PREFACE. V
which their variety, and difference of charac-
ter, suggest; and have classified the temples
under several heads, according as they re-
quire to be distinct, or united, on account of
their differences, or resemblances in plan,
and general arrangement. The columns evi-
dently began with the square pillar, derived
from the mass left to support the roof of a
stone quarry ;* and this I therefore consider
the first order. That which is directly taken
from it, by removing the angles, and forming
it into a polygonal shaft, and thence into the
round fluted column, I consider the second
order; and the remaining orders (which are
also derived from the square pillar, with its
ornamental details), follow, according as
they gradually succeeded to, or were derived
from, each other: the third, fourth, and fifth,
having been invented before the sixth, and
seventh, orders.
It is with a view to follow up this grada-
tion, that I have put off the Osiride pillar to
the eighth order, though of earlier date than
the sixth and seventh ; as its introduction
immediately after the fourth would have in-
* Vide pages u, (i, 91.
which their variety, and difference of charac-
ter, suggest; and have classified the temples
under several heads, according as they re-
quire to be distinct, or united, on account of
their differences, or resemblances in plan,
and general arrangement. The columns evi-
dently began with the square pillar, derived
from the mass left to support the roof of a
stone quarry ;* and this I therefore consider
the first order. That which is directly taken
from it, by removing the angles, and forming
it into a polygonal shaft, and thence into the
round fluted column, I consider the second
order; and the remaining orders (which are
also derived from the square pillar, with its
ornamental details), follow, according as
they gradually succeeded to, or were derived
from, each other: the third, fourth, and fifth,
having been invented before the sixth, and
seventh, orders.
It is with a view to follow up this grada-
tion, that I have put off the Osiride pillar to
the eighth order, though of earlier date than
the sixth and seventh ; as its introduction
immediately after the fourth would have in-
* Vide pages u, (i, 91.