Jump to ContentJump to Main Navigation
Sino-French Trade at Canton, 1698-1842$

Susan E. Schopp

Print publication date: 2021

Print ISBN-13: 9789888528509

Published to Hong Kong Scholarship Online: September 2021

DOI: 10.5790/hongkong/9789888528509.001.0001

Show Summary Details
Page of

PRINTED FROM HONG KONG SCHOLARSHIP ONLINE (www.hongkong.universitypressscholarship.com). (c) Copyright Hong Kong University Press, 2022. An individual user may print out a PDF of a single chapter of a monograph in HKSO for personal use. Subscriber: null; date: 27 June 2022

The Voyage and Sea Routes

The Voyage and Sea Routes

(p.55) 4 The Voyage and Sea Routes
Sino-French Trade at Canton, 1698-1842

Susan E. Schopp

Hong Kong University Press

Abstract and Keywords

Chapter 4 focuses on the voyage and sea routes that figured in France’s Europe trade with China. A knowledge not only of ship handling but also of wind systems was essential to a vessel’s safe arrival at her destination; so was the ability to deal with navigational hazards at a time when many regions remained to be accurately charted. The timing and duration of a voyage were also affected by the number of ports of call, the length of stay at each, and the route taken. Mariners were guided by sailing instructions such as d’Après de Mannevillette’s Neptune oriental, which described routes, winds, currents, navigational hazards, landmarks, port entrances, and more. While the majority of ships sailed via the Cape of Good Hope, the French also made use of routes via Cape Horn, both in the years 1706–1717, when they combined trade at Canton with that of supplying Spanish colonies on the Pacific coast of South America, and in the nineteenth century. Closer to China, the increasing use of alternative routes freed vessels from restrictions imposed by monsoons in the South China Sea, enabling ships to arrive year-round and ultimately lessening the control of Chinese authorities over international trade.

Keywords:   timing of voyages, navigation, sea routes, sailing instructions, ports of call, Cape of Good Hope, Cape Horn, Pearl River, pagodas and sight lines

Over the course of the near century-and-a-half (1698–1842) of French participation in the Canton Trade, at least 265 French merchant vessels called in China. The great majority were Europe ships, departing from, and returning to, France. A much smaller number served the country trade.1 Detailed sailing instructions spelled out the sea routes from Europe to Asia and back, as well as ports in between, and provided information on such topics as winds, currents, and tides; descriptions of landmarks and details of navigational hazards; and locations for obtaining fresh water and provisions, to name just a few.

“When sailing from Lorient or from one of the other ocean ports of France, you must first shape your course so as to pass about twenty-five or thirty leagues from Cape Finisterre. This distance is sufficient in any season whatsoever; you can even double that cape at a closer distance, depending on the circumstances; but from its latitude you will always steer for the island of Madeira.”2

So began the voyage to China. Depending on the era, the conditions encountered en route, and the number of ports of call and length of stay at each, the one-way voyage to China via the Cape of Good Hope could take as little as five-and-a-half months (though six to nine months was more typical), and the round trip, from exactly one year to as long as slightly over four. Via Cape Horn, the one-way voyage could take from eighteen months to over four years, and the round trip, from slightly over four-and-a-half years to nine years. In the eighteenth century, a round-trip voyage of about a year and a half via the Cape of Good Hope, with a stay in China that varied from two to five months, was typical. In the nineteenth century, the stay in China could be as short as three to four weeks, though two months was not unusual, and both the Cape of Good Hope and, to a lesser degree, the Cape Horn routes were used. (See Appendix 1.)

