this is only 0.67 m. below the level of the flooring of the opisthodomus. Taking this fragment of the flooring at the entrance to the opisthodomus as a level, it is found that the levelled rock at the southeastern corner of the peristyle is 2.07 m. below it; the rock on the northern side just west from the wall of the modern schoolhouse is 2.55 m. below it, and the lowest point of levelling is found in the foundation for the wall of the cella on the northern side at some distance from the eastern end. Here a slot has been made for an inlaid stone, and one stone has been set in the levelled rock; this point is 2.61 m. below the level of the floor of the opisthodomus, and the slot is 0.15 m. below this level, making a difference between the highest and lowest points of the natural rock of 2.09 m.

The plan itself (PLATE III) needs but little explanation; practically all the measurements are indicated, the levelled rock is left white, the blocks of masonry still in situ are indicated with a shading of diagonal lines, the unworked natural rock is shaded dark, the existing columns are black, and the location of the others is shown by dotted circles. The dotted lines on the present plan show the extent of Dr. Dörpfeld's excavations.

Dr. Dörpfeld discovered that the temple was hexastyle, with fifteen columns on a side (i.e. thirty-eight columns in all), a proportion which is seen at a later period in the temple of Apollo at Bassae and Temple R at Selinus. The stylobate of the temple was 53.30 m. in length and 21.36 m. in width. The foundations for the peristyle, the walls of the cella, and the interior columns were placed on the solid rock which was levelled and so descends in a series of steps or planes in accordance with the natural slope of the rock from west to east.1 These foundations consisted of walls of squared limestone masonry, the blocks of which had in general a surface measurement of c. 1.25 m. by 0.85 m. The breadth of the foundations for the peristyle was c. 2.50 m., but varies somewhat. The spaces between these foundation walls were filled with earth and chips of stone; the rock underneath was left in its original condition. The

1 Cf. photograph from northeast corner (PLATE II).

spacing of the columns from axis to axis on the west end, beginning at the corner, is as follows: 3.70 m., 4 m., 4.02 m., and 4 m.; on the side, 3.48 m. and 3.70 m. From this we may conclude that the spacing was 3.70 m. all down the sides excepting the last space at each end. Dr. Dörpfeld discovered that there was a slight curvature in the stylobate at the west end, for at the bases of the two middle columns the stylobate is two centimetres higher than under the corner column. I also tested this by a series of levellings, and obtained the same result.

The passageway between the peristyle and the wall of the cella was somewhat more than 2.60 m. in width. None of the masonry which formed the foundations for the temple proper remains, but the outline can be made out from the cuttings in the rock. The temple was a temple in antis, not prostyle; the position of the column of the opisthodomus which was standing in Stuart's time and of which he shows the position in his drawings makes this certain. This column stood behind the third column of the west end, counting from the southern corner, and was distant from it 4.29 m. (14 feet 11 inches). This measurement brings the column in position on the preserved masonry at the entrance to the opisthodomus, where it has been indicated on the plan. An examination of the surface of the masonry shows a faint circular weathering or tooling of the stone at this point and also at the place for the companion column on the other side of the entrance.

The cuttings for the bedding of the wall of the cella average about 1.50 m. in width. The chambers of the temple had a width of nearly nine metres. The entrance chamber on the

west has a much greater depth (4.23 m.) than the chamber at the east (2.71 m.). A large block of Pentelic marble found near the east end of the temple Dr. Dörpfeld thinks was the threshold of the eastern chamber. The interior between the two entrance chambers is divided by a cross-cutting into two chambers; that toward the east with a length of 16 m., that to the west of 9.60 m. This last is nearly square. Throughout two chambers run two parallel cuttings

the length of these

(c. 3.45 m. apart), where rested the foundations for the interior columns supporting the roof. Dr. Dörpfeld conjectures two pairs of columns in the western chamber and four pairs in the eastern. On account of this cross-cutting, Dr. Dörpfeld supposed the temple to have been a double temple with two separate chambers,1 where two different divinities were worshipped, but inasmuch as Pausanias mentions the structure as being sacred to Apollo only, it now seems more probable to him that the cross-cutting, which is much shallower and less sharply defined than any of the others, served as a bedding for some slight superstructure which was probably a screen placed across the cella of the temple to shut off the cult image from the larger chamber.

At a distance of 1.80 m. west of this cross-cutting were found the remains of what appears to have been the foundation for the base of the cult image. When Dr. Dörpfeld excavated here there were four blocks in place, but at present only two of these remain. The whole base had a breadth of 2.75 m. and a depth of about 2.25 m.

The columns of the temple are Doric in style with twenty flutings, which are c. 0.206 m. wide at the bottom. The columns are monolithic and the material is the rough native limestone of the neighborhood. The greatest diameter of the columns at the west end is 1.72 m., while those at the side measure only 1.63 m. The total height of column and capital is 7.215 m.; these measurements give massive proportions, i.e. 1:4.2 on the front and 1: 4.4 on the side. The proportions of the axial distance to the height of the column are 1: 1.8 on the front and 1:1.95 on the side. Dr. Dörpfeld was unable to measure the height of the columns, but Blouet gives the measurement as 7.21 m. I have measured the fallen monolith on the south side of the temple, whose full length is preserved, and

1 Professor J. B. Bury, in his History of Greece [1st ed.], p. 152, combines the idea that there were two separate chambers with the knowledge that Pausanias called this a temple of Apollo and guesses that the second chamber was dedicated to the worship of Artemis.

2 Cf. Ath. Mitth. XI, 1886, p. 304, no. 1.

found it to be 6.33 m. When the thickness of the capital and abacus, 0.885 m., is added to this, it gives a total of 7.215 m. Stuart's measurements, 21 feet 4 inches (6.50 m.), height of column to the swell of the capital, +2 feet 48 inches (0.72 m.), thickness of capital, give a total of 7.22 m. as the height.

The columns at the top have a diameter of only 1.295 m., and this great entasis and the flat, bulging capitals give an appearance of great solidity to the structure. The columns

were left slightly rough by the chisel in order to furnish a secure surface for the coat of reddish-yellow stucco with which the temple was finished in the earlier period. When the temple was repaired in Roman times a thicker coat of stucco. of coarser texture was put on over this. These different coats can be clearly seen on the larger fragment of column lying on the north side of the temple (Fig. 4).

The details of the capital and the annulets given in Figs. 5 and 6 are self-explanatory; these are redrawn from Dr. Dörp

feld's measurements. The capital proper has a thickness of 0.565 m. and the abacus of 0.32 m. The abacus of the third column from the southern

I measured them is 1.32 m., but Blouet gives the measurement as 1.292 m., while Stuart gives the total height in his time as

on the west to the outside of the corner column. On the third column of the western end the architraves are 0.27 m. from the outside edge of the abacus and 0.20 m. from the inside edge.