Section 12 – The Products of Da Qin 大秦 (the Roman Empire)

1. yeguangbi 夜光 [yeh-kuan-pi] – literally: ‘night-shining bi.’ The word bi [pi] sometimes denotes ceremonial stone disks, usually of jade but, sometimes, it is used more loosely and refers to jade in general. The Weilue mentions 夜光珠 yeguangzhu [yeh-kuan-chu], literally, ‘night-shining pearls’ among the products of Da Qin, presumably referring to the same product.

“The Chinese lore of “luminous pearls” (or “beads”) and “night-shining pearls” and “luminous moon pearls” ... goes back to Chou times, and may be ultimately of Indian origin. It has parallels and analogues in many cultures....
          Actually, the luminescent “gems” seen in China were often the eyes of whales, which, like the body parts of many marine creatures, were naturally phosphorescent....
          But there were also luminescent gems of mineral origin; some stones have this quality continually, others only when rubbed or heated. During Hsüan Tsung’s first reign an embassy from Maimargh presented the monarch with a gem called gem called simply *pi̯ɒk. This was the name of an archaic flat stone ring, a symbol of the heavenly kingship in Chou times; but it was also a word used interchangeably with *pi̯äk, “dark blue-green stone” and sometimes “luminescent blue-green stone.” If not a ceremonial jade ring, then, this gift was probably made of chlorophane, the thermoluminescent variety of fluorite, which was undoubtedly the material of the phosphorescent “emeralds” of classical antiquity, such as the green eyes of the marble lion on the tomb of King Hermias of Cyprus, though the Hellenistic alchemists had methods, seemingly magical, of making night-shining gems by the application of phosphorescent paints to stones, the most famous being their “emeralds” and “carbuncles.” Schafer (1963), pp. 237-238.

Yeh-kuang-pi (or Yeh-kuang-chu) .... (HHS, WL, CS, WS, Nestorian Stone, TT). This “night shining ring” or “night shining pearl” is also mentioned in most of the texts, but it is not clear what jewel it is. “Diamond,” “chlorophane,” “tourmaline (lychris),” and “ruby” have been suggested. Chavannes and Hirth suggest possible parallels in western literature. This night shining jewel has also occurred in earlier texts, Mo-tzu and SC. As Demiéville points out, it is often associated with “the bright (moon) pearl.” Leslie and Gardiner (1996), pp. 209- 210.

The following account is of interest here as it may indicate that the “night-shining bi” may be just a name referring to a type of “jade,” and not a truly luminescent material:

“The special local product [of Jiuquan – Chiu-ch’üan – literally: ‘Wine Spring’], ‘night-glowing cups’, are produced in a third-floor factory.... Made of locally mined black and green Qilian jade, these cups come stemmed, thimble-shaped or three-legged in the ancient style. A 2,000-year-old legend tells how King Zhoumu was presented with one of these cups, which glowed when filled with wine and placed in the moonlight. The Tang poet Wang Han wrote:

 Grape wine from a night-glowing cup is good,
 I want to drink, but the pipa urges me to mount my horse,
 Lying drunken on the battlefield would be no laughing matter,
 Tell, how many soldiers ever did return?”

(Sadly, the cups do not glow in modern moonlight!)” Bonavia (1988), p. 80. 

2. mingyuezhu 明月珠 [ming-yüeh-chu] – literally: “bright moon pearls.”

“The large pearls range from 0.5 to 1.5 inch across. There is a variety known as “pendant pearl,” which is slightly oval in shape, somewhat resembling an inverted cooking pot, with one side highly lustrous suggesting gold plating. One of these is worth as much as a thousand taels of silver. This pearl since ancient days, has been labelled “bright moon” or “light at night.” Actually, these beautiful names have been accorded to the pearls because they glimmer with a thread of light if held against the sun on a fair day, not because there are pearls that really shine in the dark of night.” Sung (1637) p. 298.

“Ming-yüeh-chu are pearls produced in the southern seas, and if compared with those produced in the fresh water inside China, they are bigger in size and of a superior quality. Since the ancient times, pearls are produced mainly from the southern seas such as the Indian Ocean, the Gulf of Persia, the Red Sea, and so on. In the Han period, pearls produced in the Red Sea were imported through the eastern territory of the Roman Empire....” Harada (1971) p. 72.

See also Leslie and Gardiner (1996), p. 209; Liu (1988), pp. 57-58.

3. haiji [hai-chi] – literally: ‘terrifying chickens’ = fighting cocks according to a personal communication from Dr. Edmund Ryden, Fujen Catholic University, Taiwan on 2nd July 1998. Dr. Ryden also kindly pointed out that, “Zhuangzi knew of fighting cocks.” Following an early, misleading Chinese interpretation, this term has usually been connected with the next item in the list, rhinoceroses. See, for example, Leslie and Gardiner (1996), p. 202.

4. xi [hsi] = rhinoceroses. The Chinese were primarily interested in rhinoceroses for their horn which was widely used as an aphrodisiac, in various medicinal concoctions, and for carving.

“Another commodity which was fed into the trade of the Indian Ocean from the Barbarā coast was rhinoceros horn, possibly the single most valuable item in the Chinese pharmacopoeia, a veritable apotropaion of apotropaia, which could also afford raw material for the jeweler. The Chinese could, of course, obtain horns from their own southern provinces and from South and Southeast Asia, but the market was so elastic that from time to time Arab merchants found it worth their while to bring to China the horn of the African rhinoceros.” Wheatley (1975), p. 106.

“Contrary to a universally held Western misconception, the rhino’s horn is not widely considered to be an aphrodisiac. Only the Romans (and, nowadays, a few Indians) believed it to have this property, presumably either because it is long, hard and pointed upwards or because the rhino itself is so generously endowed by the size of its penis and takes over an hour to complete its copulation. This is the only time that rhino’s horn has been given a medicinal value in Europe, although its value as a wondrous object associated with the unicorn existed for hundreds of years....
          In the Far East, however, it is another story and rhino horn has been on the books of traditional herbalists and exponents of folk-medicine since well before the time of Christ....
          Depending upon where one looks in the Far East, rhino horn has a variety of wonderful properties. In India, it is still – though very infrequently – offered as an aphrodisiac when mixed with herbs and swallowed in milk or honey: it was from the East that the Romans heard of this supposed property. Similarly taken, it is also said to cure arthritis, muscular pains and spasms and paralysis: fat and stomach lining are also said to cure polio and skin diseases. In the past, the horn was burnt under the anus of hæmorrhoid sufferers to alleviate their condition and to counteract constipation....
          It has been the horn of the Asian rhinoceroses which has been considered the most effective medicine but, with the decline of the Asian rhinos in the last two centuries, the Chinese have turned to the African rhinos for their supplies, dosages being increased because the African rhinos do not apparently have the concentrations of power of the Asian ones....
          Rhino horn shavings are given as a treatment for the lowering of fever such as typhus and malaria. The idea, as is so often the case with such traditional brews, is that the liquid cleanses the body of poisons. Additionally, it is regarded as a cure for laryngitis, bronchitis, tuberculosis and poor eyesight. Dried and powdered rhino’s blood is sold as a tonic for sufferers of anæmia which it probably does help to cure being, like snake’s blood, rich in iron.” Booth (1988), pp. 156–159.