The route as well as the timing of a voyage was heavily influenced by the earth’s wind systems, which determined the directions in which the ships could—or could not—sail. As one of the eighteenth-century books of sailing instructions put it, “it is essentially necessary for the navigator to have a thorough knowledge of the direction of winds as chiefly prevail through the whole extent of those seas which he is obliged to pass over.”3

(p.56) For ships sailing via the Cape of Good Hope, besides the westerlies, the trade winds, and the infamous doldrums (the intertropical convergence zone) of the Atlantic, there were the monsoons to reckon with. The southwest monsoon, which blows from late spring to mid-autumn, carried vessels through the Indian Ocean and South China Sea to China, while the northeast monsoon, blowing in the opposite direction from late autumn to mid-spring, took them back through those same waters. It was well not to sail too close to the end of the monsoon season, however, for the transitional periods between monsoons—the months of May and November—were marked by unstable winds, whose strength and direction were subject to unpredictable variation, and were accompanied by storms and typhoons.4

There was also the matter of ports of call and the length of stay at each, as well as of mastery of the sea routes. Of the twenty-seven departures from Europe during the era of the first Compagnie (1698–1719), more took place in January (seven) and March (five) than in the other months of the year, which saw between one and three departures each. Between 1720 and 1769—that is, under the second Compagnie—there were eighty-eight departures from Europe, the greatest number of which took place in December (twenty-eight) and January (fourteen), followed by November (twelve), February (ten), October (eight), March (seven), April (six), and June, August, and September (one each). Over the decades, the date of departure gradually grew later; eight of the ten in February occurred between 1762 and 1769. From 1770 to 1792—that is, from the beginning of the first open-trade era to the last French Europe voyage of the eighteenth century—the favored month of departure for the sixty-three vessels was later still: March. Of the fifty-three whose month of departure is known, the majority took place between January and March: eleven in January, five in February, and sixteen in March. April saw nine; May, one; July, two; October and November, one each; and December, seven.5 (See Appendix 1.)

In the years of the Compagnie, vessels’ arrivals at, and departures from, China were closely tracked by the Canton Conseil, which was responsible for keeping the Compagnie’s directors in France informed of the vessels’ movements. Of the twenty-one Europe vessels whose month of arrival during the era of the first Compagnie (1698–1719) is known, the greatest number (eight) arrived in July; this was followed by March (five), June (three), September (two), and May, November, and December (one each). Among the three country vessels, one arrived in July; a second, by late July; and a third, in September. (See Appendices 1 and 2.)

During the period 1720–1793, the great majority of the 171 arrivals in China, including both Europe and intra-Asian vessels, took place in the summer and early autumn. During the era of the second Compagnie (1720–1769), ninety (85.7%) of the 105 arrivals occurred between July and September. August saw the greatest number (thirty-six), followed by July (twenty-eight) and September (twenty-six). The concentration shifted slightly during the first era of private trade (1770–1785); of the 53 arrivals, the most took place in September (twenty-four), followed by August (fifteen) and then October (ten), enabling these three months to account for 92.4% of the total. The concentration changed again, slightly, in 1786–1792, during the third Compagnie’s existence; September saw six arrivals; October, four; and November, two.

In 1802–1803, there were three arrivals: one each in April, August, and September. The pattern changed noticeably, however, between 1818 and 1842: vessels arrived year-round. Of (p.57) a total sixty-three arrivals, the month is known for fifty-six of them. Thirty-six (64.2 %) fell between July and November: five in August, seven in September, ten in October, and eight in November. December, January, and February saw six each; together they constituted 32%. The remaining eight were divided among March (three), April (two), May (one), and July (two). (See Appendix 1.)

There existed alternative routes through the South China Sea that freed sailing vessels from the restrictions imposed by the monsoons; one such route, inaugurated by the 1740 Jason, took ships across the South China Sea, up the west coast of the island of Luzon in the Philippines, and then back across the South China Sea to Macau.6 The search for alternative routes increased from the 1750s onward, with significant consequences for the Canton trade as a whole, as ultimately, vessels became able to arrive at virtually any time of year.

For the return voyage to Europe, the timing of departures from China likewise reflected the influence of the monsoons. The celebrated eighteenth-century hydrographer and former Compagnie ship captain Jean-Baptiste-Nicolas-Denis d’Après de Mannevillette, whose 1745 Neptune oriental became the bible of sailing instructions for much of the century, advised vessels to depart China between mid-November and mid-February.7 There were also winds off the coast of South Africa to consider. It was well to double the Cape of Good Hope no later than mid-May, when the winds turned contrary. D’Après de Mannevillette made the reason clear: “Although ships can easily at all seasons double the Cape of Good Hope, coming from the westward into the eastern seas, they find not the same facility at their return. The winds from W.N.W. to S.W. that rage violently about this cape in the months of June, July, and August, and more frequently than at any other period of the year, expose the ships which attempt to double it at that season, to the loss of a great deal of time in tacking, besides the unavoidable accidents which almost always attend a long continuance at sea in tempestuous weather.”8

