¹ The texts were the Kuang chou chi of Ku Wei , the Kuang chih of Kuo I-kung , the Nan chou chi of Hsü Piao , and the Ming i pieh lu of T'ao Hung-ching . The fragments discussed in this study were preserved in two materia medica: (i) the Ch'ung hsiu chêng ho ching shih cheng lei pei yung pên ts'ao , the Yüan version of T'ang Shên-wei , Ching shih chêng lei pei chi pên ts'ao , Jên min wei shêng ch'u pan shê edition, 1957, and referred to as the CLPT; (ii) the Pên ts'ao kang mu of Li Shih-chên first published in 1596, Jên min wei shêng ch'u pan shê edition, 1957, and referred to as the PTKM. Of these four texts only the identity and dates of the author of the Ming i pieh lu are known beyond dispute; he was the famous Taoist T'ao Hung-ching who lived from about 452 to 536. Nothing is known of the Nan chou chi. Laufer stated that it was quoted in the Ch'i min yao shu by Chia Szῠ-hsieh , who lived between 386 and 534; Laufer, B., Sino-Iranica: Chinese contributions to the history of civilisation in ancient Iran with special reference to the history of cultivated plants and products (Field Museum of Natural History, Publication No. 201), Chicago, 1919Google Scholar (henceforth referred to as Sino-Iranica), 247, n. 7. Laufer considered that it was written before the fifth century a.d.; ibid., 460. The Kuang chih was quoted by Li Tao-yüan in the Shui ching chu , Ssῠ pu ts'ung k'an edition, 2, 1b; Li Tao-yüan died in 527. Fragments were collected in the Yü han shan fang chi i shu , Changsha edition, 1883, and the text was attributed to the Tsin period. It was listed in the Sui shu bibliography, Ssῠ pu ts'ung k'an edition, 34, 7b. The Kuang chou chi of Ku Wei was listed among the sources of the Chêng lei pên ts'ao. A work of this name was attributed to Ku Hui of the Tsin period in Wên T'ing-shih , Pu Tsin shu i wên chih , reprint of the K'ai ming shu tien edition of the Êrh shih wu shih pu pien , 1959, 3, 38. The characters for ‘Wei’ and ‘Hui’ are similar, and Wên T'ing-shih could have made a mistake. That he was capable of inaccuracy is shown in the same passage where he quoted the T'ang shu tsai hsiang shih hsi piao Ku Hui was the younger brother of Ku Yung , a Tsin minister. According to the Hsin T'ang shu, Ssῠ pu pei yao, 74 , 10a, Ku Hui was two generations senior to Ku Yung and the brother of a Wu minister. In Ku Yung's biography in the Tsin shu, Ssῠ pu ts'ung k'an edition, 68, la, Ku Yung was described as a minister during the early fourth century. It was also said that ‘Ku’ was a famous surname in the south; to that extent Ku Hui's composition of the Kuang chou chi or ‘Record of Canton’ is reasonable. In the present study it will be suggested that references to ‘Po-Ssῠ resins’ in fourth century texts are explicable, and there is no reason to upset the tradition that the Kuang chou chi and the Kuang chih were written in Tsin times. The passages of the four texts under discussion were quoted by Li Hsün in his Hai yao pên ts'ao which is also lost. According to Li Shih-chên Li Hsün was alive in the 756–63 period. Chang Hsing-lang has suggested that Li Hsün's brother, Li Hsien , wrote this work in the tenth century and that the family was of Persian extraction; Chung hsi chiao t'ung shih liao huai pien , Peiping, 1930, 4, …, 99–100. Dr. Ch'ên Pang-hsien , in his Chung kuo i hsüeh shih , 1957 edition, 146, and Drs. Huard and Wong in their ‘Bio-bibliographie de la médicine chinoise’, Bulletin de la Societé des Études Indochinoises, xxxi, 3, 1956, serial number 262, have accepted Li Shih-chên's view.

¹ The benzoin of Laos is from Styrax benzoides Craib and S. tonkinense Craib. Laos does not produce tree camphor. Borneo produces tree camphor but no species of Styrax with commercial value. The Malay Peninsula produces both but the trade in them has never been important. Western Java has a few benzoin trees but only the fossils of camphor trees.

² In Western literature the attribution was as early as the Specimen medicinae Sinicae, Frankfurt, 1682Google Scholar, serial numbers 210, 211, edited by Andrew Cleyer and believed to be based on the work of Michael Boym, S. J. There is a copy in the Library of the Royal Botanic Gardens, Kew. The transcriptions were jù hia and mŏ yŏ. For modern works see Read, B. E. (with Liu Ju-ch'ang), Chinese medicinal plants from the Pên ts'ao kang mu, a.d. 1596 …, a Botanical, chemical, and pharmacological reference list (Peking Natural History Bulletin), 1936, serial numbers 313, 340Google Scholar; Roi, J., ‘Traité des plantes médicinales chinoises’, Encyclopédie biologique, XLVII, Paris, 1955, 202, 208–9.Google Scholar

³ The Ling wai tai ta of Chou Ch'ü-fei , 1178, Pi chi hsi shou ta kuan collection, Chin pu shu chü, Shanghai, undated, 3, 3a-b, and the Chu fan chih of Chao Ju-kua , 1225, Chung hua shu chü, Peking, 1956, 74.

