small oasis eastwards, where I had to prepare rapidly for fresh exploration at the ancient site in the desert northwards. It was encouraging to learn from my old ` treasure-seeking' guide Ibrahim, ` the miller ',2 that the further search I had enjoined him to make for ancient dwellings hidden away amidst the dunes had been fruitful. It was equally pleasing to see how readily my old Niya diggers of 190I rejoined me. I was resolved this time to take out as many labourers as I could possibly keep supplied with water at the site, and thus to expedite the excavation of any newly discovered ruins. So it was encouraging that, with the example set by my old guard ' and the influence still possessed here by Ibrahim Beg, my energetic old Darogha, who had rejoined me at Khadalik, a column of fifty men, with additional camels and supplies for four weeks, could be raised within a single day's halt.
I have shown in my former Report that the present Niya corresponds to the town of Nil-fang iyt f, which Hsüan-tsang mentions as the eastern frontier station of Khotan,3 and that it is meant probably also by the Nina to which one of the Kharosthi documents excavated by me in 1901 at the Niya Site refers. Nor need I give details about the next three marches which carried me down to the present end of the Niya River, seeing that I have described them in my Personal Narrative. An account of the riverine forest belt traversed and of the curious desert shrine of Imam Ja`far Sadiq near the river's end has already been given in my former Detailed Report.4 By October 18 we reached Tülküch-köl, a small lake fed by springs from the dying river, some four miles below the Mazar, where there is a little cultivation occupying some fifteen men in a clearing amidst the luxuriant jungle (see Map No. 37). Next morning, after making a depot there of all stores not immediately needed and filling all available water-tanks and goat-skins, we left behind the last abode of the living and also the present end of the life-giving river.
I was anxious to move that day as far as possible northward amongst the ruins of the ancient site to which my thoughts had turned so often since those happy labours of the winter of 1901. But unexpected finds en route delayed my arrival. I had just passed, about five miles below Tülküch-köl, the last deserted shepherd's hut in the gradually thinning jungle, when Ibrahim and his fellow guides told me of some remains they had discovered since my visit among the high and closely packed tamarisk-cones east of the route. After making my way through these for about a mile and a half north-eastwards I found myself suddenly in a small open area measuring about two hundred and forty yards from north to south, and about half that distance across.
The much-eroded ground showed remains of fences with fallen trunks of poplars and mulberry-trees, marking ancient orchards, and near its centre the débris of somè dwellings constructed in timber and plaster. Owing to the far advanced erosion of the open ground there was no cover of sand or refuse to offer protection to these scanty relics. But in one place the foundation beams of walls could still be traced in situ. They extended over a space of about fifty-seven by twenty-six feet, a single beam, on one side of what evidently was the main room, measuring thirty-two feet in length. Little, however, survived of the wattled walls which once rose above them. Yet even thus my familiarity with constructive peculiarities previously observed sufficed to convince me that the remains dated back to the same period as those of the main site ahead, i. e. the third century A. D.
The whole occupied a small plateau rising about eight feet above the eroded ground, and isolated witnesses edged it. To the north-west close by the foot of one of the encircling tamarisk-cones a row of big trunks of dead mulberry-trees still rose upright. Long centuries of exposure to the desert winds had left nothing for me to dig here. But the mere fact of the area of ancient