Between 1720 and 1755, all but one of the sixty-one departures of Europe ships occurred during the three months of November (fifteen), December (thirty), or January (fifteen). During those same years, more than half the ships arrived back in France during the month of July (thirty-one, or 56.3%); the next closest month was August, with eight. Between 1758 and 1792, the eighty China departures for Europe for which the month is known were concentrated in December (twenty-two, or 27.5%) and January (forty-nine, or 61.2%). The remaining nine occurred in February. Among those vessels, the month of return to France is known for sixty-one. July saw twenty-seven (44.2%) returns, while the next closest month, June, saw twenty (32.7%). (See Appendix 1.)

Sea Routes

During the seventeenth and eighteenth centuries, the majority of East India ships heading for China sailed via the Cape of Good Hope, a route that is well known in scholarship on the East India and China trade. The French were no exception—at least from 1698 to 1702 and after (p.58) 1720. Less familiar, however, is the route via Cape Horn, which is heavily associated with the French, by whom it was favored between 1706 and 1715. Cape Horn also figured in certain French voyages in the nineteenth century, though with additional routes.

The early use of the Cape Horn route was directly linked to Saint-Malo navigators and their trading voyages to the Pacific coast of South America, and to the practice, by the first Compagnie des Indes, of leasing its China-trade monopoly to private merchants, many of whom came from Saint-Malo. The voyage across the Pacific to China was an extension of this route; vessels typically called first at trading ports in Chile and Peru, especially Concepción, Arica, Pisco, Callao, and Guacho (Huacho), before heading for Asia. The chief disadvantage to the route was its length, for between South America and Guam, Europeans knew few, if any, ports of call, and there was thus little to no opportunity to repair or to re-provision.

The sea route (including ports of call) that a given vessel was to follow was stipulated in the years of the second French East India Company by the Compagnie itself. Only in the case of unforeseen and exceptional events was a captain permitted to deviate from the route set out.9 Similarly, the Compagnie’s conseil at Canton was not allowed to alter either the route or the date of departure from China. Exceptions were permitted only if, in addition to unforeseen and exceptional events, a change was deemed to be an absolute necessity.10 The Compagnie also regulated the duration of a vessel’s stay at ports of call, limiting it to the time required to replenish supplies of water and wood for cooking; to take on provisions; and to provide the crew with an opportunity to rest and, in case of illness, to regain their health. Delays were permitted only if departing as instructed would pose a serious risk to safe navigation.11

To follow the sailing instructions that detailed the routes, the captain of a vessel had to know his position at sea. In 1700, ascertaining latitude was not a problem, but accurately determining longitude remained elusive until marine chronometers, which were invented in the eighteenth century, became commercially available; they came into widespread use only in the 1800s.

The inability to correctly identify one’s position at sea was a major cause of grounding and shipwreck, for in uncharted or inaccurately charted waters, vessels ran the risk of running aground or striking submerged reefs and other often unseen obstacles. So important did the ability to accurately determine longitude become that both France and England offered prizes for a solution. In 1714, Rouillé de Meslay, a councilor in the nation’s chief judicial body, the Parliament of Paris, left a bequest to fund prizes, a number of which were pertinent to navigation.12 Several months later, the British Parliament passed the Longitude Act, which offered a prize “providing a Publick Reward for such Person or Persons as shall discover the Longitude at Sea.”13

