National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

which also was called Kizil-ling.¹ The water-courses coming down from the four mountains
join at the foot of the pass to a little lake. The pass was red and barren, the grass of
the cliffs dry and short. Cold winds blow there. Therefore Sung Yün says, the *Ts'ung-ling*
is steep, and neither grass nor trees are growing there.²

The reading *Han-p'an-t'o = Alai* seems to be very improbable. But if now, as the
author of the Hsi-yü-shui-tao-chi will have it, the *Dragon Lake* is identical with the *Kara-
köl*, Hsüan-chuang's itinerary would be as follows: From Mastuj 700 li or about 200 miles
N. E. to Pamir, and the Dragon Lake 560 li or about 135 miles S. E. to Tashkurgan,
which both in relative distance and direction would be fairly correct. But the description
of the lake does not agree with the reality, as it is bitter salt and has no effluents. It
is also extremely improbable that the pilgrim, who was no professional explorer, would
have chosen to cross the Pamirs by the most difficult portions. He has certainly preferred
the usual trading and caravan roads.

As to Wood's lake and *Chakmakden-köl* both the relative distances and the second
direction are not in accordance with reality. But both lakes are fresh, and the general
description of the country agrees very well with the Pamirs as we know them now. With
perfect certainty nothing can be determined, though in the case of Hsüan-chuang we are
on surer ground than regarding the earlier pilgrims. Finally we have therefore to make
a choice between *Sor-köl* and *Chakmakden-köl* as having witnessed the passage of Hsüan-
chuang's caravan. On account of the hydrographical part played by the latter, and as it
is easier to commit a mistake regarding the discharge from this lake, and more especially
on account of the greater practicability of the road, I prefer to draw Hsüan-chuang's itinerary
along the *Chakmakden-köl*.

4. THE EASTERN LIMITS OF THE TS'UNG-LING.

From *Tash-kurgan* the pilgrim travelled 300 li or 60 miles to the S. E., 200 li or
40 miles to the N. E., 800 li or 160 miles to the east, *Wu-sha*, from there 500 li or 100 miles
to the north, *Kashgar*. If the least confidence could be attached to those distances and
bearings, Hsüan-chuang should have reached very nearly the point in the *Takla-makan
desert* not far west of Khotan-darya, where my caravan perished May 1st, 1895, that is
to say, 4° 20′ too far east, and half a degree too far south. But we must remember that
the li are reckoned much shorter in the mountains than in the plains, and that their length
depends upon the practicability of the roads. The roads of Eastern Pamir belong to the
most difficult in all Central Asia. The salient point is also the geographical names given
by Hsüan-chuang and absolutely determined by European sinologists. As regards the
physical geography we know that it was only after marching the 800 li that the pilgrim
came out from the eastern foot-hills of the *Ts'ung-ling* mountains.

According to Herrmann these indications which expressly point to the east, can only
correspond to a route to *Yarkand*. As it, from its starting point Tash-kurgan, is said to
run 300 li in a south-eastern direction, the only road that may be taken into consideration
is the one which is determined by the following stations: Kandahar-davan, the Tong valley,
Unkurluk, and Damsin.³ The present *Yarkand* is so much the more to be regarded as