“The horn of the rhinoceros played a role in the minor arts of T’ang very similar to that of ivory, and indeed the two substances were regularly linked in language, particularly in parallel verse. The demand for rhinoceros horn was very great, so that, although many rhinoceroses still lived in Hunan, as we have seen, and their horns were submitted to the court as tribute, it was also necessary to import them. From close at hand, they were obtained in Nan-chao and Annam; more remotely, they came to the port of Canton from the Indies, and in such quantities that the near extinction of the Indochinese rhinoceroses in modern times can in large part be attributed to the China trade of the T’ang....
          Rhinoceros horn was important in medieval Chinese medicine, especially as an antidote for all kinds of poison. Belief in its efficacy goes back to the fourth century, and may have originated in China, to spread to Western Asia and the Roman empire.” Schafer (1963), p. 241; see also ibid. p. 83.

5. shanhu 珊瑚 [shan-hu] = coral. See, for example, GR No. 9520.

“Coral is as highly valued among the Indians as Indian pearls. It is also found in the Red Sea, but there it is darker in colour. The most prized is found in the Gallic Gulf around the Stoechades Islands, in the Sicilian Gulf around the Aeolian Islands, and around Drepanum....
          Coral-berries are no less valued by Indian men than specimen Indian pearls by Roman ladies. Indian soothsayers and seers believe that coral is potent as a charm for warding off dangers. Accordingly they delight in its beauty and religious power. Before this became known, the Gauls used to decorate their swords, shields and helmets with coral. Now it is very scarce because of the price it commands, and is rarely seen in its natural habitat.”
Pliny NH (a), p. 281. (XXXII, chaps. 21, 23).

“Coral was exported [from Egypt] to India as well as to Arabia.... Red Sea coral, to be had all along the western coast of Arabia, hardly required importation via shippers from Egypt; moreover, it was considered of inferior quality (Pliny 32.21). The coral referred to here [in the Periplus] must have come from the Mediterranean, which produced prized varieties....” Casson (1989), p. 163. See also ibid. p. 191.

“Since the period of the Former Han dynasty coral had been an extremely valuable commodity....
          From where and on what route did coral – so highly valued by the Chinese – come to China? Red coral from the western Mediterranean and the Red Sea was one of the major items shipped to the East from the time of the Periplus (28, 39, 49). The histories of the Later Han (HHS: LXXXVIII, 2919), the Three Kingdoms (SKC: XXX, 861) and the Chin (CS: XCVII, 2544) mention coral as a product of Ta-ch’in, i.e. the Roman empire. A later Chinese account gives a detailed description of how coral was collected from the sea in Ta-ch’in: the Romans dropped iron nets on the coral reefs so that the yellowish young coral would grow on them. Three years later they came back to collect the coral once it had turned red (Hsin T’ang-shu: CCXXI, 6261).
          Those records definitely refer to Mediterranean red coral. There were three possible routes to ship the coral to China. The most frequented route was the Southern Route to India. In the time of the Periplus the primary destination of coral in Roman cargo ships was India. Pliny mentions that coral was as highly treasured in India as pearls were in Rome (XXXII, 11). Coral beads along with beads of other precious materials have been found in north-Indian sites, for example at Rajghat in the level of the pre-Kushan period (Narain 1976-8: II, 12)....
          The second possible route was through the Northern Route of Central Asia. The Wei history describes coral as originating in Persia, probably because some coral was transported through Persia and the Northern Route into Central Asia. Ferghana’s gift to the Chao state in 331 AD included coral (Wang Chung-lo 1979: 704).
          The sea route from the Red Sea to south China was the third, and the most unlikely, way. Although there are some vague references to coral imported from southern ports during Han times (Shu-I-chi: 1/3a-b), most other Chinese sources call coral one of the commodities from the Western Region.... No matter where the coral originated, north India was probably the main supplier of trans-shipped coral to China before the T’ang dynasty.” Liu (1988), pp. 54-57.

“Whatever lang-kan may have been, unambiguous coral (shan-hu), presumably a red variety, was a product of Nam-Viet in T’ang times. Indeed the revered many-branched coral trees, twelve or thirteen feet high, which ornamented the garden of the Han palace, had been the gift of Chao T’o of Nam-Viet. Coral had another honorable place in Nam-Viet tradition: there was a well called “Coral Well of Ko Hung” at Mount Lo-fou. It was said that the coral which went into its construction had been the gift of a sea-deity to an early alchemist. Independently of the precious Corallium nobile imported from the Mediterranean world, then, Nam-Viet had its own fine corals, some of them red. Specimens of lovely red corals of the genus Corallium have in fact been found from Japan to the Sulu Sea in modern times.” Schafer (1967), p. 159.

See also Leslie and Gardiner (1996), p. 210.

6. hupo [hu-p’o] = yellow amber. GR, under No. 4570, p. 124. See Pulleyblank (1963), p. 124 where he gives for hupo: “M. hou/ -phak < *ha·-phlak “amber” (Han-shu 96 A, under Chi-pin, Kashmir). This may represent Greek *ρπαξ “amber”. The equation had been suggested by G. Jakob in 1889 but was rejected by Laufer. Though the Greek word is only attested in Latinized form in Pliny as a term used in Syria, the epithet “snatcher” is an appropriate one for amber and may well have been known in other Greek speaking regions.

“In Europe the biggest and most important supplies of amber traded in early times were found at Samland on the Baltic coast and in smaller quantities on the North Sea. The chief mining area was near Kaliningrad (formerly Konigsberg).... An important eastern route ran from the Baltic coast along the Vistula and through the territory of Kiev southeastwards to the Black Sea. Here this Eastern amber route linked up with the long and ancient overland connections to the Near East, central and east Asia and India.... Some of the amber sold in Asia came from Burma.” Raunig, (1984), pp. 14-15.

“[Amber is] the most expensive product... from trees or shrubs, amber, balsam, myrrh and frankincense....”
Pliny NH (a), p. 377. (XXXVII, 204).

“The ‘Baltic’ Balts are first mentioned by Tacitus, under the name of Aestii; he praises their skill at growing crops, ‘with a patience quite unusual among the lazy Germans.’ Of more general importance was that the land of the Aestii produced (and still does produce) most of the world’s supply of amber. Beads of this substance made their appearance in Greece as early as 1500 BC, and were also exported to many other parts of the world. The Roman Empire, as usual, operated on a larger scale than anything done before. From Pliny, for example, we hear that in Nero’s reign (AD 54-68) a Roman businessman visited the amber country and brought back enough amber to decorate all the equipment for a large gladiatorial show. The biggest piece weighed thirteen pounds.” Sitwell (1984), p. 41.

See also Leslie and Gardiner (1996), p. 211.

7. liuli 流離 [liu-li] = opaque glass. There has been much discussion about whether liuli in these early texts referred to glass or to some natural gemstone. See, for example, Leslie and Gardiner (1996), p. 213.
By the first century CE glass in all its forms had become one of the major exports of the Roman Empire. This was due to three main factors:

1. Rome had recently acquired the main glass-producing centres of the ancient world which were centred mainly in Syria and Egypt. It not only controlled their production and exports but imported the latest technologies (and top craftsmen) to Italy itself where a huge new industry was established.

2. The Syrians, in particular, had not only developed techniques for producing clear glass wares (which, until foreigners became aware of the scam) could be passed off as valuable rock crystal wares, but were able to produce transparent glass in a very wide range of colours. It seems that it took several centuries before transparent (rather than merely translucent or opaque) varieties of glass was able to be produced in China – see below. 

3. The development of glass-blowing in the second half of the first century BCE – probably in Syria – meant that, for the first time, glass vessels could be cheaply and quickly mass-produced. The industry expanded rapidly and by the end of the first century CE free-blown and mould-blown glassware formed the bulk of glass objects produced in the Roman Empire.