In open ocean, as well as when approaching or departing land, mariners were dependent upon previous experience and on charts, which were often dangerously incomplete.14 The (p.59) frequent inclusion in sailing instructions of reports from sea captains detailing positions of reefs and islands reflects the concern for an improved knowledge of the world’s oceans. In the absence of an accurate mechanical means of determining longitude, navigators relied on methods of celestial navigation, which requires measuring the angle between the visible horizon and the sun, moon, or certain stars—and therefore necessitates clear skies.15 A second, older method, that of dead reckoning, required no reference to celestial bodies; instead, a ship’s current position was calculated by estimating the direction and the distance that she had traveled from a previously determined position. But the results were approximate at best and disastrous at worst, for determining the vessel’s direction and speed, which were significantly affected by winds and ocean currents, was highly problematic. The opportunities for error were manifold, and the greater the distance traveled, the greater the likely increase in error. In both celestial navigation and dead reckoning, inaccuracy could prove tragic.16

Navigational hazards took a variety of forms. Shoals and reefs were often completely submerged or only periodically visible, and the dangers they posed were compounded by incomplete or inaccurate charts. In the eighteenth century, many regions of ocean were still uncharted. D’Après de Mannevillette cautioned, “The South China Sea is so full of reefs and shoals that the greatest precautions are always necessary, even in the busy routes; a ship can hardly depart from the ordinary routes without being almost certain to see some danger.”17

The chief means of updating and correcting charts came from first-hand accounts of sightings of both known and previously unknown hazards, as well as from observations of land formations, approaches to ports, and a host of other elements, much of this gained from information recorded in ship’s logs.18 But observation was not the only means of adding details to, or correcting mistakes in, a chart. Previously unknown hazards were often discovered only when a ship struck them, leading to more than one of the reefs and shoals in the South China Sea becoming known by the name of a vessel that approached too close. In describing navigational hazards in their books of sailing instructions, hydrographers often mentioned these vessels—and the type of damage that they sustained—as a warning to mariners. Explaining to his readers how to avoid Shahbunder Shoal off the coast of Sumatra, for example, hydrographer James Horsburgh mentioned not one but three unfortunate encounters:

SHAHBUNDER SHOAL, named from a Dutch ship that narrowly escaped being lost on it, lies about 7 miles W. by N. 1/4 N. from the South Brother; but it is extensive, formed of various patches, and seems to be the outer extremity of the shoal bank that projects along, and far out from this part of the Sumatra Coast. The French ship Jupiter, returning from China, grounded and had part of her keel broken off upon this shoal. The Sandwich grounded on one of the patches, returning from China in January, 1749, by borrowing too near the coast.”19

(p.60) Horsburgh’s interest in his subject was not merely academic; at the age of 24, he had survived the wreck of the English East India Company ship Atlas in the Indian Ocean on the island of Diego Garcia. But as his and other hydrographers’ sailing instructions show, a ship that struck a reef, ran aground, or experienced some other navigational misfortune became a source of information that, if correctly reported and incorporated into a chart, contributed to the chart’s accuracy and therefore to its value in preventing future tragedy.

On the Pearl River, the required use of licensed Chinese pilots greatly reduced the likelihood of disaster. The river’s hazards included a strong current and an occasionally shallow bottom. In entering the river near the Bocca Tigris, for example, there was a shoal flat having only 1½ fathoms of water (9 feet; 2.7 meters) in some places.20 In addition, the river was tidal beyond Canton, so its depth was affected by the ebb and flow of the tide. It was just as well that the East India ships anchored at Whampoa, for on the stretch of river between Whampoa and Canton, the volume of traffic increased the closer one approached to the city, “The river craft from the size of a little sampan, or canoe boat, to that of huge clumsy jonck [sic] vessels—from 300 to 1000 tons, increased in numbers as we neared the city; & from almost noiseless sounds—or stillness on the water, a humming like, buzzing tone gradually increased, occasioned by thousands of voices along, and on the river shores … which continued on the increase say from four or five miles below all the way up to the city.”21 Consequently, not only the hazards but also the tides were among the significant factors to be taken into account.

Sea Route via Cape Horn

Between 1698 and 1715, the Cape Horn route was used by nine ships for the round trip, by ten more for the outbound voyage, and by two for the homeward voyage.22 It thus figured in twenty-one of the twenty-seven Europe voyages of this period. (See Appendix 1.)

