Sec. i] FIRST VISIT TO MIRAN 347
Mirân. Its water has been utilized by the people of Abdal for a generation or two to create a small colony called the Mirân ` Târim ', where they cultivate fields of wheat without abandoning their fishermen's life by the river.3 As was to be expected among people but recently attracted to agricultural pursuits, neither the extent nor the method of cultivation seemed equal to the opportunities offered by the lively stream, then but lightly frozen over.
On December 8 I paid my first visit to the ruined site that lies to the east.. After crossing the stream and passing for about half a mile through a fertile belt, partly cultivated, the caravan track we were following emerged upon a scrub-covered steppe, and, a short distance beyond, upon
a bare waste covered with fine gravel. It seemed to stretch away eastwards absolutely level,
broken only by a succession of low and narrow ridges. The curious straightness of these and their roughly parallel direction running from south to north at once suggested that they marked the alignment of old canals. As soon as I reached the first, I came upon a completely ruined mound of solid brick masonry, the unmistakable remains of a Stûpa, and from its top I sighted groups of
other ruins (see plan, Plate 29) cropping up on the wide level flat eastwards like low islands on the surface of a lake. There could be no doubt that I stood on an important site.
In view of the detailed account of later explorations on this site which I shall have to give in Chapters XII, XIII, the record of the observations made on this rapid survey may be restricted to the narrowest compass. The first group of ruins reached after proceeding for some 600 yards further east (Fig. II I) proved to consist of four buildings, of no great size but of remarkably solid construction, and all unmistakably of great age. Of the two which were undoubtedly Stapes one, seen in Fig. I I I to the right in the middle, immediately attracted my attention by the unusual feature of a rotunda-like structure which must once have enclosed it. Another ruin, seen to its left, seemed to contain the remains of a square shrine solidly built with sun-dried bricks of unusual hardness. But the heavy masonry débris which filled and covered it made it impossible to identify it without excavation.
Without waiting to examine smaller remains which could be seen rising on wind-eroded ter-
races both to the south and north, I hurried on towards the old fort of which Tokhta Akhun had spoken as the principal ruin of the site. Seen from afar and over ground of such absolute flatness it looked quite imposing. But when I approached it after a further tramp of over a mile, and entered it by clambering over the badly breached west wall, I could not help feeling disappointed. The crumbling walls and bastions were massive enough and rose still high in parts ; but their inferior construction and the irregular shape of the fort (as seen in the subsequently prepared plan, Plate 3o) seemed to suggest a comparatively late date. There will be occasion further on to describe the details. Here it will suffice to state that the whole formed an irregular quadrangle with walls of about 240 feet in length on the two longer faces, which looked to the east and the north-west. Oblong towers of varying dimensions jutted out at the corners. Of the bastions guarding the centre of each face the one to the south (see Fig. I 13) was particularly striking. Projecting about thirty yards beyond the line of the wall, and still rising to a height of about forty-three feet, it looked like a keep or donjon.
The interior of this desolate stronghold was quite bare, without a trace of superstructures on the surface. But I could not doubt its age when I noticed that, within the circumvallation and near the west face, wind-erosion had scooped out a depression fully ten feet below what layers of stable refuse marked as the original ground level. The surface sloped down from the east wall, behind