It seems to be now generally accepted that opaque glass only was produced in China until superior manufacturing techniques were introduced by Yuezhi merchants in the fifth century. The following account of this technology transfer also makes it clear that the liuli previously imported from the west was indeed glass:

“According to the Pei-shih . . . it was during the time of T’ai-wu of the northern Wei dynasty (A.D. 424-452) that traders came to the capital of Wei from the country of the Ta-yüeh-chih . . . , bordering on the north-west of India1 who said that, by fusing certain minerals, they could make all colours of liu-li. They then gathered and digged in the hills, and fused the minerals at the capital (near the present Ta-t’ung-fu in Shan-hsi). When ready, the material so obtained was of even greater brilliancy than the liu-li imported from the west. The Pei-shih specially states that, after this event, articles made of glass became considerably cheaper in China than they had been before....

1. According to the Wei-shu, quoted in the Yüan-chien-lei-han, ch. 364, p. 31, they came from India..”

Hirth, pp. 230-231 and n. 1. Note that Hirth quotes another story from Grosier’s Description de la Chine, edition of 1787, Vol II, p. 464, which relates this event to an “Emperor Tai-tsu” of the Sung, which Hirth maintains was another name for emperor Wen-ti of the Sung (A.D. 424 to 454). Doubt has, however, been cast on the authenticity of this latter story – see Leslie and Gardiner, p. 214 and n. 75. See also: Boulnois (1992), pp. 178-179.

   “Tracing the history of glass as a commodity in Chinese foreign trade poses several problems. Previously, scholars thought that China did not develop glass-making techniques until the fifth century AD. But since the 1930s, many glass samples have been found in tombs dating from the fifth century BC. Doris Dohrenwend recently summarized the history of Chinese glass comprehensively. She divides Chinese glass into two categories. The small opaque items pre-dating the third century AD are liu-li, and the transparent vessels from the T’ang dynasty onwards are po-li. Between the two phases during the Northern and Southern dynasties there was a ‘glass mini-boom,’ as indicated by a series of glass vessels of doubtful provenance (Dohrenwend 1980: 426-46).
          Today no one doubts that the Chinese made glass long before the Christian era. There is also clear evidence that China imported glass from foreign countries even up to the Ch’ing dynasty. The real question is: did the Chinese regard the ancient opaque items made by them or their ancestors as being the same thing as the transparent or colourful glass they imported at the same time? Obviously not. Both terms, liu-li and po-li, appeared in the Chinese vocabulary after contact with the Western Region, and both have Sanskrit origins.” Liu (1988), pp. 58-59. See also ibid. pp. 60-63, 80, 160-161.

“Glass had been familiar to the Chinese for centuries, and had been manufactured by them since late Chou times. Their language distinguished two kinds of glass, liu-li and po-li. Liu-li was colored glass, either opaque or only dully translucent, or even a colored ceramic glaze; it was akin to the lead glass which we call “paste,” and like paste was thought of as a substitute for natural gemstones, especially for green and blue ones. Indeed, it was sometimes confused with real minerals, such as lapis lazuli, beryl, and, no doubt, turquoise. Po-li, on the other hand, was transparent, either colorless, like rock crystal, and compared with water and ice, or else palely tinted. Liu-li was already old in China, but blown vessels of po-li were a novelty in T’ang.
          Little need be said of the false gem liu-li. It was familiar in both life and literature, and was doubly exotic in that it came occasionally with embassies from the West, and was also reported of distant cultures, such as Pyü in Burma....” Laufer (1912), pp. 235-236.

“The Chinese word liu-li apparently transcribes Pali veluriyam (Sanskrit vaid
ūrya) and in the Buddhist literature continues to have the same referent, that is, “beryl” or some other green gem. For this reason, Laufer (1946), 111-112, did not accept the meaning “glass” for it, and, though he admitted that certain colored glazes were sometimes called liu-li, he considered po-li the only usual word for glass in China. Po-li transcribes a form close to Sanskrit sphaṭika, “crystal.” Cf. Needham (1962), 105-106.” Schafer (1963), p. 335, n. 137.

“One of the products that Rome exported further and further afield, was glass objects, particularly coloured glasses, containers of all sorts, cut glass, glass beads for necklaces from the workshops of Syria or those of Puteoli. These necklace beads have been found from the shores of England to those of the Annam Sea, in Central Asia and the Ukraine. They were made round or oval, pear-shaped and cylindrical, in the shape of disks and amphora, in opaque and translucent glass. There were blues and greens, and whimsical beads made of alternating layers of blue glass, bronze, and white pottery.” Translated from Boulnois (1992), p. 75.

“Glass, for example, initially imported from Hellenistic nations, was first introduced in China during the Warring States period [481-221
BCE], as the fragments discovered at Jincun near Luoyang, or at Changsha have confirmed. Apart from being prized for its beauty, glass, which was as uncommon in China as jade and served as a substitute for jade, was considered priceless by the rulers of that era on account of its rarity. Under the Han dynasty, glass was imported from Syria which filled specific orders for the Chinese market. Indeed, glass was used to produce jewels and inlaid work for belt plates or bronze mirrors. Sometimes it was substituted for jade in the form of small plates that were inserted in the mouths of corpses. Authors even wrote admiring poems to praise this extraordinary substance.” Elisseeff (1983), pp. 163-164.

“The evidence for Roman trade in glass with the cities along the east coast of South India is exactly the opposite of that on the west coast. There is no written evidence, but finds from excavations are abundant. A Chinese record from the end of the second century B.C. says that, among other goods, the Chinese got glass from Kanchipuram.30 No published archaeological evidence for glass trade at Kanchipuram is known to me, nor for glass trade from its ports at Vasavasmudram and Mahabalipuram. However, the chance find of a large fragment of a Mediterranean amphora at Vasavasmudram indicates that Mediterranean wares reached this port. Therefore, the possibility cannot be excluded that glass exported from the West to Kanchipuram was destined for transit trade with China.”

    30. J. Duyvendak, China’s Discovery of Africa (London 1949) 9-10. See also Stern (infra n. 37) [E. M. Stern, Ancient Glass at the Fondation Custodia (Collection Frits Lugt) Paris (Groningen 1977) 25-30.]

Stern (1991), p. 117 and nn. 30, 37. See this article for a detailed discussion of the types of glass manufactured in the Roman Empire and exported to India and Africa and which are probably indicative of the types of glass and glass wares exported to China as well.

8. langgan 琅玕 [lang kan]. Probably a whitish chalcedony. However, there have been many definitions of langgan and, perhaps, it has meant different things at different times. It has frequently been described as a kind of branching coral or “coral tree,” and sometimes as a pearl-like type of stone.

            The oldest reference I can find to langgan is in the Guanzi 管子 which says that they originated “from the mountains nearby the Yuzhi [Yuezhi];” specifically the Kunlun mountains. Now, the compilation of the Guanzi “was probably begun by the scholars of the Chi-hsia Academy founded c. 302 B.C. in Ch’i State, that most of the chapters belong to the third century, while some may still be earlier, and others were added in the second or even the first century B.C. Thus the book was mostly written before the Han period, even though some of its ideas are of a later date….” Pokora (1973), pp. 31-32.


