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

and sixty [double] paces in width; it cannot be crossed except on boats of down; it flows towards
the south and falls into the sea. The Anavatapta Mountain is the K'un-lun Mountain. This is in
agreement with the Ta-huang hsi-ching ». Here ends, in my opinion, the quotation by Chang Shou-
chieh of the assertions of late scholars, and then comes his refutation. « The kingdom of Ta-Ch'in,
however, is on an island in the Western Sea; from the western boundary of An-hsi it is reached,
crossing the sea with a favourable wind, in three months, and the Weak River is still to the west
of that kingdom. The Weak River of the K'un-lun mountains flows north of the Kingdom of
Women and rises south of the K'un-lun Mountains. The Kingdom of Women is 2 700 li south
of Khotan, and Khotan is at 9 670 li from the capital. Calculation shows that there are some
40,000 or 50,000 li between Ta-Ch'in and the Great K'un-lun Mountains. No account has been
taken of this, and hence the error of former scholars. All this argument is based on the geography
of the Han (括 地 k'uo-ti; it may be that Chang Shou-chieh here alludes more precisely to the
K'uo-ti t'u, a geographical work the exact date of which is not ascertained, but which certainly is
prior to the 6th cent.; see « Cotton », t. I, p. 493), and I am still afraid that it may not be quite
certain; but the fact remains that by all accounts there were two 'Weak Rivers'.» It seems to me
impossible that Chang Shou-chieh, a man of real learning, should have ascribed to the Shan-hai
ching a mention of the Anavatapta Mountain; and the trend of his reasoning is to refute scholars
who maintained that the Weak River of the K'un-lun of ancient Chinese lore was the same as the
Weak River west of Ta-Ch'in and that both took their rise south of the Anavatapta, identified with
the ancient K'un-lun. Who these scholars were remains uncertain. There is some probability
that they relied on Tao-an's Hsi-yü chih, a work in which the identity of the Anavatapta Mountain
and the K'un-lun was more than once asserted; but in the quotations of this Hsi-yü chih which are
extant, no mention is made of the Kingdom of Women. Even if Chang Shou-chieh is alluding
to other texts, the fact remains that early Chinese legend and Buddhist cosmogony had combined
to locate to the extreme north-west of India a Kingdom of Women through which the Weak River
flowed.
This must be remembered to account for a puzzling section in Hsüan-tsang's Hsi-yü chi.
In his description of the countries in the basin of the upper Ganges, he devotes successive notices
to (1) Sthāṇvīśvara, (2) Srughna, (3) *Matipura, (4) P'o-lo-hsi-mo-pu-lo, (5) Ch'ü-p'i-shuang-na,
(6) Ahicchattra, (7) P'i-lo-shan-na, and (8) *Kapitha. Before we proceed, some remarks must be
made on these eight names : (1) Sthāṇvīśvara, or Sthāneśvara, Thanesar; cf. JULIEN, Mémoires,
I, 211. The transcription seems to be based on Sthāneśvara. (2) Srughna, or Śrughna; cf. Mém.,
I, 215. Hsüan-tsang's transcription renders Srughna; cf. JA, 1915, I, 36, 46, 97, for both forms.
From Srughna Hsüan-tsang visited the source of the Ganges. (3) *Matipura, unidentified; cf.
Mém., I, 219. In the north-western part of *Matipura was the city of Mayūra and, at a short distance
from Mayūra, the « Gate of the Ganges », Gaṅgādvāra, the present Hardvar; cf. Mém., I, 230;
SACHAU, Alberuni's India, I, 199. (4) 婆 羅 吸 摩 補 羅 P'o-lo-hsi-mo-pu-lo; cf. JULIEN, Vie,
110; Mém., I, 231. JULIEN restored the name as Brahmapura, and this was accepted, with some hesi-
tation, by WATTERS, On Yuan Chwang's Travels, I, 329, and is stated to be correct by A. H. FRANCKE
in JRAS, 1910, 489, where he says that it represents the modern Brahmaur, the ancient capital
of the Chamba State. A city of Brahmapura is mentioned in the Bṛhat-saṃhitā and by Al-Bīrūnī