Sec. v] SURVEY OF LIMES LINE TOWARDS AN-HSI, T. xxxi xxxv 607
the basis of the measurements I was able to make, to at least 4,000 cubic feet per second, if not more. But the wide marshy belt within sight to the west showed clearly that none of this water
could possibly be utilized for cultivation. •
Overnight the wind increased to a Burân, this time from the west, and the consequent murky Return
condition of the atmosphere for some days left no chance of a further search for the Limes line from Limes
to Tun-
eastwards. The town of An-hsi, to which it would have carried me, I was in any case bound to Huang. visit later. So I decided for an early return to Tun-huang in order to spare time for the explorations awaiting me on the west. The main object which had prompted this expedition to the north-east was already secured. It had proved that the remains of the ancient wall actually continued east of Tun-huang, as I had conjectured from the first. In addition, I now carried back indisputable evidence in the shape of exactly dated records proving the occupation of this Limes in the first century A. D. The trying experiences undergone on the three days' march back to Tun-huang town, first along the wall and then via T. xxx and Shih-tsao, amidst icy gales and driving sand, have been described in my Personal Narrative, and need not be told here again. They made me realize fully the life led by those who once guarded this desert border. On April 3 my old camping-place was regained.
IIIF Color HighRes