“In the “Qingzhong Jiapian 蜻重甲篇 of the same book [the Guanzi] it is also recorded:


     If what is valued at no less than one thousand pieces of gold are white jade discs, then we should be able to persuade the Yuezhi, who are at a distance of 8,000 li, to present tribute. If clasps and earrings worth no less than one thousand pieces of gold are made from qiulin 琳 (a kind of beautiful jade) or langgan琅玕 (a kind of white carnelian), then we should be able to cause the Kunlun Hills 昆侖之虚, which are at a distance of 8,000 li, to present tribute.


The “Qingzhong Yipian” records also: “Jade originates from the mountains nearby the Yuzhi [Yuezhi], which are at a distance of 7,000 li from Zhou周.” Yu (1998), p. 48.


Yu believes these “Kunlun” Mountains “may have referred to the Altai Mountains,” but I prefer to accept the more usual definition of them as the high chain of mountains separating the Tarim Basin from the Tibetan Plateau, in particular, the famous jade-bearing regions south of Khotan and Yarkand.

            Yu’s identification of langgan as a “kind of white carnelian” undoubtedly indicates the whitish form of chalcedony. Carnelian, a form of chalcedony is, by definition, of a reddish colour, but chalcedony comes in a variety of other colours, including bluish-white, grey, yellow, or brown. It is a waxy, fine-grained form of silica much favoured by gem engravers. It is particularly suited to the manufacture of seals, as hot wax tends not to stick to it.

            The GR, No. 6687], on the other hand, gives, among its definitions: “balas-ruby : a precious stone of yellow or red from the Indies; a stone in the form of a pearl; name of a tree : a tree of pearl.

            Williams (1909), p. 498 says: “[lang kan:] white coral of a firm texture, branched like a Gorgonia, but not susceptible of polish.”


Lang-kan 琅玕is a stone variously said to resemble pearl and jade; the term occurs in the Shu-ching (6.21a; Couvreur, p. 79; Legge, III, 127). Legge suggests that it is lapis lazuli. Schafer describes lang-kan as a fairy gem, the stuff or fruit of a tree of paradise, or of an axial world-tree” (“The Origin of an Era,” p. 545; cf. his The Golden Peaches of Samarkand, p. 246, and The Vermilion Bird, p. 159).” Rogers (1968), p. 257, n. 486.


“Since the period of the Former Han dynasty coral had been an extremely valuable commodity. In eulogies describing the court’s brilliance in Former Han time, Later Han writers such as Pan Ku mentioned ‘coral trees’, i.e. branch coral (Liang-tu-fu, 4a). In fiction written in a later period coral trees symbolize the extravagance of the Former Han court. It is said of Emperor Wu of the Former Han built a shrine with ‘coral window lattice’, and with ‘coral trees’ planted around it, where he searched for immortality in vain (Lu Hsün 1939: 347). This tradition of using coral continued after the Han.1 It seems that coral was the most precious and, hence, the ideal item of tribute. More specific records about the use of coral appear after the Han....”
          “Even as late as the T’ang. In the famous picture by the T’ang artist Yen Li-pen: ‘Foreign envoys coming with their tributes’ (Schafer 1963), many envoys carry a piece of ‘coral tree’.” Liu (1994), p. 54, and n. 1. On these “tree corals” see also, for example: Maunder (1878), p. 398 – under “Madrepore”.

9. zhudan 朱丹 [chu-tan] = ‘red cinnabar’. See Leslie and Gardiner (1996), p. 208.

“Among the mineral species found in T’ang Nam-Viet, the most distinguished in Chinese tradition was cinnabar. This red sulphide of mercury was virtually a sacred stone to the Taoists, for whom its ready conversion into shining quicksilver was a true miracle which, properly understood, could lead to the transmutation of metals and the prolongation of life. It also yielded the pigment vermilion, the color of life and blood and eternity, applied in antiquity to the bodies of the dead, to coffins, to mimics of the gods, in short, to all sorts of holy objects, and, latterly also to secular pictures. It was also well regarded as a drug for the treatment of serious diseases, and in some opinion it was the true panacea.” Schafer (1967), pp. 156-157.

“Lastly, you find there [in the plains around Bathang, just over the border from Tibet in Szechuan] mines of cinnabar (sulphide of mercury). The Tibetans get the mercury in all its purity by disengaging the sulphur by combustion, or by combining it with slack-lime.” Huc (undated), p. 123.

For a good description of the production of cinnabar in China, see Sung (1637), pp. 279-285.

10. qingbi
青碧 [ch’ing-pi]. It is difficult to decide exactly what this term qingbi refers to. Qing is notoriously difficult to translate precisely but usually refers to what we call ‘green’ or the greenish-blue of the ocean. Williams, p. 158, defines it as “The first of the five colors, the color of nature, as the green of sprouting plants, the blue of the sky, and the azure of the ocean ; but especially the dark green of plants ; the green part of a thing; wan, fading away, pale; black.”
          The GR Vol. iv, p. 986 identifies bi as a blue-green stone resembling jade; nephrite; jasper. Williams, p. 637 says of bi: “Green jade stone ; some kinds are blueish, and others are greenish, like the deep sea ; it is like jadeite and highly prized.”
          Chavannes (1907), p. 182, n. 5, accepts Geert’s identification as a sort of blue-green jasper, but I would suggest that, in the context of Roman trade at this period, it more likely refers to the highly-prized green peridot and/or emeralds – both of which the Romans were mining and exporting at this period.
          It is not possible to choose for certain between them. However, I would suggest that, as the name qingbi (or “green gem”) seems to be otherwise unrecorded, that it probably referred to peridot, a gem most likely unknown to the Chinese until they began trading with the Romans, as all the peridot of the ancient world was mined on the island of Zebirget (also known as St. John’s Island) in the Red Sea. It was probably the trade item called chrysolithum mentioned in the Periplus of the Erythraean Sea as being traded to India in the 1st century

“Whether Pliny’s chrysolithum is actually yellow sapphire or some other stone of India, it cannot be what the author of the Periplus has in mind, for he is referrring to a stone imported by India and hence not obtainable there…. As it happens, there is a desirable stone not found in India which the shippers of Roman Egypt had at their very doorstep, so to speak, namely peridot. The sole source of peridot in ancient times was St. John’s Island in the Red Sea…, only a few miles away from Berenicê, and the stone was so prized it served as gifts to royalty (Pliny 37.108) and the kings of Egypt maintained a guard over the quarries (Strabo 16.770). Both Strabo and Pliny (37.107) call the stone topazos, but since they name the island as the source, there can be no doubt that they are referring to peridot (cf. Warmington 253; Eichholtz’s note to his translation of Pliny 37.108). Peridot, the gem grade of olivine, is green rather than “golden,” but there seems to have been some confusion about the color; whereas Pliny properly says it has “its own greenish nature,” Strabo describes it as being “like gold.” Casson (1989), p. 190.

It is, of course, as I mentioned above, also possible that qingbi was an early name for emeralds. The Romans controlled the famous ancient mines in the region of Mons Smaragdus (‘Emerald Mountain’) in Wadi Sikait, not far northwest of the major Red Sea port of Berenice which handled much of the Roman maritime trade with India and the Far East. For an excellent account of these mines and their importance, see the article, “Emerald City,” Rivard, Foster and Sidebotham (2002).

11. lajinlüxiu 剌金縷繡 [la-chin-lü-hsiu]  – literally: ‘drawn gold-thread multi-coloured embroideries.’ The first word la can mean either ‘cut’ or ‘drawn’ in the sense of ‘to pull’ or, as in forcing gold through a hole to make a thread.

