easternmost ridge. We found in it the body of a woman laid out on a rough woollen mat (III. 4. 03 is a fragment of it) and covered by a hollowed-out trunk. The body, though poorly preserved, still retained parts of an upper garment of silk canvas edged with crimson silk, III. 4. 02, and remains of trousers of a strong woollen material, III. 4. (DI. The head was wrapped round with strips of a fluffy material resembling cotton-Wool ; but there was no head-band nor any metal ornament. The feet were cased in coarse leather moccasins. The only sepulchral deposit traceable was a circular wooden tray of rough make with a lamb's head on it.
This was the last of the graves opened by us at the site. Having regard to the time which the digging involved, with only a few men available for the work, and to the work still to be done in
desert ground ahead, I did not consider myself justified in continuing an investigation which had
already disclosed, on the one hand the poor preservation of the contents of the graves, and on the other the uniformity of the customs illustrated by them. In respect of these burial customs it is
easier to recognize the difference between them and those previously observed, whether in the
ancient Chinese graves of Lou-lan or in the indigenous graves of L.F., L.Q., L.S., and L.T:, than to indicate definite evidence which would enable us to fix the approximate period and the
race to which the occupants of these graves belonged. The use of miscellaneous rags of old clothing
for wrapping up the dead, which forms so characteristic a feature of the remains of Chinese burials of Han times found at L.C. and L.H., was certainly not adopted here. At the same time the
employment of silk fabrics for shrouds and the different character of the sepulchral deposits plainly
show that the people here buried lived under the influence of Chinese civilization and were considerably advanced beyond the manner of life of those autochthonous Lou-lan people with whose remains
we have become acquainted at L.F. and at other Lou-lan cemeteries of the same primitive type.
Looking for analogies on other comparatively near ground, we certainly find the use of plain silks for shrouds, and of simple but complete garments beneath them, illustrated by a number of the
Astâna burials in Turfân. But the regular face-cloths so common there are absent in the Ying-p`an graves, while of the swathing of the heads and of head-bands tied across the forehead no instance was met with at Astâna.
The indications gathered from these few, though in essentials uniform, burials of the Ying-p`an site may be insufficient in themselves for any safe conclusion. Yet no harm will be done by recording
the impression they conveyed to me, purely conjectural as it is. Taking into account the fact that
the graves are all orientated with the foot end to the west just as at L.S. and L.T., and that the rows of posts marking them on the surface may be derived from the solid stockades found above
the graves of these indigenous burial-grounds, it occurred to me that the bodies buried at Ying-p`an
may, perhaps, be those of local people settled around the old Chinese station at a period when prolonged contact with Chinese civilization had considerably modified their habits. It might
thus be possible to account on the one hand for the adoption of certain Chinese customs not merely in the dress, &c., of the living but also in funeral equipment, and on the other for the maintenance of traditional arrangements in respect of the abode of the dead.
But whether the above suggestion is accepted or not, it cannot help us much towards an approximate dating of the graves. The influence of Chinese civilization must have been strong and continuous in these parts from the first century B.C. onwards, and there is ample evidence, both in the Tarim basin and in Turfân, that it outlasted the decline of China's direct political power in the ` Western Regions ' which set in during the second century A. D. In the Lou-lan region, moreover, the existence of direct Chinese control down to the first half of the fourth century A. D. is attested by documents from L.A i1 Thus, on general grounds, the graves might be attributed
11 Cf. Serindia, i. pp. 408 sq.