458 RETURN TO THE MÏRAN SITE [Chap. XII
it a constant blast of fine gravel and coarse sand, made it difficult for any one to hold out long there, and even my hardy Loplik diggers had to be relieved in brief shifts. That erosion had not progressed much further and had not yet succeeded in cutting through the wall altogether, after breaching the parapet, might well have caused me surprise. But in the refuse-filled rooms which my excavations brought to light immediately behind the east wall-face I had the clearest explanation of its ability to withstand the wind's onslaught for over a thousand years.
Interior Even before, on my first visit to the ruined fort, as described in Chapter X, the force of
near west wind-erosion had been fully brought home me b ht h to m by the observation that within the west wall-face
wall. the ground had been scooped out to a depth which, as comparison with the lowest layer of refuse
behind the north face showed, lay fully ten feet below the original level of the enclosed area. It was obviously by the agency of the fine gravel and sand which the prevailing winds were driving over the east wall that this natural excavation had been effected, and it seemed safe to conclude that this western portion of the enclosure had not been occupied by dwellings, which might otherwise have protected the soil from erosion. The obvious explanation of this was that any quarters built along the west wall-face would have been exposed far more to the cutting east-north-east wind than those near the sheltering east wall. The eroding force of the wind here also accounted for the badly-decayed condition of the west wall between its central bastion and that guarding the northwest corner. The stumps of three massive posts, still rising in a row above the ground by the side of the broken rampart where it showed a complete gap, evidently marked the position of the gate leading into the fort. Everything else had been slowly ground down and blown away.
Evidence of A few general observations connected with the outer appearance of the ruined fort call for
earlier occu record here. In many places, but especially y in the big south bastion, I noticed that the stamped
pation. y p p S big p
clay of the rampart contained plentiful fragments of pottery which had become embedded at the time of construction. Most of these potsherds were of a well-levigated clay and of superior make. This was clear proof that the ground close by, from which the clay for building the walls and bastions must obviously have been obtained to save labour and transport, had been occupied during an earlier historical period. To this important fact I shall have repeated occasions to refer hereafter.
Ground Some quasi-chronological interest attached also to another observation. At no point outside
surface near the foot of the walls was the effect of wind-erosion noticeable except at the north-west angle, where fort.
the ground immediately adjoining the ruined corner bastion showed erosion to a maximum depth of six
feet. I noticed this increased scouring capacity of the wind at the north-west angle in some of the other ruins of Mirân, and it may be connected with the whirl produced by the wind sweeping round the end of a wall which extends approximately parallel to its own course, and seems, somehow, to act like a funnel. At first the absence of any indications of a general lowering of the surface level around struck me as a sign suggesting a somewhat late origin for the ruin. But on closer observation I found that I must also take into account the peculiar surface conditions resulting at this site from the thin coating of gravel with which preceding deflation has covered the alluvial soil. The absence of indications of a lowering of the surface level was all the more curious in view of the fact that less than a quarter of a mile to the south of the fort I found a group of small Yardang terraces rising above wind-eroded troughs fully i 7 feet deep in places. Fig. i 32 shows one of these terraces occupied by the ruin M. x.
Stones A minor puzzle, which, however, was quickly solved, was presented by the abundance of big
stored on stones which lay scattered along the foot of the walls, and particularly along that of the big south or walls. bastion. I did not anywhere find evidence that such stones (no doubt brought from the higher rubble-filled stream bed) had been used as building material in the walls. I was therefore led to