“The art of weaving gold-thread into cloth was of very ancient origin. I quote from the Very Rev. Rock’s “Textile Fabrics” (South Kensington Museum Art Handbooks, No. 1, p. 23): “The process of twining long narrow strips of gold, or gilt silver, round a line of silk or flax, and thus producing gold-thread, is much earlier than has been supposed; and when Attalus’ name was bestowed upon a new method of interweaving gold with wool or linen, thence called “Attalic,” it was probably because he suggested to the weaver the introduction of the long-known golden thread as a woof into the textile from his loom. It would seem, from a passage in Claudian, that ladies at an early Christian period used to spin their own gold-thread.” According to the same author, the superior quality of Cyprian gold-thread was known to the mediæval world. Attalicæ vestes are mentioned in Porpertius, but, as Rock (l. c., p. 14) points out, “the earliest written notice which we have about the employment of gold in the loom, or of the way in which it was wrought for such a purpose, is in the Pentateuch. Among the sacred vestments made for Aaron was an ephod of gold, violet and purple, and scarlet twice dyed, and fine twisted linen, with embroidered work; and workmen cut also thin plates of gold and drew them small into strips, that they might be twisted with the woof of the aforesaid colours.”
   The combination of several materials (silk, wool, linen, byssus) in the same texture… was well known in ancient manufacture, and the Syrian school of art is especially known for the great variety of musters produced by means of coloured threads.” Hirth (1875), pp. 253-255.

          Some extraordinary discoveries were made in 1978-79 during the joint Soviet-Afghan excavations of six graves dating to the beginning of the 1st century CE at Tillya-tepe (literally: “The Golden Mound”), 5 km north of the present town Shibarghan – or just over 100 km by road east of Balkh. See note 13.8 for more details on this site.
          A spectacular treasure of hundreds of precious gold and jewelled objects were uncovered in these graves which were presumably those of Kushan nobles. Among these treasures is clear evidence in one of the graves of cloth woven with threads of gold:

“Besides the sewn-on plaques, the gold threads of a decayed cloth were also uncovered ; they were found exclusively beneath the vertebrae and nowhere else, from which it may be deduced that they are either from the shroud that lined the inside of the coffin, or, more likely, from a short cape or cloak. As the threads retained a wavelike form, one may presume that the weft threads had been of either wool or silk, which had completely rotted away before the grave was excavated.
   The gold threads were disposed beneath the skeleton, in isolated multi-layered patches not in one unbroken pattern, thus indicating large ornamental designs. Considering that the gold threads had had a multitude of pearls on them and that the intricate designs they had thus created were edged with gold plaques, one may gain some notion of how dazzling a sight the funerary garments must have presented.” Sarianidi (1985), p. 20. Also: Sarianidi (1990-1992): p. 106.

The following note from the China Daily dated 17th May 1999 is also of great interest:

“URUMQI (Xinhua) – A garment made of fabrics with dazzling gold foil sewn on applique work, dating back to the Western Han Dynasty (206 BC – AD 24), was recently unearthed from a tomb in Lop Nur, a desolate area in Northwest China’s Xinjiang Uygur Autonomous Region.
          Chinese archaeologists believe this is the earliest woven material with gold patterns ever found in China. The discovery pushes the history of fabric-making with gold back 1,000 years earlier than previously believed,” said Li Wenying, an archaeologist participating in the excavation.
          The gold foils were sewn by two different methods. One was to cut coloured silk in the shape of flower buds, petals, stamen and fruits, which were then pinned to the collar, sleeves, lower hem, and back of the garment. Then gold foils were pasted on the silk designs. The other way was to spread gold powder onto the design.
          One archaeologist, Zhou Jinling, described the embroidery as “distinctive, dazzling and harmonious in colour.” This garment was one of the 200-plus rare cultural relics unearthed from 32 tombs built in the period between the Han and Jin dynasties (206 BC – AD 420) at the Yingpan Ruins in Lop Nur.
          The site lies 200 kilometres east of Loulan, one of the busiest commercial cities on the ancient Silk Road which served as a transportation artery for the flow of goods from China to the West more than 2,000 years ago. The flourishing trade route began to decline in the fourth century.
   The recent excavation, which lasted for more than a month, was the continuation of a protective excavation begun in 1995. During the three-year period, Chinese archaeologists opened 32 ancient tombs and cleared more than 100 robbed tombs over a large area.
          One-third of the unearthed objects were burial accessories, including garments, wooden, bronze, and lacquer wares, gold and silver ornaments, and pearls. The most significant finding was three woolen robes with designs of flying men, eagles, and snakes woven with a jacquard technique.”

12. zhichengjinlüji成金縷罽 [chih-ch’eng-chin-lü-chi], literally: ‘woven with gold thread net.’ It is difficult to determine exactly what is meant here, as the dictionaries are not unanimous. The word ji can either mean, according to GR 878, either a “fisherman’s net,” or “cloth of rough material, of fur or hair.” Couvreur, p. 726 gives: a “fisherman’s net” or “fabric of hair coming from a western country; rug.” Williams, p. 67 defines it as “a kind of fishing-net or seine made of hair,” I have chosen “net” as the more likely meaning here, as have Leslie and Gardiner (1996), p. 215. However, it should be kept in mind that it may just as well have referred to some sort of rug or tapestry with gold threads woven into it.

13. zaseling zuohuangjintu 色作黄金 [tsa-se-ling tso-huang-chin-t’u] – literally: ‘delicate polychrome silks painted with gold.’ The word ling usually refers to a fine silk. See note 12.11 above and Leslie and Gardiner (1996), p. 215. According to GR No. 1169, tu carries the meanings of: to apply (a product); to spread; to colour-wash or whitewash; to coat or paint.

14. huowan bu 火浣布 [huo-wan pu] = asbestos cloth. Literally: ‘fire-washed cloth’ which, as noted by Leslie and Gardiner (1996), pp. 215-216: “ almost certainly asbestos.” See, for example, GR Vol. III, p. 300, No. 5231.

“The wonderful quality of asbestos was familiar to both Romans and Chinese from about the beginning of the Christian era. The men of Han regarded it as a Roman product, quite properly since this mineral fiber was very well known to the Romans, who also understood that it came from a rock. Here is Apollonius Dyscolus [2nd century CE] on asbestos napkins:

When these napkins are soiled, their cleansing is performed not by means of washing in water, but brush-wood is burnt, the napkin in question is placed over this fire, and the squalor flows off; while the cloth itself comes forth from the fire brilliant and pure.

This natural but somewhat ostentatious display is said to have had its counterpart in China in the second century, when a man purposely soiled his asbestos robe, and hurled it into a fire with simulated anger, only to bring it our fresh and clean. These anecdotes make the Chinese name for the mineral fabric understandable – it was “fire-washed linen.” But asbestos was also called “fire hair,” which illustrates another (and false) theory of the origin of the stuff. In the Hellenistic Orient it was sometimes thought to be of vegetable origin, like cotton, but among the Chinese, until the sixth century, and after that among the Arabs, the most popular theory was that it was the fur of the salamander-rat (but sometimes the phoenix) which was cleansed and renewed by fire.” Schafer (1963), p. 199.

15. Sea silk and wild silks. See: Appendix B: The Story of “Sea Silk”, and Appendix C: Wild Silks in Ancient Times, at the end of these notes.

16. suhe蘇合 [su-he] = storax. Classical storax (Storax officinalis) was produced in the Mediterranean region and was used primarily as an incense. It should not be confused with the medicinal “Liquid Storax,” a balsam obtained from the trunk of Liquidambar orientalis which has been used in Western medicine up until recent times to treat scabies and other skin parasites, as an expectorant, and in perfumery.