⁴ Bretschneider, E., On the knowledge possessed by the ancient Chinese of the Arabs and Arabian colonies and other western countries, mentioned in Chinese books, London, 1871, 16Google Scholar, n. He translated Po-Ssῠ ju and An-hsi as ‘Persian’; Botanicon Sinicum, Part III, 1895, 460, 466. Many years later Moens associated himself with the colonies theory; ‘De Noord-Sumatraanse rijken der parfums en specerijen in Voor-Moslimse tijd’, Tijdschrift (Madjalah), LXXXV, 4, 1955–1957, 353–5.Google Scholar

⁵ The first to express this view seems to have been Phillips, G., ‘Notes on Sumatra and the Po-szu’, Notes and Queries on China and Japan, III, 6, 1869, 90–2.Google Scholar His texts reflected the surmise in the Ming shih that Su-mên-ta-la = Samudra = Pasai might have been the Po-Ssῠ and Ta-shih countries of T'ang times. The ‘Pasai’ theory was specifically stated by Tsuboi Kumazo in his study of Ju-kua, Chao, Actes du douzième Congrès International des Orientalistes, II, 1899, 92, 121–2.Google Scholar

⁶ The existence of a Po-Ssῠ country apparently on the borders of Burma was noted by Parker, E. H., Burma with special reference to her relations with China, Rangoon, 1893, 14.Google Scholar He quoted from the Hsin T'ang shu account of the P'iao which described a situation at the beginning of the ninth century. Later Laufer and Ferrand thought that there was a Po-Ssῠ in Burma as well as in Sumatra. Ferrand suggested that in Burma it referred to ‘Bassein’, but Laufer did not commit himself (Sino-Iranica, 474Google Scholar); G. Ferrand's review of Sino-Iranica in the Journal Asiatique (henceforth referred to as JA), XI^e Sér., XVIII, 1921, 286.Google Scholar

¹ Chavannes and Takakusu, in their translations of I Tsing's works, assumed that in 671 I Tsing's ship from Canton to Fo-shih , later identified with Ṥrīvijaya, was Persian. Hirth and Bockhill developed the middlemen theory in their introduction to Chau Ju-kua. His work on the Chinese and Arab trade in the twelfth and thirteenth centuries …, St. Petersburg, 1911, 7–8.Google Scholar They noted, however, the reference to the ‘Po-Ssῠ’ country in the Ling wai tai ta and the Chu fan chih and thought that it was a Negrito country in or near the Malay Peninsula; ibid., 152, n.

² Sino-Iranica, 468–87.Google Scholar

³ In his review of Sino-Iranica. He had previously accepted the theory of Hirth and Rockhill.

⁴ A history of Persian navigation, 1928, 81–3, 99–103.Google Scholar

⁵ , Chung hsi chiao t'ung shih Kao huai pien , Peiping, 1930, , 4, 188.

⁶ ‘Les traditions manichéennes au Fou-kien’, T'oung Pao (henceforth referred to as TP), xxii, 1923, 196, n. 3.Google Scholar

⁷ Relation de la Chine et de l'Inde, 1948, xxxvi.Google Scholar

⁸ In an unpublished thesis entitled The Nanhai trade, 1954Google Scholar, presented to the University of Malaya for the degree of Master of Arts. It is understood that this thesis has now been published in the Journal of the Malayan Branch, Boyal Asiatic Society.

¹ The Chinese attitude towards the pine is discussed by Stein, R., ‘Jardins en miniature d'Extrême-Orient’, Bulletin de l'École Française d'Extrême-Orient, XLII, 1942, 83–4.Google Scholar

² Li Shih-chên's account of the pine included several quotations from Taoiste which show clearly the value the latter attached to it; PTKM, 34, 1351–4. On techniques for resisting the decay caused by the ‘Five Cereals’ see Maspero, H., ‘Le taoïsme’, Mélanges posthumes …, II, 1950, 100–2.Google Scholar Maspero discussed the subject at much greater length in ‘Les procédés de “nourrir le principe vital” dans la religion taoïste ancienne’, JA, 1937, 2 and 3, 177–252, 313–430.Google Scholar

³ Chung kuo ku tien i hsüeh ts'ung han edition, 1955, 28. On the date of this work see Roi, J. and Yun-joei, Ou, ‘Le taoïsme de les plantes d'immortalité’, Bulletin de l'Université l'Aurore, Sér. III, Tom. II, 4, 1941, 539Google Scholar, and Huard, P. and Wong, M., ‘Evolution de la matière médicale chinoise’, Janus, XLVII, 1958, 7.Google Scholar

⁴ PTKM, 34, 1352.

⁵ The Shên nung pên ts'ao ching, 28Google Scholar, said of it: … . Ta Ming , in his Jih hua pên ts'ao of the tenth century, was more explicit: OLPT, 12, 291.

⁶ Fu ling has been identified as Pachyma cocoa Fries.

¹ Maspero, H., ‘Influences occidentales en Chine avant les Han’, Mélanges posthumes …, III, 1950, 37–51.Google Scholar

² Huai Nan tzῠ mentioned an immortality drug supplied by the Queen Mother of the West and the many drugs in the K'un lun region; T'ai p'ing yü lan, Ssῠ pu ts'ung k'an edition, third series (henceforth referred to as TPYL), 984, 4b. The Han wu ku shih attributed Tou-mo perfume to her; it was given as tribute by a central Asian country and effected a miraculous cure (TPYL, 983, 6b–7a). The Queen Mother had once been regarded as the goddess of epidemics but, for the Taoists, she became the Queen of the Immortals and the purveyor of recipes for long life and other drugs; Maspero, ‘Le taoïsme’, 127. Among the famous perfumes which reached China at an early date was storax from the Middle East, traded by central Asian middlemen.

