XXXIII., p. 448.

PERSIAN.

With regard to the knowledge of Persian, the only oriental language probably known by Marco Polo, Pelliot remarks (Journ. Asiat., Mai-Juin, 1912, p. 592 n.): "C'est l'idée de Yule (cf. par par exemple I., 448), et je la crois tout à fait juste. On peut la fortifier d'autres indices. On sait par exemple que Marco Polo substitue le lion au tigre dans le cycle des douze animaux. M. Chavannes (Toung pao, II., VII., 59) suppose que 'cette dernière différence provient sans doute de ce que Marco Polo connaissait le cycle avec les noms mongols des animaux : c'est le léopard dont il a fait le lion.' Mais on ne voit pas pourquoi il aurait rendu par 'lion' le turco-mongol bars, qui signifie seulement 'tigre.' Admettons au contraire qu'il pense en persan : dans toute l'Asie centrale, le persan šir a les deux sens de lion et de tigre. De même, quand Marco Polo appelle la Chine du sud Manzi, il est d'accord avec les Persans, par exemple avec Rachid ed-din, pour employer l'expression usuelle dans la langue chinoise de l'époque, c'est-à-dire Man-tseu; mais, au lieu de Manzi, les Mongols avaient adopté un autres nom, Nangias, dont il n'y a pas trace dans Marco Polo. On pourrait multiplier ces exemples."

XXXIII., p. 456, n. Instead of Hui Heng, read Hiu Heng.

BOOK SECOND.

PART II.-JOURNEY TO THE WEST AND SOUTH-WEST OF CATHAY.

XXXVII., p. 13. "There grow here [Taianfu] many excellent vines, supplying great plenty of wine; and in all Cathay this is the only place where wine is produced. It is carried hence all over the country."

2

Dr. B. Laufer makes the following remarks to me: "Polo is quite right in ascribing vines and wine to T'ar Yüan-fu in Shan Si, and is in this respect upheld by contemporary Chinese sources. The Yin shan cheng yao written in 1330 by Ho Se-hui, contains this account1: 'There are numerous brands of wine: that coming from Qara-Khodja (Ha-la-hwo) is very strong, that coming from Tibet ranks next. Also the wines from P'ing Yang and T'ar Yuan (in Shan Si) take the second rank. According to some statements, grapes, when stored for a long time, will develop into wine through a natural process. This wine is fragrant, sweet, and exceedingly strong: this is the genuine grape-wine.' Ts'ao mu tse, written in 1378 par Ye Tse-k'i,3 contains the following information: 'Under the Yuan Dynasty grape-wine was manufactured in Ki-ning and other circuits of Shan Si Province. In the eighth month they went to the T'ai hang Mountain, 4 in order to test the genuine and adulterated brands: the genuine

1 Pen ts'ao kang mu, Ch. 25, p. 146.

2 Regarding this name and its history, see PELLIOT, Fourn. Asiatique, 1912, I., p. 582. Qara Khodja was celebrated for its abundance of grapes. (BREtschneider, Mediaval Res., I., p. 65.) J. DUDGEON (The Beverages of the Chinese, p. 27) misreading it Ha-so-hwo, took it for the designation of a sort of wine. STUART (Chinese Materia Medica, p. 459) mistakes it for a transliteration of "hollands," or may be "alcohol." The latter word has never penetrated into China in any form.

* This work is also the first that contains the word a-la-ki, from Arabic 'araq. (See T'oung Pao, 1916, p. 483.)

4

• A range of mountains separating Shan Si from Chi li and Ho Nan,

kind when water is poured on it, will float; the adulterated sort, when thus treated, will freeze.1 In wine which has long been stored, there is a certain portion which even in extreme cold will never freeze, while all the remainder is frozen: this is the spirit and fluid secretion of wine. 2 If this is drunk, the essence will penetrate into a man's armpits, and he will die. Wine kept for two or three years develops great poison." For a detailed history of grape-wine in China, see Laufer's Sino-Iranica.

XXXVII., p. 16.

VINE.

Chavannes (Chancellerie chinoise de l'époque mongole, II., pp. 66-68, 1908) has a long note on vine and grape wine-making in China, from Chinese sources. We know that vine, according to Sze-ma Ts'ien, was imported from Farghânah about 100 B.C. The Chinese, from texts in the T'ai p'ing yu lan and the Yuan Kien lei han, learned the art of wine-making after they had ' defeated the King of Kao ch'ang (Turfan) in 640 A.D.

XLI., p. 27 seq.

CHRISTIAN MONUMENT AT SI-NGAN FU.

The slab King kiao pei, bearing the inscription, was found, according to Father Havret, 2nd Pt., p. 71, in the sub-prefecture of Chau Chi, a dependency of Si-ngan fu, among ancient ruins. Prof. Pelliot says that the slab was not found at Chau Chi, but in the western suburb of Si-ngan, at the very spot where it was to be seen some years ago, before it was transferred to the Pei lin, in fact at the place where it was erected in the seventh century inside the monastery built by Olopun. (Chrétiens de l'Asie centrale, T'oung pao, 1914, p. 625.)