“The classical storax [Storax officinalis] imported to China long ago from Rome and Parthia had been dark purple in color, and some said it was lion’s dung – a fearful drug. This scented resin was, it seems, popular and well-known in pre-T’ang times….
          The place of this Western resin in China can be compared with that of another, myrrh, but unlike it, myrrh was the least noted of the exotic resins.” Schafer (1963), pp. 168-169.

“Storax is made by mixing and boiling the juice of various fragrant trees; it is not a natural product. It is further said that the inhabitants of Ta-ts’in gather the storax [plant, or parts of it], squeeze its juice out, and thus make a balsam [hsiang-kao]; they then sell its dregs to the traders of other countries; it thus goes through many hands before it reaches China, and, when arriving here is not so very fragrant.” From the Liang-shu, “written about A.D. 629, and comprising the period A.D. 502-556, ch. 54: the account of Chung T’ien-chu,” translation from Hirth (1885), p. 47.

See also: Liu (1988), p. 8 and n. 6.

17. “Ten silver coins are worth one gold coin.” I can find no reference to 10 silver coins equalling one gold coin in the Roman system. However, this probably refers to the relative value of silver and gold, presumably as reported by Gan Ying after his journey to Parthia in 97 CE. The silver content of the denarius was reduced by Domitian in 85 CE to about 80%, making the value by weight of silver coins relative to gold coins portantly, the term wang in this context probably does not carry its basic meaning of “king, ruler.” Wangmu is a cultic term referring specifically to the powerful spirit of a deceased paternal grandmother. So Xi wangmu probably means “Spirit-Mother of the West.” this paper discusses occurrences of wang as “spirit” in ancient texts, and concludes with a consideration of some etymological reasons as to why wang is sometimes used in this less common sense.” Goldin (2002), p. 83.

          The original legend placed the ‘Western Paradise,’ the abode of Xi wangmu, at the top of the Kunlun mountains, and is considered to be the centre of the world, where the Yellow River (and many other rivers had their source). This legend is possibly related to the Hindu, Bön, and Buddhist traditions of Mt. Meru or Sumeru (usually identified with Mt. Kailash in southwestern Tibet) as the centre of the world, and the source of several of the major rivers of Asia including the Indus, the Sutlej, the Ganges, and the Bramaputra. For a detailed discussion the Xiwangmu cult, see Fracasso (1988), pp. 1-45. See also Leslie and Gardiner (1996), p. 275. The Shiji says:

“The old men of An-hsi [Parthia] say they have heard that in T’iao-chih [Characene and Susiana] to be found the River of Weak Water and the Queen Mother of the West, though they admit they have never seen either of them.” Dubs (1962), p. 268.

The early references to the Liusha or ‘Shifting Sands,’ the Kunlun, and the ‘White Jade Mountain,’ all suggest that the original abode of Xi wangmu was thought to be in the mountains to the south of Khotan. As the Chinese got to know the regions to their west better, it seems they kept moving the supposed location further west to cover their inability to actually find Xiwangmu.
          By the time of the Weilue (which was composed between 239 and 265
CE) the supposed locations of Ruo Shui, Liusha, and Xi wangmu become extremely confused:

“In earlier times it was also mistakenly thought that the Ruo Shui (‘Weak River’) was west of Tiaozhi (Susiana and lower Tigris). Now it is (thought to be) west of Da Qin (Roman territory).
          In earlier times, it was also mistakenly thought that if you left Tiaozhi (Susiana and lower Tigris), and travelled more than two hundred days to the west, you reached the place where the sun goes down. Now, (it is thought that) you travel west from Da Qin (Roman territory) to reach the place where the sun sets.” From the Weilue.

“West of Da Qin (Rome) is a sea. West of the sea is a river. West of the river there are big mountains running south to north. West (of this) is the Chi Shui (‘Red River’). West of the Chi Shui (‘Red River’) are the Baiyu Shan (‘White Jade Mountains’).
          In the Baiyu Shan (‘White Jade Mountains’) lives Xiwangmu (‘Queen Mother of the West’). West of Xi wangmu (‘Spirit-Mother of the West’) are the long Liusha (‘Shifting Sands’). To the west of the Liusha (‘Shifting Sands’) is the kingdom of Daxia (Bactria), the kingdom of Jiansha (‘Stable Sands’), the kingdom of Shuyao (Sogdiana), and the kingdom of the Yuezhi (Kushans).
          West of these four kingdoms is the Hei Shui (‘Black River’), which is as far west as I know of.” From the Weilue.

It is quite likely that the Liusha of these accounts originally referred to the infamous Liusha or “shifting sands” on the Southern Route west of Cherchen. The 6th century Zhoushu, juan 50, provides a graphic description of their terrors, and places them to the northwest of Jumo [Chü-mo] – situated on the east bank of the Cherchen River, opposite the modern town of Cherchen:

“Northwest [of Jumo] there are shifting sands for many hundreds of li. On summer days there is a hot wind which is disastrous for travelers. Only the old camels know that the wind is about to strike; then, crying out and huddling together, they stand burying their mouths and noses in the sand. Whenever this happens, the men recognize it as a sign, and themselves take felt and press it over their noses and mouths to cover them up. This wind is swift, but after a while it is completely calm. Still, those who do not take precautions are sure to come to grief and perish.” Miller (1959), pp. 8 and 24, n. 44.

“Stein speaks of “tame deserts”: those found in Arabia, America, and South Africa that are deserts in their sense of solitude and emptiness, but “tame” because in them whole tribes can wander about for long periods of time sure of finding water at least at certain regular seasons. How different the true desert, “the dune-covered Taklamakan and the wastes of hard salt crust or wind-eroded clay of the Lop desert which stretch almost unbroken for a length of eight hundred miles from west to east. In them the absence of moisture bans not only human existence but also practically all animal and plant life”....
          Like ocean swells, the dunes move, the effect of northeast winds that rage over the desert much of the year, that are also constantly abrading the soft clayey soil unless it is already covered by dunes or anchored by desert tamarisks and poplars. At the ancient sites ruins of buildings or what were once orchards and arbors often rise above the wind-eroded bare ground on island-like terraces: these preserve the original level while around them the ground has been scooped out lower and lower.” Mirsky (1977), p. 113.

Leslie and Gardiner (1996), p. 276, state: “There can be little doubt that all of these refer to unreal places, not seen or visited by Chinese, nor by their West Asian informants.” See also: Shih chi, chap. 123. Translation in: Dubs, (1944), p. 268; the account in the Hanshu in CICA, p. 114 and n. 260 Shiratori, (1956c), p. 135, n. 131.

Finally, in regard to the legend of where the sun goes down:

“The actual wording of the Qian Hanshu (chap. XCVI, a, p. 6a) is a little different: “On leaving Tiaozhi, when on goes by water about a hundred days, one reaches the spot where the sun goes down”.” Translated and adapted from Chavannes (1907), p. 185, n. 4.

