RECORDS OF THE JOURNEY
distances from each other. About 2 miles before reaching Lianchow we crossed a last river bed and then gradually began to ascend a gentle slope of nothing but gravel. Small heaps of gravel, half in the shape of dunes, half sugar-loaves, were thickly scattered over it, all with a little window at the bottom. These were Chinese burial mounds with an opening to allow the air to circulate.
At Kanchow I was struck by the bogs and morasses that almost penetrated into the January 21st. house I was living in; at Lianchow it was the gravel and stones that left an indelible impres- Lianchow. sion on me. The town is about i 1/3 mile from W to E and 2/3 of a mile from N to S, and is enclosed on the S, W and halfway from the N by a huge Chinese cemetery, its innumerable small stone cairns, packed tightly close to each other, making the most desolate impression imaginable. Beyond this sea of stones there are single houses surrounded by some trees or temples embedded in a shady grove, from which the graceful lines of the roofs stand out beautifully against the background of mountains or sky. They look like small oases in a desert of stone. The mountains in the S arc a few dozen li distant. To the N there is an open plain with a slight drop northward. The land seems to be more tilled and the houses more numerous in that direction. In the immediate vicinity of the town, to the S and especially to the N, there are several larger groups of temple buildings, built high up and looking picturesque among the surrounding trees. From these a belt of trees and gardens runs along a narrow river bed to the NE corner of the town and to the N wall of the suburb of Tung-kuan. Small groups of houses have formed near the 3 other gates of the town, but outside the E gate there is quite a little town-quarter enclosed by its own wall, though very dilapidated. It is called Tung-kuan. Here Dungans are allowed to live and here you find the biggest number of sarais, some of them quite good. The town is like all the other larger Chinese towns, at any rate in Kan Su. Two big streets, as straight as arrows, connect the four gates of the town and divide the space into four equal parts. At the point where they cross, the customary kulo tower is replaced by 4 large wooden gates which form a square with openings towards the four points of the compass. The principal sight worth mentioning is an old stone with an inscription said to be in the Sisia language. It stands in the NE part of the town at the foot of two very tall and similar, pillar-like towers, visible at a distance of many miles from Lianchow. The stone is in a very bad condition. On its front is carved the inscription in the Sisia language, on the back another in Chinese. The Sisia letters might be taken for Chinese characters, but Chinese cannot read them. When Bonin was here, he was said not to have been able to photograph it or take an impression of it, but an impression is said to have been taken by the Chinese authorities at the request of a French ambassador at Peiping. We tried in vain for two days. It was so cold that everything froze before we could get the cloth into the hollows. Finally I had all the hollows painted with white paint and photographed the stone to-day after working all night. It is possible that mistakes may have occurred, especially as the work had to be done at night by candle light. Opposite this stone there are two others with inscriptions. These stones seemed to be in an unusually good state of preservation, and on making enquiries I found that at any rate one of them had only been
) 489
IIIFカラー高解像度
