Sec. iii] THROUGH RL DAN TO OLD HORMUZ 179
about 44 yards in diameter and 15 feet in height, could not be examined closely owing to the huts and refuse heaps which are packed on and around it. With the exception of a few glazed fragments found on its top, the pottery exposed on the surface consisted of coarse plain ware, undeterminate as to its period.
Passing thence about 6 miles to the south-east, across two wide rubble-filled beds of the river and a cultivated island between them, I proceeded next to a conspicuous ruin known as Qal`a-i-Kumiz from the name of a neighbouring hamlet. Built on a steep rocky hillock and surrounded at its foot by a much-decayed outer wall, some 280 yards square, the fort looked imposing. The structure within, measuring about 88 by 33 yards and built with small mud bricks, though much dilapidated, was manifestly of recent date. The pieces of glazed pottery and incised ware, with comb-drawn designs, which were picked up on the slopes below, confirmed this impression.
It was different with the indications gained at a site known as Tappa-i-Sultân Miri, which was shown to us amidst well-tilled fields about a mile to the west. There we found close together two small mounds which unmistakably date from a prehistoric settlement. They were both being terraced for cultivation and irrigated on the lower slopes by means of water raised from a canal with the help of leather buckets worked from poles. The trouble taken over this reclamation was doubtless prompted by the special fertility of the soil due to deposits of ancient refuse. The mound to the east, obviously much reduced in size by these operations, measured about 40 yards across its top, which rose some 15 feet above the present field-level. The other mound, also of the same height, quite close to the west appeared to have been detached from it by the levelling down of the intervening strip of ground now all irrigated. The size of this mound was at the time some 110 by 88 yards.
On the surface of both mounds there could be picked up plenty of pieces of well-levigated plain pottery, of decidedly ancient appearance, besides a considerable number of fragments of equally well-made painted ware, showing simple geometrical designs familiar from the chalcolithic pottery of Balûchistan sites. The body of these fragments varies between light red, cream, grey, and a pinkish-yellow. The specimens reproduced in Pl. XXV (S. Miri 15, 16, 18, 19, 26, 34, 39) will help to illustrate the painted patterns. It deserves to be specially noted that in several pieces, e.g. 26, broad bands of deep terra-cotta colour are combined with black ones for decoration. The appearance of this polychrome treatment might perhaps be taken as an indication of a somewhat later stage of prehistoric occupation. But note must be taken also of a few worked flints which were found here. The site if visited under conditions other than those imposed would certainly have invited trial excavation.