upon the surface of the
pavement prepared for it, until it accurately fitted, and was left with its flutes coinciding with the
plan traced, as above-mentioned. It was then ready for the reception of the second drum. But
between these two the bed is differently formed. Exactly in the middle of each frustum is a
square sinking about .37 foot wide at the surface (rather less below), and .23 deep (see the repre-
sentation in the previous woodcut), into which was firmly fitted a piece of wood, either flush with, or
sunk a little below the surface of the drum. A cylindrical hole about. 16 foot in diameter was then
bored in the middle of each of these small blocks, and the two drums were then connected by
a round wooden pin, accurately fitting the hollows in the two inserted blocks of wood. A sort of
centre was thus formed by means of which one drum could be made to rotate upon the other.2
Around this centre (taking the case of one of the lower drums where the diameter is about 6.2 feet) we
find that the surface of the drum is smooth for a diameter of about .81 foot, after which it is sunk away
about .01 in a ring of .42 wide, then sunk .005 in a ring 1.5 in width, leaving a ring of about .75 of smooth
work in the same plane as the surface round about the centre; so that the weight of the column was borne
on the circular portion .81 in diameter about the centre, and a ring of about .75 wide at the circumference.
It was no doubt found to be more easy to secure a perfectly fine bed joint by this method, than if the
column had been allowed to bear over its full surface.3 The object of the wooden centre could not have
been to sustain any part of the weight of the block for the purpose of turning it cylindrically as in a lathe,
for which its strength would have been inadequate, but merely to keep it centrally over the drum below,
while moved to and fro through a small arc.
The stones were turned round with the help of ears (w-ra, ancones),1 which were at first left on the
cylindrical surfaces of the drums. The drums were probably also hoisted by ropes fastened round these
projections. The greater part of the operation of forming the beds was no doubt done on the ground ; but
the wooden centres were left between the drums when they were fixed, partly to secure to them the exact
positions in which they fitted most closely, when prepared below; and partly also to enable the operation
of rubbing to be renewed should any imperfection be found in the beds at the time of fixing the drum.
And until reduced by shrinkage, they would serve to prevent any lateral motion arising from accidents
or natural causes. When the columns were thus completed, the ancones were worked off, and the shafts
were left of a cylindrical or, more correctly, of a conical shape, until the last completion of the building;
in the state described in the Erechtheum inscription, ;uc<ntify<na,, dpd/3SaTa, unsmoothed and unfluted!1
The beds of the stones composing other parts of the building are somewhat similarly formed,
excepting that no parts of them are sunk. The middle of the bed is roughened by lines cross-hatched
and dotted, with a rim of smooth work left round the edges.
These stones seem to have had a full bearing over the whole surface; but a smooth junction only
towards the edges; which seems also to have been produced by rubbing them together till the edges,
at first left a little above the rest of the bed, were worn down to the general surface.
The vertical joints6 are formed like those of the drums of the columns, excepting that the apfj.it, or
1 See Plate XLVI., where is given an elevation of a similar drum.
- One of these pins from the Propylcea is preserved in the Museum of
the Institute of British Architects. It resembles olive wood; it is, however,
stated to be of the juniper cedar (Leake's Morea, ii. 8).
'■' These sinkings would have had the advantage of allowing the dust
produced in the rubbing the two blocks together, to escape readily into the
hollow parts.
4 Instances of these are shown, Plate XLVI. and vignette at the head
of Chap. I.
5 Besides the presence of the wooden centres, the measurements of the
scamilli impares given in Plates VII. and VIII. afford direct proof of the
stones having been rubbed one upon the other. Had the surfaces of the
beds been smoothed by the help of a common surface, they could not have
fitted in the perfect manner in which we find them, unless they were all
absolute planes. But among the dimensions given of the upper drums
(which as they could be measured with more certainty than the lower
are preferable for the present inquiry) it may be seen that in some cases
a slight difference exists between the sum of the measurements taken on
the north and south, and those taken east and west. The difference in
the second column from the northern angle is considerable ; but to avoid the
chance of this difference (.01) being thought a clerical error, I would refer
to the third from the south-east, which gives .004. This, though small,
is an appreciable quantity, and inconsistent with the two surfaces being
perfect planes.
0 Very great difficulty occurs in explaining how the vertical joints of the
architrave stones could be brought to the exquisite fineness which many of
them still retain. We may understand how the drums of the columns
would have been rubbed together by rotation, and how others of the
beds may have been formed by sliding one stone upon another: but how
the vertical joints of stones upwards of 14 feet in length and of con-
siderable depth could have been so perfectly formed is very wonderful ;
for they must have been too unwieldy to enable this object to be effected
by rubbing them together. Modern masons sometimes produce vertical
joints of considerable fineness by passing a thin saw blade between the two
contiguous surfaces, already carefully prepared, and afterwards wedging the
stones closely together. But some light has been thrown on this question,
H