³ The identification of the shên perfume with fan hun was contained in the Yüeh shih shih chê chuan , quoted by the Ch'ing encyclopaedia Ko chih ching yüan (henceforth referred to as KCCY), 57, 24b. The date of the epidemic was given as 88 b.c. Similar epidemics seem to have broken out in 99 b.c. and 90 b.c. The epidemics are interesting partly as early examples of the powerful fumigatory properties which the Chinese came to associate with foreign resins and partly because shên, as will be noted below, had a connexion with the origins of the South East Asian resin trade.

⁴ In his review of Hirth, and Rockhill's, Chau Ju-kua, TP, XIII, 1912, 475–9.Google Scholar

⁵ In addition to being described in the Nan fang ts'ao mu chuang , it seems to have been mentioned in the Nan chou i unt chih of Wan Chên , an official of the Wu dynasty (222–80). TPYL, 982, 2a. The Wei lüeh also mentioned it. San kuo chih , Ssῠ pu ts'ung k'an edition, 30, 33b. Of some interest is the statement in the T'ang tzῠ that it came from Shih-tzῠ . TPYL, 982, 2b. The author was the Taoist T'ang P'ang who lived in Wu times; Sui shu, Ssῠ pu ts'ung k'an edition, 34, 4a. Shih-tzῠ was the name by which Ceylon was known in the Liu Sung shu which covered the 420–76 period. If T'ang P'ang meant ‘Ceylon’, it would lessen the likelihood that Ssῠ-tiao , mentioned in the Nan chou i wu chih, meant ‘Ceylon’ in the third century. The writer is preparing a study of the early historical geography of western Indonesia and hopes to return to this matter.

¹ See p. 327, n. 5, on T'lang P'ang.

² PTKM, 34, 1372. He attributed it to an unknown country in the ‘Southern Ocean’.

³ ; PTKM, 34, 1372.

⁴ According to T'ao Hung-ching it was used for ; PTKM, 34, 1371. According to the T'ang pên ts'ao it cured ; CLPT, 12, 309.Google Scholar

⁵ T'ao Hung-ching's expression was: ; PTKM, 34, 1371–2. Ta Ming said of pine resin: ; PTKM, 34, 1352.Google Scholar

⁶ Shang mo yin shu kuan edition, 1955, , 7. The sentence was as follows: ; Hirth translated as ‘straight like’. China and the Roman Orient, 1885, 268, n. 1.Google Scholar

⁷ TPYL, 982, 2a. The peach was the fruit of immortality and grew in the garden of the Queen Mother of the West.

⁸ TPYL; 982, 2a, quoting the Tien shu . According to the Sui shu it was written by Chien P'ing-wang in the Liu Sung period (420–76). T'ao Hung-ching was born about 452.

¹ According to Su Sung: CLPT, 12, 291.Google Scholar

² ; PTKM, 34, 1371.Google Scholar

³ According to the text used by Hirth, and Bockhill, , Chau Ju-kua …, 195.Google Scholar In Feng Ch'êng-chün's edition, Chung hua shu chü, 1956, 93, the tree was said to be like the yung , translated in the Tz'ῠu hai as Ficus wightiana.

⁴ CLPT, 12, 309Google Scholar; PTKM, 34, 1371.Google Scholar Laufer omitted t'ou when he mentioned this resin; Sino-Iranica, 470.Google Scholar The Kuang chih's attribution of hsün-lu to Ta-ch'in is contained in TPYL, 982, 2a, under the irregular transcription of . The Yü han shan fang chi i shu confined its quotation to the Kuang chih passage in the TPYL, rendering hsün-lu as .

⁵ Facsimile edition published for the Peking National Library, Ta tung shu chu, Peking, 1936, , 44, 2a–b.

⁶ ; PTKM, 34, 1371.Google Scholar He was the author of the Pên ts'ao shih i .

⁷ ibid.

¹ T'ao Hung-ching described the properties of hsün-lu and concluded: ; ibid. Bretschneider misunderstood T'ao Hung-ching to refer to one drug; Botanicon Sinicum, Part III, 1895, 460.Google Scholar

² Wei shu, Ssῠ pu ts'ung k'an edition, 102, 15b.Google Scholar

³ The best grade was ‘transparent’. One grade was ‘chopped and pared ’ and was composed of pieces of inferior quality. The worst grade was dust swept off the floor; PTKM, 34, 1371.Google Scholar

⁴ For the Ṥrīvijayan embassy see Sung hui y ao kao, , 7, 21a–b. The list of goods covered by imperial control is contained in SHYK, , 44, 2a–b.

⁵ See p. 329, n. 1.

⁶ CLPT, 12, 292Google Scholar, quoting the T'ang Ping pu show chi as follows: .

