Sec. iii] SITE OF ZAHIDAN AND LATER RUINS TO THE NORTH-WEST 933
cultivable area. But within the Persian portion of the northern delta, to the west of the Helmand-iKalàn and its terminal continuation, the Siksar river, such areas as had at different periods been deserted owing to failure of irrigation or other causes, such as destructive invasions, have again and again been brought under cultivation when changes in the river restored irrigation facilities or human factors permitted the reclamation of fertile ground that had been abandoned. Thus extensive ruins are known to have disappeared there within living memory, buried under the heavy alluvium deposited by river branches escaping into new channels, or effaced by resumed cultivation.' These brief observations will suffice to explain why, apart from a few sites such as Shahristàn and Atish-kadah occupying little island-like outliers of the ` Dasht ' plateau, practically all remains surviving above ground within the limits above indicated belong to Muhammadan times.
By far the most extensive and conspicuous of such remains are those of the ruined town known as Zàhidàn, which local tradition, probably with good reason, considers to have been the Sistàn capital besieged and taken by Timûr (A. D. 1383). It is situated about 6 miles to the north-west of Shahristàn, on a low ridge of clay stretching to the north-west between the old river-bed known as R[id-i-Nàseru and the wide belt reached by the floods of the Rûd-i-Pariûn. Slight as the elevation of this ridge is, it precludes inundation from either side. At the same time changes in the level of the ground which existing canals from the Rûd-i-Sistàn can conveniently be made to command have prevented the cultivation of the greater part of this broad ridge since Zàhidàn was abandoned, probably not very long after Timûr's invasion of Sistàn. This accounts for the survival of extensive remains at the site and also for the freedom with which the wind has exerted its erosive action upon them wherever irrigation has not reached.
The ruins of Zàhidàn have been described at some length by Mr. Tate,2 and this, together with Walls of their comparatively late date, renders a detailed account here unnecessary. It will suffice to note citadel.
briefly the essential features of the site and to refer for their illustration to the sketch-plan, Pl. 56.3 The best-preserved ruin within the walled town is that of the citadel. It consists of an oblong
inner fort strengthened by massive towers or semicircular bastions and of two outer enclosures pro-
tecting its NE. and SE. faces, also provided with towers. These defences are constructed of sun-dried masonry resting on foundations of stamped clay. The bricks show a fairly uniform size
of 12" X 6" x 21". Single courses of hard-burnt bricks are to be found inserted at intervals in the
walls of the towers and also as a flooring between the stamped clay and the brickwork. Winderosion, which is very noticeable in the circumvallation outside and on the ground enclosed by it,
has affected the walls of the citadel far less. This seems to point to a continuance of the occupation
of the latter and to the repair of its defences for some time after the area outside had been abandoned.
Outside the citadel there are no structures of large size, but a group of detached buildings which Outer cir-
evidently served as quarters. Much drift-sand accumulated amidst and above them has helped
to protect them: I noted that their vaultings and arches all showed voussoirs of the usual Western cumvalla-
type, instead of the peculiar construction with bricks set on edge with their longer sides along the curve of the arch, as invariably observed at Ghàgha-shahr and also at Atish-kadah. The walls of the outer circumvallation form an irregular oblong, truncated at its southern end. Its length is a little over 12 miles and its maximum width about 4 mile. The enclosing walls, badly decayed for the most part, seemed, as far as I examined them, to consist of a rampart of stamped clay probably surmounted by a parapet of sun-dried brickwork. Semicircular bastions strengthened
1 Cf. Tate, Seistan, pp. 115, 177, 202, 236 sq.
2 See ibid., pp. 219 sqq.
3 The preparation of this plan proved necessary since the one inserted after section iii of Seistan shows inaccuracies,