In 1907, a Danish gentleman, Mr. Frits V. Holm, took a photograph of the tablet as it stood outside the west gate of Si-ngan, south of the road to Kan Su; it was one of five slabs on the same spot; it was removed without the stone pedestal (a tortoise) into the city on the 2nd October 1907, and it is now kept in the museum known as the Pei lin (Forest of Tablets). Holm says it is ten feet high, the weight being two tons; he tried to purchase the original, and failing this he had an exact replica made by Chinese workmen; this replica was deposited in

This is probably a phantasy. We can make nothing of it, as it is not stated how the adulterated wine was made.

2 This possibly is the earliest Chinese allusion to alcohol.

the Metropolitan Museum of Art in the City of New York, as a loan, on the 16th of June, 1908. Since, this replica was purchased by Mrs. George Leary, of 1053, Fifth Avenue, New York, and presented by this lady, through Frits Holm, to the Vatican. See the November number (1916) of the Boll. della R. Soc. Geog. Italiana. "The original Nestorian Tablet of A.D. 781, as well as my replica, made in 1907," Holm writes, "are both carved from the stone quarries of Fu Ping Hien; the material is a black, subgranular limestone with small oolithes scattered through it" (Frits V. Holm, The Nestorian Monument, Chicago, 1900). In this pamphlet there is a photograph of the tablet as it stands in the Pei lin.

Prof. Ed. Chavannes, who also visited Si-ngan in 1907, saw the Nestorian Monument; in the album of his Mission archéologique dans la Chine Septentrionale, Paris, 1909, he has given (Plate 445) photographs of the five tablets, the tablet itself, the western gate of the western suburb of Si-ngan, and the entrance of the temple Kin Sheng Sze.

Cf. Notes, pp. 105-113 of Vol. I. of the second edition of Cathay and the Way thither.

II., p. 27.

KHUMDAN.

Cf. Kumudana, given by the Sanskrit-Chinese vocabulary found in Japan (Max MÜLLER, Buddhist Texts from Japan, in Anecdota Oxoniensia, Aryan Series, t. I., part I., p. 9), and the Khumdan and Khumadan of Theophylactus. (See TOMASCHEK, in Wiener Z. M., t. III., p. 105; Marquart, Eranšahr, pp. 316-7; Osteuropäische und Ostasiatische Streifzüge, pp. 89-90.) (PELLIOT.)

XLI., p. 29 n. The vocabulary Hwei Hwei (Mahomedan) of the College of Interpreters at Peking transcribes King chao from the Persian Kin-chang, a name it gives to the Shen-si province. King chao was called Ngan-si fu in 1277. (DEVERIA, Epigraphie, p. 9.) Ken jan comes from Kin-chang King-chao Si-ngan fu.

=

=

Prof. Pelliot writes, Bul. Ecole franç. Ext. Orient, IV., July-Sept., 1904, p. 29: "Cette note de M. Cordier n'est pas exacte. Sous les Song, puis sous les Mongols jusqu'en 1277, Si-ngan fou fut appelé King-tchao fou. Le vocabulaire houeihouei ne transcrit pas 'King-tchao du persan kin-tchang,' mais, comme les Persans appelaient alors Si-ngan fou Kindjanfou (le Kenjanfu de Marco Polo), cette forme persane est à son tour

transcrite phonétiquement en chinois Kin-tchang fou, sans que les caractères choisis jouent là aucun rôle sémantîque; Kin-tchang fou n'existe pas dans la géographie chinoise. Quant à l'origine de la forme persane, il est possible, mais non par sûr, que ce soit King-tchao fou. La forme 'Quen-zan-fou,' qu'un écolier chinois du Chen Si fournit à M. von Richthofen comme le nom de Singan fou au temps des Yuan, doit avoir été fautivement recueillie. Il me parait impossible qu'un Chinois d'une province quelconque prononce zan le caractère tchao."

XLI., p. 29 n. A clause in the edict also orders the foreign bonzes of Ta T'sin and Mubupa (Christian and Mobed or Magian) to return to secular life.

Mubupa has no doubt been derived by the etymology mobed, but it is faulty; it should be Muhupa. (PELLIOT, Bul. Ecole franç. Ext. Orient, IV., July-Sept., 1904, p. 771.) Pelliot writes to me that there is now no doubt that it is derived from mu-lu hien and that it must be understood as the "[religion of] the Celestial God of the Magi."

XLIII., p. 32.

"The chien-tao, or 'pillar road,' mentioned, should be chan-tao, or 'scaffolding road.' The picture facing p. 50 shows how the shoring up or scaffolding is effected. The word chan is still in common use all over the Empire, and in 1267 Kúblái ordered this identical road ('Sz Ch'wan chan-tao') to be repaired. There are many such roads in Sz Ch'wan besides the original one from Han-chung-Fu." (E. H. PARKER, As. Quart. Rev., Jan., 1904, P. 144.)

XLIV., p. 36. SINDAFU (Ch'êng tu fu).-Through the midst of this great city runs a large river. . . It is a good half-mile wide. . . .

...

"It is probable that in the thirteenth century, when Marco Polo was on his travels, the 'great river a good half-mile wide,' flowing past Chengtu, was the principal stream; but in the present day that channel is insignificant in comparison to the one which passes by Ta Hsien, Yung-Chia Chong, and HsinChin Hsien. Of course, these channels are stopped up or opened as occasion requires. As a general rule, they follow such contour lines as will allow gravitation to conduct the water to levels as high as is possible, and when it is desired to raise it higher than it will naturally flow, chain-pumps and enormous