⁷ The pên ts'ao literature from the seventh century attached importance to expulsive agents. This may reflect a new emphasis in Chinese medicine as a consequence of Indian medical influences arriving with Buddhist scholars during the period between the fall of the Han and the T'ang dynasty. In the same period Taoists were experimenting more freely with vegetal drugs for non-magical purposes. As a result of these developments Chinese pine resin may have been relegated to the status of a cure for skin infections, while hsün-lu and ju, originally known as ‘pine resins’ and recommending themselves initially for longevity techniques, became specially popular as expulsive agents in a new context of medical ideas. Dr. Needham has commented on the need for more study of the influence of Indian medicine on China. A related study would be the extent to which an Indo-Indonesian corpus of medical experience developed in connexion with Indonesian resins and other drugs and reached China with the products. The present study is only concerned with the manner in which a trade in hsün-lu produced a supplementary trade in ju; hsün-lu seems originally to have recommended itself as a form of ‘pine resin’ at a time when very early Taoist ideas about pine resin still dominated Chinese medical practice. For the development of Chinese medicine in this period see Roi, J. and Yun-joei, Ou, ‘Le taoïsme et les plantes d'immortalité’, Bulletin de l'Université l'Aurore, Sér. III, Tom. II, 4, 1941, 535–46Google Scholar; Pang-hsien, Ch'ên, Chung kuo i hsüeh shih, 1957 edition, 148–54.Google Scholar Dr. Needham's views are contained in Science and civilisation in China, I, 1954, 208–13.Google Scholar

¹ Students of Chinese botany working in China in modern times relied chiefly on the analysis of specimens bought at the local drug shops for identification of the Chinese names. They were therefore the victims of chance purchases. Tatarinov in 1851 sent ju to Horaninov and it turned out to be sandarac. Bretschneider sent it to Flückiger and it was found to be Boswellia. Dr. Bead and Father Roi have identified it as Pistacia ierebinthus and Pistacia lentiscus respectively.

² China and the Roman Orient, 266–8.Google Scholar In Hirth, and Rockhill's, Chau Ju-kua, 196Google Scholar, n. 1, hsün-lu was derived from the Arabic kundur.

³ TP, XIII, 1912, 475–9.Google Scholar He pointed out that the early sound of lu was *luk. This seems to be confirmed by the variant character lu embodied in the name in the Pei shih, Ssü pu ts'ung k'an edition, 97, 16b, and in the Kuang chih's reference in TPYL, 982, 2a.

⁴ Sino-Iranica, 470–1.Google Scholar

⁵ Documentary evidence of hsün-lu in South East Asia is limited to the statement of the Kuang chih that it came from Tongking; TPYL, 982, 2a. The trade has to be reconstructed from indirect evidence discussed below in connexion with the Po-ssῠ. In 428 Kaviri in India sent it as tribute to China; Liu Sung shu, Ssῠ pu ts'ung k'an edition, 97, 11b. It was called ma-lo which was one of the alternative names given by Li Shih-chên. Pelliot ignored it because he was uncertain when it was first en vogue. TP, XIII, 1912, 477.Google Scholar

¹ For Boswellia see Carter, H. J., ‘A description of the frankincense tree of Arabia, with remarks on the misplacement of the “Libanophorous region” in Ptolemy's Geography’, Journal of the Bombay Branch of the Royal Asiatic Society, II, 1848, 380–90Google Scholar; Howes, F. N., Vegetable gums and resins, Waltham, Mass., 1949Google Scholar; Beek, Gus W. van, ‘Frankincense and myrrh in ancient South Arabia’, Journal of the American Oriental Society, LXXVIII, 3, 1958, 141–51.CrossRefGoogle Scholar The writer of the present study had the benefit of discussing botanical aspects of the subject with Dr. Howes.

² Yule, H., The book of Ser Marco Polo, 3rd edition, 1926, II, 449Google Scholar, where the editor noted from Bretschneider's translation that the Chinese had also compared the ju tree with the pine.

³ It comes mainly from Boswellia carteri and B. frereana. The T'ang pên ts'ao referred to hsün-lu in India which would presumably be Boswellia serrata of northern and central India.

⁴ The Periplus of the Erythraean Sea, translated from the Greek and annotated by Schoff, W. H., 1912, 38.Google Scholar

⁵ Hou Han shu, Ssῠ pu ts'ung k'an edition, 88, 14a.

⁶ Flückiger, F. A. and Hanbury, D., Pharmoeographia, second edition, 1879, 608.Google Scholar

⁷ According to Pliny the ground-pine was known to some as tus terrae. ‘False frankincense’ is scraped off the trunk of Pinus palustris where it gathers spontaneously. Li Shih-chên noted that ju flowed naturally from the tree as well as being tapped.

¹ On the Sumatran pine see Heyne, K., De nuttige planten van Nederlandsen, Indië, second edition, Buitenzorg, 1927, I, 118–20Google Scholar, and Howes, F. N., Vegetable gums …, 111–13.Google Scholar The Atjeh hinterland seems to be well-endowed with pine forests.

² Bor, N. N., Manual of Indian forest botany, 1953, 14.Google Scholar

³ In the Achinese area, the base for the Dutch-sponsored pine industry, the local name for the pine is sala. Ṥala-veṣṭa has been used in Sanskrit for ‘pine resin’, though Ṥāla is the usual name for Shorea. The various species of Shorea in Indonesia are known by indigenous names. There is a slender possibility that the Achinese pine was given its name as a result of an early Indian interest in it.

⁴ The Ming travellers did not mention it and the Arabs, close to the source of genuine frankincense, naturally ignored it. Tomé Pires mentioned ‘pitch’ from the Batak country, Aru, and Kampar but also from the southern end of the island where no pines grow. It probably came from a non-fragrant dammar-producing Shorea.