20. Wuyi [Wu-i] = Kandahar and/or the kingdom of Arachosia of which it was the capital. See note 8.5 above.

21. Haibei 海北 [Hai-pei], literally: ‘North of the Sea,’ must refer to the lands between Babylonia and what is now Jordan and/or Syria. See note 10.12 above and also Graf (1996) – especially the section on ‘The Western Regions’ on p. 204 and the map at the end of his article.
          However, Graf (1996), p. 204, argues that the use of the terms Haixi
海西 [Hai-hsi], Haibei 海北 [Hai-pei], and Haidong [Hai-tung] indicate: “that the Chinese of the Han era were ignorant about the existence of the Arabian peninsula. For them, the great sea adjacent to the Persian coasts stretched westward forming an immense bay that extended all the way west to the coasts on Ta-ch’in. Their belief in this imaginary body of water resulted in the creation of the three coastal districts.”
          I cannot agree with this analysis. It is true that the Chinese, like the Romans, and the Greeks before them, considered the Indian Ocean and its two major Gulfs, the Red Sea and the Arabian (or ‘Persian’) Gulf as a whole. The Greeks referred to it as the Erythraean Sea. This is perfectly reasonable and accurate, as easily navigable entrances join all the waterways. Because the Chinese accounts do not mention the Arabian Peninsula does not mean they were necessarily ignorant of it.
          In my view the Chinese division of these regions makes excellent sense. Thus we have: ‘West of the Sea’ (= Egypt); ‘East of the Sea’ (the lands on the east coast of the Persian Gulf) and, finally, ‘North of the Sea,’ the region in between and joining them: (probably northern Saudi Arabia, Jordan and southern Israel). See also the discussion under Heixi in note 10.12. The Weilue mentions an overland route through Haibei from Parthian territory to Egypt (Haixi):

“Now, if you leave the city of Angu (Gerrha?) by the overland route, you go due north to Haibei (‘North of the Sea’), then due west to Haixi (Egypt), then turn due south to go through the city of Wuchisan (Alexandria).”

This would appear to be an alternative to the long route up the Euphrates and through Palmyra and Dura Europa, then to turn south, and later west, to Egypt. There were two rather more arduous, but shorter and more direct alternatives. One route ran west from the head of the Persian Gulf across the desert to the oasis of al-Jawf (Dumatha). Here the road forked, one could either head north up the Wadi Sirhan towards Damascus, or west towards Petra, Rhinocolura, and Egypt. It seems these routes were guarded by Roman patrols after their annexation of Nabataea in 106 CE. Bowersock (1996), pp. 157-159; Millar (1993), pp. 138-139.
          Another route left the region of the prosperous trading state of Gerrha in eastern Arabia and travelled across the peninsula either to al-Jawf or to the Nabataean city of Taima and on to the port of Leuke Kome
(literally, ‘White Village’), which has still to be located exactly, but it is probably to be located in the vicinity of modern ‘Aynūnah, in the bay to the southeast of the entrance to the Gulf of Aqaba. It is probably the sane as the town called Xiandu 賢督 [Hsien-tu] in the Weilue.

“Much of the merchandise of the Orient was brought overland from the port of Gerrha on the Persian Gulf to the Arabian port of Leuke Kome on the east side of the Gulf of Aqabah and then shipped or transported by caravan northward to Aila. From there it was carried to Petra, to which a direct, overland route led also from Meda’in Aleh in Arabia. And “thence to Rhinocolura (modern el-Arish in Sinai on the Mediterranean) . . . and thence to other nations,” according to the Greek geographer Strabo, who wrote about the Nabataeans at the beginning of the first century A.D.” Glueck (1959), pp. 269-270.

“For this trade [with Elymais and Karmania] they opened the city of Carra [Gerrha] where their market was held. From here they used to set out on the twenty-day march to Gabba and Syria-Palestine. According to Juba’s report they began later for the same reason to go to the empire of the Parthians. It seems to me that still earlier they brought their goods to the Persians rather than to Syria and Egypt, which Herodotus confirms, who says the Arabs paid 1,000 talents of incense yearly to the kings of Persia. Juba (c. 25 BC-AD 25) and Pliny, NH (AD 77) 12. 40. 80).” Potts (1990), p. 91.

“The merchants of Palmyra were also active in Egypt. One group resident in Coptos was engaged in the commerce of the Red Sea and thus by implication possibly also with India and East Africa. Others used the overland route from Mesopotamia to Denderah in Egypt.” Raschke (1976), p. 644.

22. Haixi 海西 [Hai-hsi] – literally: ‘West of the Sea’ = Egypt. See notes 10.12 and 12.21 above.

23. This passage has caused some confusion to modern commentators. A ting [t'ing] in China was basically a shed or simple lodge for travellers to stop at, which I have called a ‘stage,’ and a zhi [chih] was a ‘postal station’ or inn that could provide not only shelter, but fresh horses, food and supplies.
          This is similar to the Roman postal system along the main routes of having mansiones or stationes (inns) every 25 to 35 Roman miles (37 to 52 km) – though closer in densely populated areas, with one or two mutationes, simple hostels or ‘changing-places,’ in between. Casson (1974), pp. 184-185. The only difficulty here is the short distances recorded between the ting (ten li or 4.2 km) and the zhi (30 li or 12.5 km). Lionel Casson (1974): 182 says: “In China the post-stations were some eleven miles (16.3 km) apart, with two or more substations in between.”
          The Roman and Parthian systems of postal relays were developments of the famous Achaemenid system developed by Darius I circa 515
BCE. The road from Sardis to Susa was 2,475 km in length, and had 111 relay stations [i.e. one every 4 parsangs, or about one every 22 km]. At normal rates of travel, the whole could be covered in 90 days (average speed = 27.5 km/day). However, by changing mounts and couriers, over 350 km could be covered in a day, and messages could be taken from Sardis to Susa in seven days. From: Ciolek (2000). See also: Dandamayev (1994), p. 52.

24. There are references to this long feiqiao 飛橋  [fei-ch’iao] = ‘high,’ ‘rapid,’ ‘raised,’ or ‘elevated’ bridge, in several early texts including the Hou Hanshu, the Hou Hanji and the Weilue. The GR Vol. II, p. 598 gives two interpretations for this term: 1. A raised bridge or foot-bridge (crossing a valley). 2. A floating bridge.
          For discussion of the various theories and arguments I refer the reader to the discussions in Hirth (1885), pp. 187, 192 ff; Leslie and Gardiner (1996), pp. 191-193; and Graf (1996), pp. 205-206.
          Instead of revisiting these arguments here I would like to make a totally new proposal – that it referred to a raised road or highway that crossed the Nile floodplains between the Egyptian border town of Pelusium in the east and the delta cities further west.
          To begin with, I think it is time to get rid of the use of the term flying bridge (first used by Hirth, and repeated by most writers since) to translate the term feiqiao. Flying bridge suggests some unusual or exotic type of bridge whereas, in fact, feiqiao is a common and standard Chinese term for a high or elevated bridge.
          All the texts remark on the great length of the bridge. The Hou Hanshu says it is “several hundred li long,” the Weilue says it was 230 li (96 km), while the great 13th century encyclopaedist, Ma Duanlin, records it as being 240 li (100 km). Previously this great length has always been considered to be either a gross exaggeration or a simple mistake. I suggest, however, that the “bridge” must have been unusually long or it would not have been mentioned at all. I have never come across any other reference to the length of a bridge in the Chinese histories; it is certainly most unusual. Also, there is no qualification by the historians, as one would expect if they were just repeating some story, such as “it is said” It is clearly stated as a fact in all the texts.
          Clearly, no conventional ‘bridge’ could be so long. One can only assume that it was some sort of raised road to take traffic across a wide expanse of water – not a ‘bridge’ in the usual English sense of the word, but something more like a ‘highway;’ a roadway raised above the surrounding country.
          Unfortunately, where this long ‘elevated bridge’ was situated is very difficult to establish because the location of the countries it is said to have joined are still being disputed by scholars.
          The Hou Hanshu has it leading from Haixi (which I identify as Egypt) to Haipei (‘North of the Sea’), which I identify as the lands stretching between Egypt and the head of the Persian Gulf. The Roman dependencies mentioned in the corresponding passage from the Weilue are very difficult to confidently identify – so I have qualified my suggestions with question marks:

“The king of Zesan (Azania?) is subject to Da Qin (Rome). His seat of government is in the middle of the sea. To the north you reach Lüfen (Leuke Kome?). It can take half a year to cross the water, but with fast winds it takes a month.
          Leaving from the city of Lüfen (Leuke Kome?) you head west to go to Da Qin, crossing over the sea by an ‘elevated bridge’ 230 li (96 km) long; then you take the sea route southwest, travelling around the sea (coast), and then head due west (to reach Da Qin).”

