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

what we are to understand by « domestic brazil-wood » (the word domesce is left out by RR, 280);
Yule's interpretation (Y, II, 276), « the brazil we make use of », which follows the reading of FA,
cannot be retained; only Moule's version, « cultivated brazil-wood », makes good sense. Yet
I was in doubt about the correctness of F and Z, until I noticed that both « verzino dimestico » and
« verzino salvatico » occur in Pegolotti (ed. Evans, 295, 296). Although Polo does not use the
word « domesce » in the chapter on « Lambri », he gives there (Vol. I, 376) a description of what
certainly is the process of cultivating brazil-wood. In modern times, the brazil-wood of Malabar
is both wild and cultivated (cf. Y, II, 380).

275. LOP

job VL TA¹, TA², V, VA, lope VB
lop F, Fr, t, FB, L, LT, P, Z; R loup FA

For « Lop », the modern Charkhlik (Čarḫlïq), on the southern side of the Lop-nör, cf. Stein,
Serindia, index, 1546. The etymology is unknown, but the name, with an initial n-, goes back
to pre-T'ang times. Hsüan-tsang's 納縛波 Na-fu-po, *Navapa, of the 7th century, is a sanskrit-
ization of *Nop, which is written Nob in Tibetan documents of c. 800. A colony from the Lop
region, which settled in the 6th cent. west of Qomul (see « Camul »), was known as 納藏 Na-chih,
*Napčïq, in T'ang and Sung times, and its name is still Lapčuq (cf. my paper in JA, 1916, I,
117-120). F. W. Thomas (BSOS, VIII, 793-794; and cf. Bacchi, Deux lexiques sanskrit-chinois,
II, 360) has proposed to see another transcription of *Napčïq in the « Dapici » of a « Saka » (Khotanese)
document of the 10th cent. The correspondence would be satisfactory if we could account for
the absence of the final guttural consonant in Khotanese.
Navapa presupposes the form « Lop », with -p, and the name is thus spelt by Polo; the Tibetan
Nob proves nothing, as the Tibetan script admits of no final -p. On the other hand, our Lop-nör
is a Mongol form (with Mong. nör, « lake »), and theoretically there is no -p in Mongolian, so that
we ought to transcribe Lob-nör; the transcription in Chinese texts of the 13th cent. renders Lob;
but the Turkish local pronunciation is now Lop-nör.
Postal stages were established at 羅卜 Lo-pu (Lob) in 1282 and 1286 (YS, 12, 3b; 14, 1a).
The city of 卜 Pu or 格卜 Ko-pu, near Čärčän (see « Ciarcian »), in a Ming itinerary (China Review,
v, 233) is probably a misreading for 洛卜 Lo-pu, Lob (Lop).
In JA, 1916, I, 119, I had proposed to trace the name Lop (or Lob) to Han times, and to reco-
gnize it as the first element in the name of the kingdom of 樓蘭 Lou-lan (sometimes written 牢蘭
Lao-lan); this suggestion must be abandoned. We know now, from other sources, that the native
name of Lou-lan must have sounded *Krorän (or *Grorän?); cf. TP, 1931, 459-460. This would
confirm the otherwise theoretical view that 樓 lou was pronounced as *glou in Han times; but
then the same would hold good for 牢 lao (*lâu < *glâu), and the conclusion would be that,
contrary to the usual opinion, lao is probably not a purely figurative character (« an enclosure for
oxen »), but is phonetically connected with 牛 niu (*ngigu), « ox » (? cf. Tib. glaṅ, « ox »).