¹ About 12 species of Commiphora are found in Arabia and about five in India. The famous Commiphora mukul, sometimes called Balsamendron mukul, is found in the dry zone of Sind, Kāthiāwār, Rājputana, Berar, Khandesh, and also Mysore. Watt, G., The commercial products of India, 1908, 400Google Scholar; Howes, F. N., Vegetable gums…, 153.Google Scholar

² The Periplus stated that bdelliums were exported from Barygaza and Barbaricum. On this subject see Warmington, E. H., The commerce between the Roman Empire and India, 1928, 201.Google Scholar Bdelliums or ‘false myrrhs’ are believed to come from species of Commiphora; Howes, F. N., Vegetable gums …, 153.Google Scholar

³ For example, Stuart, G. A., Chinese materia medica, 1911, 61.2.Google Scholar Father Roi found it difficult to determine the origin of the Commiphora producing the myrrh sold on the Chinese market. ‘Traité des plantes …’, Encyc. biol., XLVII, 202.Google Scholar

⁴ That it was a myrrh is suggested by a fragment of the Lien hua fo ching sutra preserved in the Ko chih ching yüan which stated that mo was called in Sanskrit kan-ta which suggests itself as a transliteration of gandha or gandhārasa = Balsamendron mukul. A specific resin rather than ‘perfume’ must have been intended. KCCY, 57, 26a.

⁵ Pei shih, Ssῠ pu ts'ung k'an edition, 97, 29b; T'ai p'ing huan yü chi , 1754 edition, 182, 11b.

¹ CLPT, 13, 330.Google Scholar Also see PTKM, 34, 1373.Google Scholar

² Sino-Iranica, 462, 480.Google Scholar Bretschneider did not mention it in Botanicon Sinicum, 1895.Google Scholar

³ ; PTKM, 34, 1373.Google Scholar The comparison may have been made by Hsü Piao, author of the Nan chou chi; the text is ambiguous.

⁴ ibid.

⁵ Chinese medicinal plants, serial number 336. ‘Traité des plantes’, 208.Google ScholarPistacia nuts, under the name of wu ming tῠ , were mentioned by the Nan chou chi which stated that the Po-ssῠ chia called them A-yüeh-hun-tzῠ ; PTKM, 30, 1294.Google Scholar Laufer showed this to be a likely transliteration of an Old Iranian word; Sino-Iranica, 248–53.Google Scholar Species of Pistacia grow in Persia and in this instance Po-ssῠ must mean ‘Persian’. On Pistacia in Persia see Hooper, D. and Field, H., Useful plants and drugs of Iran and Iraq (Field Museum of Natural History, Publication No. 387), Chicago, 1937, 153–4.Google Scholar

⁶ Natural history, Book XXIV, XX.

⁷ According to Ma Chih and to Su Sung . Li Hsün said that it was ‘reddish-black ’; PTKM, 34, 1373.Google Scholar

⁸ In I Tsing's translation of the Golden Sutra. Tripitaka, daizōkyō, Kokuyaku, 1919–1928, vol. 26Google Scholar, , 55. Guggula is identified with Balsamendron mukul in Avinash Chandra Kaviratna's translation of the Caraka Samhitā. Also see Pelliot, , TP, XIII, 1912, 480Google Scholar, and Laufer, , ‘Bird divination among the Tibetansy’, TP, xv, 1914, 5, n. 1.Google Scholar

⁹ See p. 347 below.

¹ For differences of opinion on the subject see Hasan, Hādī, A history of Persian navigation, 65.Google Scholar The Periplus stated that this coast contained nothing except bdelliums.

² Sui shu, Ssü pu ts'ung k'an edition, 83, 15b. TPYL, 982, 1b, quoting the ‘T'ang shu’. Wei shu, Ssῠ pu ts'ung k'an edition, 102, 10b. PTKM, 34, 1373.Google Scholar The T'ai p'ing huan yü chi, 182, 11b, attributed it to Kashmir.

³ CLPT, 13, 330.Google Scholar Laufer did not quote this passage in his Sino-Iranica.

⁴ On Sumatran benzoin see Heyne, K., De nuttige planten, II, 1256–62.Google Scholar Marsden described the location of benzoin as almost exclusively in the Batak country. On benzoin from mainland South East Asia see Crevost, Ch., ‘Le benjoin de l'Indochine’, Bulletin Économique de l'Indochine, 146, 1921, 286–7.Google Scholar

⁵ Burkill, I. H., A dictionary of the economic products of the Malay Peninsula, 2 vols., London, 1935, II, 2102.Google Scholar He accepted the view of Flückiger and Hanbury, Pharmacographia, second edition, 1879, 404–5. Laufer doubted whether the description of the tree contained in the Yu yang tsa tsu of about a.d. 800 corresponded to Styrax benzoin because the tree it described was too large; Sino-Iranica, 466.Google Scholar Mr. Burkill disagreed and described Styrax benzoin as a ‘fairly tall tree’. The Ta Ming i t'ung chih and Garcia d'Orta also stated that it was a large tree.

⁶ Vuuren, L. van, ‘De handel van Baroes, als oudste haven op Sumatra's westkust, verklaard: en voor de toekomst beschouwd’, Tijdschrift van Kon. Nederlandsch Aardrijkskundig Gen., second series, XXV, 6, 1908, 1389–1402.Google Scholar One reason given by Mr. Burkill for the later appearance of benzoin in trade was the difficulty of extraction. This applies equally to camphor; magic is a part of both processes. These processes are wasteful in trees, and in early times the distribution of benzoin and camphor in northern Sumatra may have been more widespread. One of the features common to Ming, Portuguese, and early nineteenth century accounts of Sumatra was the way benzoin and camphor found markets on the east as well as on the west coast. In the nineteenth century camphor was extracted in the Siak area, and Heyne has noted camphor trees on the east coast. Joustra suggested that benzoin trees had become rare in Karoland in the hinterland of the north-east coast; Batakspiegel, Leiden, 1926, 53.Google Scholar

¹ , PTFTKM, 34, 1375. Li Shih-chên explained its name incorrectly by saying that this perfume ‘warded off all evil and quelled (an hsi ) all harmful things . But he also noted that An-hsi was said to be the name of a country.

