National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

   Sec. vii]   THE GREAT MAGAZINE OF THE LIMES   713

render the height of the whole main structure still more impressive. Its walls, in spite of the decay

they had suffered, especially on the south face, still rose in places to about 25 feet, being

highest on the north side of the central hall. The small figures of men seen in the photograph at

different points in front of the building will help to convey some scale.

Either on account of erosion undercutting the base or of some other cause, considerable portions Purpose of

of the south walls of the halls had fallen. These masses of clay had covered up the face of the base building

puzzling.

in some places, while in others the drainage of occasional rainfall, escaping on this side, had scooped

out deep hollows in the floor and base. All this made it difficult to ascertain where the main

entrances of the three big halls and the stairs once giving access to them might have been, or to

make quite sure that there were no remains of any. I was struck, however, from the first by the

absence of large windows such as halls of this great size would have needed for their proper lighting.

Curiously enough, the parts of the walls still standing, both on the north and on the south, were

instead pierced at fairly regular intervals by triangular openings, about 3 feet high and about as

wide at the base, one row being on a level flush with the floor and another about 14-15 feet

above it. Five of them could be traced below in the north wall of both the central and the

western halls. Obviously they must have been intended not so much for lighting as for ventilation.

No remains could be traced of internal fittings or arrangements which might have suggested the

purpose of the whole building. A plastered platform or plinth, about a foot wide and 8 inches

high, which on clearing the interior was found to run along the foot of the north wall in the western

and central halls, except in front of the triangular openings, did not furnish any clue.

Nor was direct guidance as to the purpose of the main building to be derived from any Inner en-

structural remains that could be traced around it. On the north side, and less clearly on the east, closure with

there survived portions of what appeared to have been an enclosing wall of stamped clay, about towers.

5 feet thick and built parallel to the walls of the main structure at about 4o feet from its base.

Fig. 182, on the right, shows the least injured segment of this enclosure on the north side. To the

west, below the steeply cut face of the clay ridge previously referred to, there was no clear indication

of such an inner enclosing wall, and along the south face I felt sure from the look of the level fore-

ground that it had never existed. Yet the intention of an inner enclosure or court for the big

building was plainly indicated by towers of stamped clay, massive but apparently not uniform in

plan, of which the remains rose at the four corners, marking a rectangle about 56o feet

long outside and nearly 200 feet across. The south-west tower, seen on the left of Fig. 186,

still stood to a height of about 20 feet. The north-west one, less well preserved, occupied a small

detached clay terrace, and is seen also in Fig. 182, on the left. The plan (Plate 41) shows

that the three towers which adjoined still traceable walls were built within the inner enclosure.

This fact clearly suggested that they were meant, not for defence against attack from outside, but as

points of vantage for sentinels posted to guard the building—or, rather, its contents.

Apart from this inner enclosure there survived also remains, far less distinct, of an outer one in Remains of

the shape of two badly decayed mounds running parallel to the south and north faces of the building. enclosure.

The one to the north ran about 8o feet outside the inner enclosing wall. The mound or

embankment on the south kept at a distance of over loo yards from the base of the building,

thus leaving a big open court in front of it. Owing to the subsoil being moist through the vicinity

of the marshes, the material composing the mounds had decayed into mere earth, and their original

construction could not be determined. Near what seemed the centre of the south mound there was

a gap about 4o feet wide which I thought might mark a gateway. Wind-erosion may account for

the disappearance of corresponding remains of an outer enclosure on the east and west sides, where

I could not trace any mounds.

   1374   4Y