28 Field Museum of Natural History — Anth., Vol. XIII. 



mention the jewel in two passages,— first, as a product of the country 

 of Sogdiana (K'ang) corresponding to the region of Samarkand, and 

 secondly as a product of Persia (Pose, from Pars)} The text of the 

 Pei ski, with the same indications, is found also in the "Annals of the 

 Wei Dynasty" {Wei shu, Ch. 102, pp. 5a and 9b) .^ But this passage 



1 It is noticeable that si-se as products of Persia are mentioned in Pei shi and 

 Sui shu, but not in the two T'ang shu. The Kiu T'ang shu (Ch. 198, p. 11) enu- 

 merates as precious objects of Persia coral-trees, ch'i-k'ii, agate, and "fire-pearls" 

 (huo chu). The T'ang shu mentions only coral as a product of Persia and the gift to 

 China of a couch of agate (Chavannes, Documents, pp. 171, 174). The exact 

 history of the term ch'e-k'ti which in general denotes a large white conch (Tibetan 

 dung, Sanskrit gankha, Arabic shenek: M. Reinaud, Relation des voyages faits par 

 les Arabes, Vol. I, p. 6), and sometimes seems to refer to a precious stone remains to 

 be ascertained (compare Hirth and Rockhill, Chau Ju-kua, p. 231; Pelliot, 

 T'oung Pao, 1912, p. 481). The "fire-pearls" were lenses of rock-crystal, alleged to 

 have been used for producing fire (F. de Mely, Les lapidaires chinois, p. 60; Cha- 

 vannes, Documents, p. 166; Pelliot, Bulletin de V Ecole franqaise d' Extreme-Orient, 

 Vol. Ill, 1903, p. 270; Pen ts'ao kang mu, Ch. 8, p. 18 a). In the Sui shu, se-se are 

 enumerated together with genuine pearls, glass, amber, coral, lapis lazuli, agate, 

 rock-crystal, huo ts'i, and diamond. The name huo ts'i (the alleged identity with 

 huo chu remains to be proved) has not yet been properly identified. In the Nan shi 

 (Ch. 78, p. 7) these stones are mentioned as products of central India and described 

 as having the appearance of yiin-mu and the color of violet gold (Pelliot, /. c); the 

 difficulty is that also the word yiin-mu which according to Pelliot seems to designate 

 mica and mother-o'-pearl is not yet determined beyond doubt. Possibly, huo-ts'i 

 designates the garnet. The word se-se is, in the text of the Sui shu, followed by the 

 words hu lo kie lii t'eng. At first I was inclined to take the verb hu in its literal sense 

 "called, designated," and to believe that the words following it represent a gloss, 

 being the Persian or Arabic name of the stone in Chinese transcription. Recon- 

 structing the ancient sounds of those Chinese characters we would arrive at the read- 

 ing lok (or rok)-ket-li-dang; but there is no word in Persian or Arabic to be identified 

 with such a form. M. Paul Pelliot, to whom I submitted this difficult point, has 

 been good enough to write me that this passage had already attracted his attention, 

 and that he does not regard the incriminated words as a gloss; he thinks that the word 

 hu is also part of the transcription, and that two further products are enumerated in 

 their Persian names. The passage, accordingly, should be understood in the sense 

 that Persia produces se-se, hu-lo{k), and ket-li-dang. The two latter names, "however, 

 are as yet unidentified, but with M. Pelliot's very plausible point of view, a better 

 attempt at identification might be pursued. Indeed, Prof. A. V. Williams Jackson 

 had called my attention to the fact that katlidang may be a compound of the word 

 qatldn, "link" or "scale," used alike in Arabic, Turkish and Persian, and the Persian 

 word tan "body," the content of the term implying scale or chain armor. This is 

 very suggestive, as indeed Persia was the country which supplied China with chain- 

 mail (ancient specimens in the Field Museum). The T'ang shu, in the account on 

 Samarkand (K'ang) states that in the beginning of the period K'ai-yiian (713-741) 

 Samarkand sent as tribute to China chain-mail {so-tse k'ai). This question will be 

 shortly discussed by me in another place. Liang shu (Ch. 54, p. 14 b) attributes to 

 Persia coral-trees one to two feet high, amber, agate, genuine pearls, and mei-hui. 

 Hirth and Rockhill (Chau Ju-kua, p. 16, St. Petersburg, 1912), treating the prod- 

 ucts of Persia after Wei shu and Sui shu, entirely omit the se-se (and several others). 

 It seems doubtful if, as stated so positively by the two authors, "most of these prod- 

 ucts came, of course, from India, or from countries of south-eastern Asia, only a few 

 being products of Arabia, or countries bordering on the Persian Gulf" (and again on 

 p. 7). This is true only to a certain extent; the se-se, at any rate, are not mentioned 

 by the Chinese as products of India or south-eastern Asia, but exclusively as products 

 of Persia and Sogdiana, to which, later in the T'ang period, Fu-lin, Tashkend, Tibet, 

 and the Man are joined. 



^ The Wei dynasty ruled from 386 to 556; the Wei shu was written by Wei Shou 

 (506-572; Giles, /. c, p. 867) and presented to the throne in 554. 