² Su Kung's observation of the seventh century is contained in PTKM, 34, 1375.Google Scholar

³ ibid., 1373.

⁴ A dictionary …, I, 1065–7.

⁵ ibid., II, 2102, 2106. Sumatran grades of benzoin vary more than the mainland ones.

⁶ The Suma Oriental, edited by Armando Cortesāo, Hakluyt Society, 1944, I, 135–65, 226–7.

Garcia d‘Orta and Marsden also noticed the same two colours.

⁷ Ying yai sheng lan , edited by Feng Ch‘êng-chün, Chung hua shu chü edition, 1955, 17, 27. Ma Huan made it clear that there were degrees of whiteness and blackness.

⁸ Journal of the Straits Branch of the Royal Asiatic Society, xxx, 1897, 306–7Google Scholar; Journal of the Royal Asiatic Society, 1913, 1, 168.Google Scholar Mr. Burkill should not have derived kĕmĕnyan from ‘An-hsi hsiang’ which means ‘Parthian perfume’.

⁹ Chu fan chih, 96–7, 99–100.Google Scholar An earlier reference to kĕmĕnyan seems to be contained in a fragment preserved in the Ko chih ching yüan from the Wu tsa tsu in which it was stated that chin-yen ‘snow’ perfume was burnt in the palace in the 1119–26 period. KCCY, 57, 24a.Google Scholar

¹ Burkill, I. H., A dictionary of the economic products …, II, 2102.Google Scholar Flückiger and Hanbury, assuming that Arab and Persian traders carried South East Asian products to China between the tenth and thirteenth centuries, could not find benzoin among them; Pharmacographia, 404.Google Scholar This is incorrect. An-hsi appears among the imports mentioned in the Sung hui yao kao at the end of the tenth century; Sung hui yao kao, , 44, 2b.

² CLPT, 13, 321.Google Scholar

³ Cunninghamia sinensis which belongs to the Pinaceae family.

⁴ PTKM, 34, 1376.Google Scholar

¹ Yu yang tsa tsu, Ssŭ pu ts‘ung k‘an edition, 18, 7a.

² Liang shu, Ssŭ pu ts‘ung k‘an edition, 54, 18a, as a product of Lang-ya-hsiu or Langkasuka, probably in the Patani region and a minor source of camphor trees. Wheatley, P., ‘Langkasuka’,. TP, xuv, 4–5, 1956, 387–12.Google Scholar The Liang shu dealt with the 502–56 period and was compiled before 637. In the Indonesian languages neither camphor nor benzoin have Sanskritic names which suggests that they did not enter trade under the auspices of Indian traders.

³ CLPT, 13, 321Google Scholar; PTKM, 34, 1377.Google Scholar According to the T‘ang pên ts‘ao it was beneficial for

⁴ For cummin see PTKM, 26, 1204.Google Scholar The Hsin T‘ang shu, reference to camphor is in 222, 5a. Ssŭ pu ts‘ung k‘an edition.

¹ P‘o-lu = ‘Barus’, according to Li Hsün, sent an embassy in the 627–50 period; CLPT, 13, 321.Google Scholar There is no other record of an embassy from a state transliterated in this way.

² An explanation of these embassies was given in the T‘ai p‘ing huan yü chi, 182, lOb, written in the tenth century. They were said to be a pretext for trading: .

³ Blagden made this point; ‘Some remarks on Chau Ju-kua's Chu fan chi’, JRAS, 1913, 1, 168.Google Scholar He was commenting on the possibility that Pasai was known under that name in 1178 which is at least 600 years later than the references discussed in this study.

⁴ Yüan shih , Ssŭ pu ts‘ung k‘an edition, 210, 18a. Marco Polo knew it under a similar name.

⁵ Gerini, G. E., Researches on Ptolemy's Geography, 1909, 682.Google Scholar

⁶ Li Hsün quoted the Kuang chou chi that sulphur was produced in the K‘un-lun and Po-ssŭ countries; PTKM, 11, 702.Google Scholar It may be noted that Ma Huan said that sulphur was found in Su-měn-ta-la = Pasai and that Tomé Pires said that the Pasai and Pedir sulphur came from the islands off Atjeh. Another example of the Po-ssŭ—K‘un-lun antithesis is provided by the Yu yang tsa tsu, 18, 8a, which stated that the Po-ssŭ envoys admitted that the K‘un-lun lac insect gum was of a better quality than that of the Po-ssŭ. The K‘un-lun country in question was Chên-la = Cambodia but the identity of the Po-ssŭ country was not disclosed; it was only said that the envoys were Po-ssŭ. The view that the Po-ssŭ were a people persisted in T‘ang times, for the Nan i chih referred to the P‘o-lo-měn, Po-ssŭ, Cho-p‘o, P‘o-ni, K‘un-lun, and other ‘foreign tribes who traded with Nan Chao; TPYL, 981, 5a. This seems to be a list of trading peoples in which Cho-p‘o and P‘o-ni meant ‘Javanese’ and ‘Borneo people’.

