masonry, with no particular characteristics. At E there is a cistern, 14 feet deep, which we have not yet completely cleared out. Many skulls and bones had been thrown in. Two of the skulls bore marks of hard blows; what was the battle that took place here, and who were the antagonists 1
I was anxious to trace the scarp of the tower, the north side of which was visible under the wall of the cemetery. It was not long before we found the west scarp at the foot of the great heap of rubbish referred to above. That I should find remains of the tower itself I did not hope, though it might have been inferred from the mass of debris. However, we had not worked far along the trench H—H' before we found a large drafted stone set back a few inches from the top of the scarp, which is chamfered. Another and another appeared joined together by strong mortar. A second course soon appeared, and when we reached the corner (a distance of 3912 feet from the cemetery wall) three courses were in place. The masonry was missing at the north end, the entire length of the lower course still in place to the south corner is 27 feet.
We are now opening up a trench along the line H'—I. The rock is visible at I, 11 feet above H'; the distance from H' to I is 32 feet, being the length of this side of the tower. We have not as yet reached the sloping rock all along the trench, but the masonry thus far uncovered appears to be of the same character as that below. It is, however, extensively but irregularly plastered over, and the plaster is Everywhere curiously indented with marks made by a small wedge-shaped tool whether for ornamentation or for securing the plaster I cannot say. Squeezes will be sent. I have seen nothing like it.
The drafted stones vary in size; the three largest seen measure—(1) 4 feet 11 inches long by 1 foot 4 inches high, the top draft is 2 inches wide, and the right side draft 3 inches; (2) 3 feet 6 inches long by 2 feet 3 inches high, to]) draft 3 inches, side draft 2 inches; (3) 2 feet 11 inches long by 2 feet 3 inches high. This latter has a rough centre projecting 9 inches from the draft. The projections of centres of the stone vary, some being almost flush with the draft. Thus far the depth of the stones has not been seen. The courses are regular in level, the lesser height of (1) is due to its being in the lowest course where the scarp is not level. The drafts bear no sign of the comb-pick, but of what Petrie calls the "pock-marking" dressing, which was used in Phœnician times. How much later it came down I am not prepared to say. Our stones here are similar to the masonry at the south-west angle of the city wall. Mr. Hanauer pointed out to me that the lowest stone of the corner projects considerably beyond the corner, as if old masonry in situ, which did not follow the exact lines of the new wall, had been taken advantage of. A shaft would determine the question.
On Conder's plan, parallel to the north side of the tower on which the school is built, is marked a "Modern wall of old masonry." These stones were found during the excavations, and are supposed to have fallen from above. Conder thinks them to be of Roman date. I have compared
