Sec. ii] THE NORTH-WESTERN GROUP OF RUINS 217
instance of the covering-tablet of a double-wedge document in wood having subsequently been transformed into a detached seal case, such as I had found in the course of my former excavations.' Nor should it pass without notice that the rectangular double tablet N. xIII. ii. I shows on the reverse of its under-tablet the unusual feature of an endorsement of which the first line is in Brâhmi, and that, according to information kindly supplied by Professor Rapson, N. xiii. ii. 2 is an exact duplicate of this tablet, but wanting its endorsement.
The narrow apartment, iii, to the north-east also yielded half a dozen Kharosthi documents on Miscella-
wood, though their condition was not so perfect. To judge from the fragments of earthenware pots noun im-
and the miscellaneous household utensils found there, it had evidently served as a living-room. found in Among the latter articles, seen in the photograph Fig. 52, were a mouse-trap (N. )(III. iii. 001 ; N. xIII• Plate xIx) ; a boot-last (N. mu. iii. 002) ; a weaver's wooden comb ; a bucket and a pitchfork in wood, and two baskets (see Fig. 47). In the small room, iv, adjoining was found a flat wooden tray, measuring about two feet by one foot five inches, which is seen in the photographs ; this, on account of its raised edge decorated in chip-carving, my labourers took for an eating-tray--in all probability rightly. Finally, from the narrow room, v, there came the boldly carved wooden panel, N. XIII. v. I (see Plate xix), once belonging to some piece of furniture, and bearing in its decorative floral design unmistakable resemblance to the carved chair already referred to. The large round board in wood, seen in Fig. 47, probably served to close a sunk wooden trough, while the use of the long roller-like implement seen on the extreme left of Fig. 52 is uncertain.
After clearing this dwelling we retraced our steps in the afternoon to the remains of a far larger Clearing of structure, N. xiv, situated about a quarter of a mile to the south-west, and quite close to our camping-
ruin N. XIV.
place. Posts, bleached and splintered, but still rising up to nine feet in height, marked here the position of the walls of a hall, of which the ground-plan (see Plate 9) showed the imposing dimensions of fifty-six by forty-one feet. The massive pillars once supporting its roof near what must have been an atrium-like central opening, still rose in their places and looked also most impressive (see Fig. 49). Unfortunately, except for these solitary posts, the walls of this hall, i, together with any objects which may have once been left between them, proved completely eroded, and the sand covering the ground lay only one or two feet high. Quite close to the north-east there lay some scattered timber débris from a dwelling, ii, also completely destroyed by erosion. Neither here nor in the hall did the clearing effected yield any finds.
But as I traced eastwards the scanty indications left by the walls of another large and badly Discovery of
eroded structure, iii, I realized quickly that most of the ground underneath it was made up of the large refuse
layers of a huge refuse heap. From north-east to south-west it extended for over fifty-three feet, and its width seemed about fifty. Previous experience, such as I had gained at that rich mine, N. xv, in 1901, supplied sufficient reason for digging into this unsavoury quarry, though the pungent smells its contents emitted, even after more than sixteen centuries of burial, were made doubly trying by a fresh easterly breeze which drove the fine particles of dirt, etc., impartially into one's eyes, throat, and nose. A wind-eroded depression eastwards facilitated the work of my trenches, and thus it was soon ascertained that the great mass of these rubbish accumulations consisted of horse and camel dung mixed with plentiful straw and twigs. Rags of various fabrics, apparently woollen and hempen, as well as torn pieces of felt, dressed leather, and skin, also turned up in them. By the evening I was able to recognize that some wooden posts, of which the tops had just been visible above the surface of the ground, belonged to a small boarded enclosure, about eight by six and a half feet, completely embedded in the refuse. The clearing commenced here