¹ ; CLPT, 14, 358.Google Scholar Laufer translated this as ‘western ocean and Po-ssŭ’; Sino-Iranica, 482.Google Scholar The reading suggested above is consistent with T‘ao Hung-ching's description, noted below, of the Lu country, the source of camphor, as the ‘western ocean country’. In that case there could be no question of two places. On the marking nut also see PTKM, 35Google Scholar, , 1411. It comes from Semecarpus anacardium Linn. f.

² The Hou Han shu described An-hsi = Parthia in terms of the direction from Loyang in northern China. Po-ssŭ = Persia was first mentioned in the Wei shu and was classified among the countries in the ‘western regions’. In the Liang shu it was a ‘north-western Jung country’ in the ‘western regions’.

³ ; CLPT, 9, 232.Google Scholar Li Shih-chên quoted Li Hsiin that it grew in ‘the western ocean and western Jung and Po-ssŭ countries’; PTKM, 14, 812.Google Scholar Laufer suspected that Li Hsün in this instance confused the two Po-ssŭ and was erroneously thinking of Persia; Sino-Iranica, 481–2.Google Scholar It is not certain, however, that it was Li Hsün who introduced the word ‘Po-ssŭ’.

⁴ PTKM, 14, 828.Google Scholar In the CLPT, 8, 214Google Scholar, ‘P‘iao’ is omitted and only the ‘western ocean’ mentioned, but T‘ang Shên-wei mentioned P‘iao at the beginning of the section as a source for the drug and quoted the Kuang chih in support. Dr. Read very tentatively identified it as Platycarya strobilacea.

⁵ CLPT, 13, 321Google Scholar; PTKM, 34, 1377.Google Scholar

⁶ The writer, after examining sailing conditions in the Straits and taking into account Chia Tan's sailing times, finds it impossible to believe that Chia Tan's P‘o-lu was Barus on the west coast.

¹ ying yai sheng lan, 27.Google Scholar

² ibid., 33. The ninth century compiler of the Relation de la Chine et de l'lnde said that Rᾱmnῑ, a general expression for northern Sumatra, was watered by the seas of Harkand and Salᾱhit or the Indian Ocean and the Straits.

³ TPYL, 790, 7b, quoting the Nan chou i wu chih written by Wan Chên who was an official of the Wu dynasty (220–80).

⁴ Also written as

⁵ Wheatley, P., ‘The earliest Chinese accounts of the Malay Peninsula’, Journal of the Malayan Branch of the Royal Asiatic Society, XXVIII, 1, 1955, 1–23.Google Scholar Dr. Wheatley suggested that it might be in the Kuantan area at the south of the Peninsula, but in 1958 he revised this view and located it somewhere in the ‘isthmian tract’; The historical geography of the Malay Peninsula before a.d. 1500, a thesis presented to the University of London for the degree of Doctor of Philosophy, 1958, 59 and n. 1.

¹ Dr. Wheatley did not regard it as a Peninsular state. Pelliot, who seems to have been the latest to take an interest in Ko-ying, thought, for reasons which he did not disclose, that it was in the southern part of the Malay Peninsula; Études asiatiques, 1925, II, 250.Google Scholar Apart from being ‘south’ of Chü-chih which does not suggest that it was in India, Ko-ying was only one month's journey away. Funan on the Mekong delta was said to be one month's journey from Tun-sun at the head of the Peninsula or, alternatively, more than 3,000 li. This supplies a hint of the distance involved between Chü-chih and Ko-ying.

² TPYL, 359, 1b.Google Scholar The traders were Indo-Scythian horse-dealers.

³ Lo yang chia lan chi , K‘o hsüeh ch‘u pan she edition, 1958, 89. The details about Ko-ying in this work were probably borrowed from the Nan chou i wu chih. For similar borrowings see Maspero, H., ‘Un texte taoïste sur 1'Orient Romain’, Mélanges posthumes …, III, 1950, 99.Google Scholar

⁴ The occasional voyage may have been made through the Straits and up the east coast of the Peninsula to join the main international trade route through the northern Peninsula and Funan, but if the volume had been great Ko-ying would have sent embassies to China. A definition of the Straits route to China must include the voyage across the South China Sea with land-fall considerably to the east of Funan. This was the short cut which later made the route worth-while, and there is no evidence that it was in use in the third century.

⁵ P‘o-huang, also known under other slightly different transliterations, sent embassies in 442, 449, 451, 455, 456, 459, 464, and 466. Liu Sung shu, Ssŭ pu ts‘ung k‘an edition, 5, 21b, 30a, 32b; 6, lOb, 18a; 97, 8a–b; Ts‘êfu yüan kuei, 968, 12a–b.Google Scholar

⁶ Ho-lo-tan was also transcribed as . The location of Cho-p‘o is the outstanding problem of early Indonesian historical geography, and it is sufficient to note here what seems to have been Pelliot's latest view which was that it represented Java-Sumatra; Études asiatiques, 1925, II, 250. The evidence that it lay south of Ch‘ih T‘u is contained in the Sui shu, 82, 3aGoogle Scholar. For Ch‘ih T‘u see Dr. Wheatley's thesis, 74.

¹ On P‘o-li's first embassy in 473 see Pelliot, ‘Meou-tseu ou les doutes levés’, TP, xix, 1920, 267, 433.Google ScholarPo-li , a variant form, was said to be 3,000 li south of Chin-li-p‘i-shih TPYL, 788, 6bGoogle Scholar, quoting the ‘T‘ang shu’; TFYK, 957, 8b. Chin-li-p‘i-shih was, according to the same sources, 1,500 li east of Ch‘ih T‘u.

