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

of R. B. Lāl Singh's route in the south-
western corner of the sheet is adjusted on
the positions of Āltmish-bulak, its starting
point, and of Turfān.

This route at Donglik, near its northern
end (A. 4), struck an old desert track from
the terminal basin of the Hāmi river, once
used by hunters of wild camels before certain
salt springs along it had completely dried
up. A more northerly route which M.
Muhammad Yakūb surveyed from the same
basin to Chik-tam (B-D. 3) has also become
impracticable to traffic for the same reason.

Apart from these desert plateaus in the
south, which in their utter barrenness fully
share the character of the eastern Kuruk-
tāgh, there falls within this sheet the small
portion of the Turfān basin around Pichan
and Chik-tam. Like the rest of the cultiva-
ted area of the basin these oases owe their
irrigation almost exclusively to Kārēzes

Astronomically observed latitudes.
1906-08. Tung-yen-tzu, Camp 262 (close to Chinese station ; C. 2) 43° 29′ 10″
Pichan, Camp 265 (Bēg's house above W. tank of river bed, circ.
1 mile from town ; A. 3) ... ... ... 42° 51′ 56″
1913-15. Jam-bulak, Camp 236 (B. 1) ... ... ... 43° 39′ 16″
Jōjan-kārēz, Camp 239 (village inn, 2 miles N.W. of Chik-tam
post ; B. 2) ... ... ... 43° 1′ 13″
Tügemen-tāsh, Camp 271 (in patch of serub ; A. 4) ... 42° 6′ 13″

NOTES ON SHEET No. 32 (ANCIENT LOP LAKE BED)

With the exception of the caravan track
leading along the southern shore of the an-
cient salt-encrusted Lop sea bed which had
been followed before in 1907, all the surveys
shown in this sheet date from my third
expedition. Those in the northern half of
the sheet belong exclusively to R. B. Lāl
Singh's work of the winter of 1915, while
most of those further south were carried
out by Miān Afrāz-gul and myself.

In the S.E. corner the position of Kum-
kuduk (D.4), on the caravan track from
Charkhlik to Tun-huang, was fixed by ad-
justing the traverses on the positions adopt-
ed for An-hsi (see Sheet No. 38) and Mirān
(No. 30. B.2). The longitude thus derived,
91° 55′ 30″, was found to agree very closely
with the one shown in Sheet No. 67 of the
1906-08 Map. For the correction of its
latitude the observations taken in 1914 on

which catch the subsoil drainage from the
elevated portion of the eastern T'ien-shan.
To the east of the meridian of Chik-tam the
crest-line of the range falls considerably, and
the moisture it receives is even on the nor-
thern slope too scanty to permit of cultiva-
tion in more than a few small patches (see
B-D. 1). It is only to the west of that
meridian that conifer forest is found on the
northern face of the range, and a small
village tract (Mu-li-ho, A. 1) with some
grazing.

The southern slope of the range is far
more barren still, and the passage of the
Chinese highroad along its foot is made
possible only by rare springs and wells and
some scanty vegetation found in small ba-
sins (B-D. 2). The pass by which it crosses
from Ch'i-ku-ching to the north of the
range is low enough to be practicable for
carts.

the route both west and east of Camp 95
could be utilized.

The traverses of the routes followed to
Kum-kuduk from Āltmish-bulak and the
ruins north-east of the Lou-lan Site (A.3),
were adjusted on the positions adopted for
the latter two points as well as on Kum-
kuduk. ¹⁸ The correction in longitude which
the revised computation of R. B. Lāl Singh's
triangulation indicates for Āltmish-bulak
and the Lou-lan Site, has been referred to in
the Notes on Sheet No. 29. The traverse
carried by R. B. Lāl Singh from Yetim-
bulak (A.3) northward through wholly un-
explored parts of the Kuruk-tāgh was ad-
justed on the adopted positions of Āltmish-
bulak and Deghar (in the Turfān depression;
No. 28. D.4). A useful check was afforded
by the series of latitude observations taken
along this route (see below).