Vessels departing from Saint-Malo headed out to the Atlantic and then followed a course that led southward past the Canaries and the Cape Verde Islands. Past Cape Verde, the route took them across the Atlantic toward Brazil. Instead of making the southeasterly sweep back across the Atlantic toward the southern tip of Africa and the Cape of Good Hope, however, they continued southward along the Brazilian coast. Slightly beyond Rio de Janeiro, they could put in at Ilha Grande, where in addition to obtaining provisions, they could also replenish their supplies of firewood and fresh water. From Ilha Grande, they continued southward toward Cape Virgins, the promontory that marks the southeastern corner of continental Argentina.23

Cape Virgins was also the landmark for the bay leading from the Atlantic to the Strait of Magellan, the narrow, hazardous passage separating the southern edge of continental South America from the Tierra del Fuego archipelago. European vessels sailing around South America had used the strait since Magellan first passed through it in 1520. But its dangers, especially its strong winds and currents, were legendary. (p.61)

The Voyage and Sea Routes

Map 4.1: French sea route to China via Cape Horn, 1707–1718

Then, in 1616, the captain of the Dutch vessel Eendracht, Willem Schouten, and his supercargo, Jacob Le Maire, succeeded in finding a passage that was somewhat further south. Foregoing the Strait of Magellan, they passed between Tierra del Fuego and the much smaller Staten Island, through the strait that now bears Le Maire’s name. Then they doubled Cape Horn, sailed on to the Pacific—the South Sea or mer du Sud—and headed up the west coast of Chile before altering course at the Juan Fernandez Islands to sail westward across the Pacific.24

The new route was not without its own dangers; the hazards of doubling Cape Horn in the east-to-west direction have since become well known. La Barbinais Le Gentil, recounting the voyage on which he set out in 1714, describes how both the ship on which he was sailing, Vainqueur, and her crew were nearly all lost when their pilot mistook a “cul de sac” for the Le Maire Strait.25 Once in the Pacific, Vainqueur called at ports in Chile and Peru, including Concepción, Coquimbo, Callao, and Guacho [Huacho], where he disembarked and boarded the ship Jupiter for the voyage across the Pacific to Asia.26

“No voyage in the world is as long as that from Peru to China,” he declared of the seemingly interminable expanse of ocean between Peru and the southeast coast of China.27 The heat and air quality, which he found oppressive, weighed heavily on him, as did what he perceived as an endless, never-changing view: “During three months and nearly four thousand leagues, I saw but several birds and some fish.”28

(p.62) Jupiter’s long voyage across the Pacific was broken only by a stop at Guam, which offered a welcome respite in the long voyage. There the ship encountered three other French vessels: Martial, Marquis de Maillebois, and Bien Aimée. The four departed Guam in company, then separated at Cape Engano and the Babuyan Islands to its north. After sighting Taiwan, Jupiter proceeded to Amoy, while Martial, Marquis de Maillebois, and Bien Aimée went to Canton.29

The route homeward from China took vessels back across the Pacific and around Cape Horn, whose eastward winds and currents made for an easier passage than on the outbound voyage. After doubling the Cape, ships were likely to forego the Le Maire Strait in favor of passing to the east of Staten Island. They then proceeded either between Staten Island and the Falklands (les îles Malouines, named after the Malouins, as residents of Saint-Malo were known), or eastward of both Staten Island and the Falklands, before turning toward the South American continent and heading up its Atlantic coast. After rounding the northeastern coast of Brazil, they made their way back to France by one of several routes, either around the Azores or via one of France’s American colonies.30

The seven vessels that departed France between 1706 and 1710 made the round trip via the Horn: Patriarche, Saint-Antoine de Padoue, Deux Couronnes, Découverte, Princesse, Solide, and Eclair. So also did Martial, which departed France in 1713. Bien Aimée, departing in 1712, completed the return as far as the Pacific coast of Mexico, where she was captured. Twelve more vessels circumnavigated the globe. The circumnavigation could be made westward, in which case ships made the outbound voyage via Cape Horn and the return via the Cape of Good Hope, or eastward, via the Cape of Good Hope on the outbound voyage and then doubling Cape Horn on the return. The first circumnavigation westward by French merchantmen trading with China was made by four ships that departed France in 1711: Grande Reine d’Espagne, Grand Dauphin, Saint-Louis, and François.31 They were followed in 1713 by Notre Dame de Lorette, and in 1714 by Comte de Lamoignon, Pontchartrain, Jupiter, Marquis de Maillebois, and Grand Dauphin.32

