The ruined dwelling N. XLII (Fig. 98) comprised two portions, both badly decayed through erosion. The smaller one to the west was built with wattle of vertically placed tamarisk branches, which were still standing to about two feet in height. On clearing the smaller of the two rooms, i, a number of miscellaneous small objects were recovered, including a carnelian bead, the bowl of a horn spoon, glass fragments, &c. In the other the roofing of reed bundles had fallen in and was found covering part of the floor. From one of the corners a quantity of oat straw was brought to light mixed with sheep-dung, indicating that at one time or another the place had been used as a sheep pen. The larger and more solidly built structure to the east had suffered worse by erosion, and of only two of the outer walls could the timber foundations be traced, together with the lowest portion of the wattle-and-plaster. One well-worked foundation beam measured 23 feet in length with a thickness of 8 inches. Among the plentiful debris of fallen timber, all badly splintered and shrivelled, was a circular post about 8 feet in height, which once supported the roof of a hall as well as the double cantilever inserted between the post and the ceiling. Scanty as were these finds, they left no doubt that the ruin belonged approximately to the same early period as the remains previously explored at the site farther north. This also agreed with the depth of erosion, 14 feet, observed in the immediate proximity of the structure. A row of dead Terek (Populus alba) trees close to it still stood upright to a height of 10-12 feet.
The second ruin, N. xi.iii, reported by `Azim was found about a furlong to the west, situated among close-set tamarisk-cones and consequently deeply buried in sand (Fig. 94). Its two main rooms (see plan, P1. 5) were filled with sand to a height of about eight feet, the removal of which was a laborious task and yielded only a few small objects in wood and metal (Pl. XVI), besides pieces of large pottery jars. The construction of the dwelling, though rough, showed two peculiarities of interest. The room to the east, i, provided with a sitting platform running round three sides, as so often in previously explored dwellings of the Niya Site, had wattle walls of a make that I had not observed elsewhere ; the elevation in Pl. 5 is intended to illustrate it. It consisted of two thicknesses of matting, closely woven in a diagonal pattern, with a layer of about three inches of mud plaster between them. I found no trace of a plaster facing outside, as is usual in the case of wattle walls with diagonal matting. The other peculiar feature was the recess formed by a wall with re-entering angles in room ii (see plan, P1. 5). Its purpose remains uncertain ; it might possibly have served for the accommodation of stairs leading to an upper story.8 The posts securing the wattle walls of vertical reeds in this room stood to a height of only 6 feet 6 inches, while the central pillar found in room i indicated for the latter a clear height of over 8 feet. In both rooms the finds consisted only of a few fragments of household objects, as described in the List.
The reports of the search-parties sent out in different directions on my first arrival at N. XLII indicated the presence of old remains at a few points to the S. and SW. A visit paid to these on December i4th showed that the ruin in the latter direction, about half a mile distant, was only that of a small and poorly built rush-wall structure, probably used as a cattle-shed. Another ruin, found at a distance of about one mile to the south, was that of a dwelling-place, measuring about 62 by 42 feet outside and built with wattle of closely woven diagonal matting like the one in N. XLIII. From the badly splintered condition of the wooden framework disclosed by a trial clearing, it was evident that the remains had been long exposed before being covered by drift-sand, and as this lay now to a height of six feet or so I did not spare time for complete excavation. Two small structures still farther to the south, from which the few relics marked N. XLII.0i 3 had been brought, were reported to be in a similar eroded condition ; want of time obliged me to leave these unvisited.
7 See, e.g. Ancient Khotan, i. p. 317 ; ii. PI. VI; Serindia, 8 Cf. Ancient Khotan, i. p. 377 ; ii. Pl. XXxIV.
i. p. 215, Fig. 47.