国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

32

D'ANVILLE.

country is here too little known to make it possible for anyone at all to settle the question, but d'Anville may be right, or rather the sources which he has used.'

Of great interest is d'Anville's Temen M. N.E. of Tengri-nor, which probably corresponds to Abbé Hue's Tangla. Klaproth shows that the name Temen belongs to a plain, Temen tala, situated north of Iké Nomkhoûn oubachi and south of the river Baka Akdam. »Dans les cartes de d'Anville, et par conséquent dans toutes les postérieures, cette montagne (Iké Nomkhoûn oubachi) est nommée par erreur M. Temen.»2 Very likely this Temen M. or Tangla may be the eastern continua-

tion of one of the great Kara-korum ranges.

If nothing else can be learnt from d'Anville's maps regarding Transhimalaya, one would always suspect, when looking attentively at these maps, that there could not possibly exist one continuous range, but very probably a whole labyrinth of different and more or less separated ranges. One would think that perhaps in reality there existed a great number of ramifications stretching out their arms in different directions from a principal range. But there is no sign of such a range on d'Anville's map in the region corresponding to the one I explored. If HODGSON and SAUNDERS had studied the Chinese maps attentively they would have hesitated to enter on their maps a continuous mountain range parallel with the whole of the Tsangpo and without ramifications. But as they did so they must have regarded d'Anville's maps and other maps founded on Chinese material as very unreliable.

To do so would indeed be unjust, for we have seen that the map is very uneven and of different value in different parts. We have seen that extremely seldom, practically only in the cases of the sources of the two great rivers, the mountains are correct. In three or four cases the lakes are rather good, in some cases rivers which do not exist, as the Tarcou Tsanpou and probably the Nacoi R., are drawn on the map. We arrive at the conclusion that some parts have really been roughly surveyed, others filled in from verbal information, which is always misleading. Therefore such lakes as Shuru-tso and Dangra-yum-tso have been placed in a way which has not the faintest resemblance to the truth. Such lakes as Teri-nam-tso, Poru-tso, Shovo-tso, Nganglaring-tso and others cannot be found on d'Anville's map. On the Ta-ch'ing map the Dzolmié thang is the lake which both in form and situation has the greatest likeness to Nganglaring-tso. It is to be found on d'Anville's map as well, north of M. Mouron. If, on d'Anville's map, Cal M. should be meant to represent Ka-la, a very important pass on the »gold-road», the lake, from which Nacoi R. comes would be Nganglaring-tso. But one loses oneself in guesses when trying to identify the representation on d'Anvilles map.

I The marked water-parting close to the whole northern bank of the Tsangpo and Raga-tsangpo as represented on the map of the Tibet Frontier Commission, 1904, which gives the impression of a continuous range running the whole way parallel with and very near the Tsangpo is certainly wrong. That this water-parting was really accepted by geographers as an uninterrupted range, can easily be

seen on map N:o 62 in Stieler's Hand-Atlas for 191o. Compare also edition 1911, Pl. XXXI below. 2 Mémoires Relatifs à l'Asie, p. 391.

.t

,e

'   I r