98 EXPLORATIONS IN PERSIAN MAKRÂN [Chap. III
have been abandoned a few generations ago. A fine perennial spring issues between the two forts. Older, perhaps, is a small circumvallation which I was shown on the top of the long stony ridge of Kurangi, about a mile to the east of Tumpi. This measures some 150 yards from north to south and about half that across. The much-decayed old walls of rough stone work had been repaired with still coarser masonry when a tribal rising occurred some eight years before my visit, and the old well within was then cleared. The coarse plain pottery found both within and outside affords no clue to an approximate dating; but it may be noted that traces of completely decayed stone structures are found over an extensive portion of the ridge to the south.
More interesting were the remains of the large Muhammadan burial-ground, known as Chihil-dukhtarân, the `forty daughters', a name frequently applied in Persia to ruined sites. Its remains occupy a terrace rising some 40 feet above the western bank of the Gungh Kaur and situated approximately opposite to the Sirdâr's fort. The area, covered with a multitude of graves and a number of small mounds marking completely collapsed sepulchral structures, extends for some 450 yards from north to south and more than 180 yards across. Many of the graves are marked by rows of carefully laid large pebbles, often also by low headstones. Three of the larger mounds, undoubtedly formed by the debris of sepulchral domes (gumbad), had been dug up. Within the foundation walls of a circular structure there was disclosed a vaulted tomb chamber, 13 feet 8 inches long and 10 feet 8 inches wide. The burnt bricks of the masonry measured 9 inches square with a thickness of 1 inches. At its northern end was a recess for a coffin, covered with large bricks set aslant. Here remains of a `wooden box containing earth' were said to have been found. Some 16 yards to the north, the wall foundation of another domed tomb could be traced, indicating a diameter of approximately 19 feet. A hole in the centre marked the place where a sepulchral chamber had been dug into.
At a third tomb there survived the lowest courses of massive walls from a rotunda having an approximate diameter of 25 feet. Here the vaulted tomb chamber, built with rough stones set in hard plaster, was laid in an east to west direction. This direction is certainly opposed to the universally observed Muslim practice of laying the body with the head to the north and the feet to the south, but apparently in agreement with Jewish custom. We subsequently observed at least three more graves at Chihil-dukhtarân and a few at a smaller burial-ground situated a short distance farther south across a ravine, the alinement of which was either from east to west or from north-east to south-west. It appears hence probable that these cemeteries were used also for the interment of non-Muhammadans living at Geh—Jews, and perhaps also Christians. The