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

Clementi's latitude of 39° 46′ 42″ for Marâl-
bâshi town and several latitudes observed by
Dr. Hedin on the Yârkand river were used.
The longitude adopted for Marâl-bâshi (78°
15′ 15″) was obtained by fitting the several
traverses passing through it to the accepted
positions of Kâshgar and Ak-su and propor-
tionately dividing the error in distances. ⁶
Similarly the position of Tumshuk was
adapted to that shown by Sheet No. 7 for
Kelpin and Yaka-kuduk. It may be noted
that the several isolated rocky heights of
the Mazâr-tâgh and other hills rising above
the alluvial plains are of great help to the
surveyor for the correct location of details
in the vicinity of Marâl-bâshi.
For brief descriptive accounts of the
routes followed in this area, see Desert Ca-
thay, ii. p. 428 ; Geogr. Journal, xlviii. pp.
112 sq.; Serindia, iii. pp. 1309 sqq.

The oasis of Marâl-bâshi is of interest
on account of its irrigation system in which
the inadequate water-supply received from
the terminal Kâshgar river is supplemented
by big reservoirs (A. 1) fed from the Yâr-
kand river. The dykes on the latter (Sheet
No. 5. D. 2) securing water for the Zai canal
and those around the reservoirs of Kölde and
Anâr-köl represent the largest irrigation
work in the Târim basin. The geological
interest presented by the Mazâr-tâgh and
the other isolated hills, as the remnants of an
ancient range distinct from the T'ien-shan,
has been touched upon above, p. 26, as well
as the formidable character of the drift-sand
desert which extends south of the Yârkand
river.
Correction 8. C. 1. Cultivation patches
ought to have been shown around the habi-
tation symbols of Kizil-âirat-terelgha.

NOTES ON SHEET No. 9 (KHOTAN, SANJU)

This sheet shows almost the whole of
the mountainous area which extends from
the edge of the Taklamakân between Gûma
and Khotan and southwards to the main
K'un-lun range and the upper Kara-kâsh
river. The routes from which it was sur-
veyed all belong, with a slight exception in
the extreme north-east, to the first and second
expeditions. For the southern portion of
the sheet, the triangulation of 1900 supplied
a considerable number of fixed points ;⁷ in
addition astronomically observed latitudes
were available for numerous camps. For
Khotan town a longitude of 79° 55′ 51″ was
adopted, this being the mean of the astro-
nomical values obtained by several observers
(Dutrenil de Rhins, Piewzoff, Deasy) and in
close agreement also with the position which
our triangulation of the hills due south of the
oasis indicates for the town. ⁸ The positions
adopted for Khotan and Karghalik (see
above under Sheet No. 6) were used in fitting
the traverses between the two cases, both by

the main road and along the outer hill; on
the former route available latitude observa-
tions were also utilized.
The approximate levels indicated for the
snow-line in different portions of the main
range are uncertain, most of the work having
been done in the early summers of 1900 and
1908 and by different surveyors. In the
mountains due south of Khotan my own
observations of the autumn of 1900 and late
summer of 1906 have led me tentatively to
assume an average snow-line at 17,000 feet.
The routes followed by myself in the
plains and mountains and the observations
attending my repeated sojourns in the
important Khotan oasis have been described
in Ruins of Khotan, pp. 169 sqq, 462 sqq.;
Desert Cathay, i. 155 sqq., ii. 416 sqq. The
geography of the Khotan region, in its physical,
economic and historical aspects, together
with the route from Karghalik and the topo-
graphia sacra of the Khotan district, is fully
discussed in Chapters v-viii of my Ancient