² Ferrand thought that Kan-t‘o-li was an ancient name for Sumatra; JA, XI^e Sér., xiv, 1919, 238–41. Kan-t‘o-li is the sixth century transcription by which the state is usually known. In the fifth century it was Chin-t‘o-li The Nan shih, Ssŭ pu pei yao edition, correction at the end of chapter 78, stated that sent an embassy in the reign of the Sung emperor Hsiao Wu (454–65) which was recorded in the Liu Sung shu, 6, 11b, as an embassy from in 455. See n. 5 below for another variant of the name.Google Scholar

³ For P‘o-huang see p. 343, n. 5, above.Google Scholar

Ho-lo-tan: 430, 433, 434, 435, 436, 437, 452.

Kan-t‘o-li: 455, 502, 518, 520, 563.

P‘o-li: 473, 517, 522.

The evidence is contained in the Liu Sung shu, 5, 9b, 13b, 14a, 16a, 33b; 6, lib; 9, 3b; 97, 5b–8a; Liang shu, 2, 8a, 28b, 29b; 3, 3a; 54, 16b–18a, 20a–21a; Ch‘ên shu, 3, 14b; Ts‘êfu yüan kuei, 968, lla–13a, 18b; 969, 1a.

⁴ Ts‘êfu yüan kuei, 968, 18b–19a.Google Scholar

⁵ Sui shu, 34, 32b.Google Scholar Their titles were: in ten chapters; in four chapters. Dr. Ch‘ên Pang-hsien considered that these books, with other ones mentioned in the Sui shu bibliography, were evidence of Indian and Buddhist influences on Chinese medicine; Chung kuo i hsüeh shih, 151.Google Scholar

¹ Cosmas Indicopleustes, The Christian topography, translated from the Greek, and edited, with notes and introduction by McCrindle, J. W., Hakluyt Society, 1897, 363–6.Google Scholar

² History of the Wars, translated by Dewing, H. B., Loeb Classical Library, 1914, I, xx, 12, p. 143.Google Scholar

³ Giles, H. A., The travels of Fa-hsien (a.d. 399–414), 1956Google Scholar impression, 66–7. T‘ang P‘ang may have mentioned Ceylon in the third century. See p. 327, n. 5, above.

⁴ Chung hua shu chü edition, 1957, 3, 99, 1769.

¹ The CLPT quotation from the Yu yang tsa tsu added ‘Po-ssŭ’ to the sentence about the cutting; 13, 321. For the Po-ssŭ clothing see Sung shih, Ssŭ pu ts‘ung k‘an edition, 489, 17a, in connexion with a Cho-p‘o embassy in 992. For the Malay numerals see Kumazo, Tsuboi, Actes du douzième Congrès International des Orientalistes, II, 1899, 121, n. 1.Google Scholar Laufer thought that the Po-ssŭ word for rhinoceros, mentioned by the Yu yang tsa tsu, represented the Malay hitam or ‘black’; Sino-Iranica, 473.

² The goods associated with Persia in the Wei shu have a close similarity with those of Ta-ch‘in in the Wei lüeh and indicate that Persia was thought to have access to an equally magnificent range of costly articles. But the Persian trade was superior to that of Ta-ch‘in because it also comprised peppers from India. Persia was in every sense the trading heir of Ta-ch‘in. Both countries were connected with frankincense.

³ Dr. Wang Gungwu also rejected Hirth's view but did not attach the same importance to the scale of the Indian Ocean trade and the Persian connexion with it in the period of the Southern Dynasties. The present writer believes that the Indonesian trade was initially supplementary to the western trade and not vice versa.

¹ Liang shu, 54, 7a.Google Scholar The account was probably based on third century information. Tun-sun was mentioned in the Nan chou i uni chih of that century but never sent an embassy in the Liang period.

² Wei shu, Ssŭ pu ts‘ung k‘an edition, 5, 5b.

³ The Wei lüeh mentioned An-hsi. It referred to events in the third century.

⁴ The Tsin shu had no section on An-hsi but mentioned it in connexion with Ta-ch‘in and also in connexion with perfumes. For the latter see TPYL, 982, Ib.Google Scholar These embassies are listed in the Tsin shu, Ssŭ pu ts‘ung k‘an edition, 97, 8a–b.

⁵ Pei shu, Ssŭ pu ts‘ung k‘an edition, 97, 20a.

⁶ TPYL, 981, 6a, quoting from the Hou Chou shu. The Hou Chou reigned in northern China from 557 to 581.Google Scholar

⁷ The Latin version of Scholasticus by St. Ambrose, contained in Coedès, G., Textes d'auleurs grecs et latins …, Paris, 1910, 101.Google Scholar

¹ Liu Sung shu, Ssŭ pu ts'ung k'an edition, 97, 29b–30a.

² Ammianus Marcellinus, quoted in Professor Coedès' Textes d'auteurs … 92.Google Scholar

³ The coins are discussed in Warmington, E. H., The commerce between the Roman Empire and India, 120–5.Google Scholar

⁴ Liang shu, Ssŭ pu ts'ung k'an edition, 54, 24b.

⁵ This would account for references to genuine Persians in the fragments forming the basis of this study. The Nan chou chi mentioned them in connexion with Pistacia nuts. See p. 335, n. 5, above.

¹ Coectès, G., Les états hindouisés d'Indochine et d'Indonésie, 1948, 51Google Scholar; Bosch, F. D. K., Local genius en oud-Javaanse kunst (Medcdeelingen der Koninklijke Nederlandsche Akademie van Wetenschappen, Amsterdam, NS, xv, 1), 1952.Google Scholar