This route from Lüfen to Da Qin probably refers to the section of the ancient caravan route from Judea to Egypt along the isthmus between Pelusium and Pi-Ramses/Qantara Sharq, which separates Lake Manzala from Lake Balah, and then on to Tanis and Alexandria.

“The great Roman coastal road from Tangiers to Alexandria (in Egypt) was, if one measured in a straight line, 2,100 miles long. Its last course lay along the menacing sands close to the sea, and the way was marked only by cairns of stones although there were regular way-stops which sheltered water sources. The goal of this coastal road was, of course, Alexandria.
          There were five principal roads in Egypt. The first was the coastal road, which Strabo called ‘the way of the sea’. It crossed seven of the streams of the Nile and followed the coast to Palestine, Sidon, Tyre and Lebanon. Under Trajan after
AD 100 the road was rebuilt along the Mediterranean and extended from Alexandria to Antioch and along the serrated outer edge of Anatolia to the Bosporus.” von Hagen (1967), p. 106.

          There was a very ancient and important trade route from Rafah via Pelusium and Qantara Sharq into Egypt known as the Horus Military Road or “Road of Kings.” The Weilue states that from Lüfen [Rafah?] you travel west “crossing over the sea” over a 230 li (96 km) long “elevated bridge” [or causeway] and then go by sea to reach Da Qin.
          The main route leading from Pelusium to the crossing at al-Qantara (‘the bridge’) and on to Tanis would have, at that time, skirted the southern bank of Lake Manzala. This road must have been raised to keep it dry and passable. It is likely that the main road was along top of the embankment beside the Butic canal which ran, in Roman times, south of Lake Manzala, in practically a straight line directly west from Daphnae across the delta lands via Tanis to Sebannytos (at the Damietta branch of the Nile), and then on to Lake Mariut, thus connecting with Alexandria. The section from Daphnae via Tanis to Sebannytos would have been just about exactly 96 km
          Professor Eric Uphill very kindly responded to a query of mine, and sent (on 30 March 2003) not only his personal comments on this matter, but also an article of his on Egyptian canals – see Uphill (1988) – and some abstracts of presentations he made at the “Eighth International Congress of Egyptologists – Cairo 2000 from which is the following quote from “The Butic Canal: Its Date and Functions,” pp. 186-7:

“This artificial waterway had an estimated length of 180 kilometres, and starting from Tell Defenneh [Daphnae], connected eleven Lower Egyptian nomes on or near its route. Inscriptional evidence suggests it was created by, or else completed under King Psamtek I (664-610 BCE). Among its varied uses, it could have served to transport grain and commodities by boat, and help irrigate lands on either side of it. In addition it could also have been used for moving troops as was done by Emperor Titus [reigned 79-81 CE]. At a time of military threat by the world power Assyria, a major canal protected by and communicating with Greek and Egyptian troops at Marea in the west and Daphnae in the east would also clearly serve as a first line of defence for the Saite rulers.”

          Professor Uphill suggested, reasonably enough, that where there were canals merchants would tend to have used them to cart their goods rather than the embankments beside them, due to the relative cheapness of water transport.
          However, my feelings on this are that the routes reported to the Chinese are obviously ones provided by merchants carrying goods from the region of the head of the Persian Gulf into Egypt. As such, for most of the way, the goods would have been carried by pack animals and it may have proved cheaper and/or more efficient to continue take these laden animals directly into the delta cities along the canal embankments rather than off-loading them onto watercraft and paying others to carry them the rest of the way. They would have also probably wanted to take their wares personally to the main market centres to get the best possible prices for them.
          Lake Manzala extended considerably further south during the Roman era than it does now, and the Pelusiac branch of the Nile still emptied into the Mediterranean, with its mouth near Pelusium.

    “The roads [in the “Land of Goshen” – i.e. the fertile lands along the easterly branch of the Nile and immediately to the east of it] are usually high above the rest of the country. They run along the canals, and consist of the dirt banked up to hold back the waters. The side roads are chiefly camel paths or foot paths, and one sees everywhere the traffic moving along through the fields. Even on the main roads there are few wagons. Most of the freight is carried on donkeys and camels, which are the common riding animals as well.” Carpenter (1928), p. 8

One can well imagine that when the Nile flooded it was only the main highways between the cities that could be traversed and they could accurately be described as long bridges “across the sea.”

“But major roads, perhaps military in origin, developed to connect some of the main towns and there were local paths which might not be useable during and just after the time of the inundation: ‘If the roads are firm I shall go off immediately to your farmer and ask him for your rents, if indeed he will give them to me,’ writes an anxious woman from Oxyrhynchus around the year 200.” Bowman (1986), p. 153.

“Now the Nile, when it overflows, floods not only the Delta, but also the tracts of country on both sides of the stream which are thought to belong to Libya and Arabia, in some places reaching to the extent of two days’ journey from its banks, in some even exceeding the distance, but in others falling short of it.” Herodotus (5th cent. BCE): 124 (II.19).

“When the Nile overflows, the country is converted into a sea, and nothing appears but the cities, which look like islands in the Aegean. At this season boats no longer keep the course of the river, but sail right across the plain. On the voyage from Naucratis to Memphis at this season, you pass close to the pyramids, whereas the usual course is by the apex of the Delta, and the city of Cercasorus. You can sail also from the maritime town of Canobus across the flat to Naucratis, passing by the cities of Anthylla and Archandropolis.” Herodotus (5th cent.
BCE): p. 155 (II.97). Note 205 by George Rawlinson; ibid. on p. 212, says: “This still happens in those years when the inundation is very high.”

The extensive canal system and associated roads required a huge investment of time and work.

            “But as a first step, the floodwaters had to be directed to where they were most wanted or where they could not reach without human assistance. to that end an irrigation system was first engineered under the Pharaohs, and expanded under the Ptolemies. The system was most extensive in the broad basin of the Arsinoite nome. Under the feeble and financially strapped government of the last Ptolemies Much of the system fell into disrepair for want of proper maintenance. Octavian, determined to make his newly acquired province a reliable source of grain to help feed the city of Rome, set his soldiers as well as locally conscripted labour gangs to the work of rebuilding dikes and clearing the cluttered canals as speedily as possible. Regular maintenance was assured thereafter through the imposition of a corvée: each year every able-bodied adult male who worked the land, plus his slaves if any, was required to perform a certain quota of unpaid labour on the dikes and canals. The normal stint was five days a year, but additional days were imposed when exceptional conditions necessitated it. In Upper Egypt, and sometimes in other parts, the performance of the corvée was measured in quantities of earth moved rather than in time worked.” Lewis (1983), pp. 111-112.