The first circumnavigation eastward by China-bound merchantmen was completed by Comtesse de Pontchartrain, which departed France in 1714; she sailed via the Cape of Good Hope to China and returned home via Cape Horn. Her consort, Brillant, completed the circumnavigation as far as Arica in northern Chile, where she was seized in 1717 while on her homeward voyage.33

(p.63) Sea Route via the Cape of Good Hope

The better known of the sea routes from Europe to China, that via the Cape of Good Hope, was the route typically used by the French.34 Six of the twenty-seven voyages by French Europe ships to China during the period 1698–1719 were made by ships that sailed via this Cape round-trip. Between 1698 and 1702, Amphitrite made two voyages and Chancelier and Saint-François, one each, while more than a decade later (1714–1718), two more did likewise: Comte de Toulouse and Comte Amelot.35

The Good Hope route became the standard from 1720 onward, used not only by ships of the Compagnie des Indes but also, after 1769, by those in the private sector. After setting out from France and reaching the Atlantic, ships followed a southwesterly course that took them past Spain and Portugal, with a possible stop at Cadiz if they needed to buy Spanish piasters, and then on past Madeira, which they were advised to sight in order to confirm their position. They passed the Canaries and the Cape Verde islands, where they could stop if they needed to take on water, firewood, or provisions. Past Cape Verde, they headed toward Brazil to take advantage of the trade winds. Before reaching the Tropic of Capricorn, they altered course again near Trindade Island off the Brazilian coast, and then traced a broad southeasterly arc that took them across the southern Atlantic, past the island of Tristan da Cunha, and on toward the southern tip of Africa and the Cape of Good Hope. D’Après de Mannevillette advised against calling in Brazil, citing delay of arrival in China and exposure to disease as deterrents; if necessary, it was better, he counseled, to stop at the Cape of Good Hope.36

The route that vessels sailed after doubling the Cape depended on which of three routes they planned to take across the Indian Ocean. As The Oriental Navigator pointed out, “The course you are to take Eastward, after doubling the Cape, ought to be always in proportion to the distance of the places to which you are bound, so that when you leave the westerly winds, those from the S.E. and E.S.E. which you will meet with afterwards, might be fair winds for the course you intend to steer.”37

The first route, which was the least utilized by the French, led vessels up the southeast coast of Africa and through the Mozambique Channel between the continent and Madagascar before taking them across the Indian Ocean. The second route led across the Indian Ocean to the Mascarenes. The economy of the Isle de France, in particular, was directed toward the servicing of vessels and crews on long-haul voyages. Not only did the island offer the sheltered port of Port-Louis (not to be confused with the Breton Port-Louis near Lorient), but it had the added (p.64) advantage of being conveniently located for vessels trading at ports across the Indian Ocean as well as in China. This gave it appeal to shipowners and sea captains of a wide range of nationalities. In 1810, Félix Renouard de Sainte-Croix wrote, “Whether one comes from Copenhagen, Amsterdam, London, Bordeaux, Marseille, Boston, Mexico, Peru, the Philippines, China, Pegu, Bengal, Madras, the Malabar coast, the Persian Gulf, Moka or Mozambique, one can always get there [Isle de France] in four months.”38

After calling at the Mascarenes, vessels going directly to China steered a course that took them toward the islands of Amsterdam and Saint Paul in the southern Indian Ocean, roughly 3000 kilometers (1,864 miles/ 1,620 nautical miles) southeast of Isle de France. After passing the islands, they made for Sunda Strait, not quite 4,500 kilometers (2,796 miles/2,430 nautical miles) to the north-northeast. Once through the strait, they sailed along the coasts of Sumatra and the Malay Peninsula as far as Pulo Timon (Tioman Island) between the peninsula and the Anambas Islands. From Pulo Timon they steered for Pulo Condore (Côn Sơn) off the southern coast of Vietnam, and then for Pulo Sapata, which it was essential to sight before proceeding up the coast of Cochinchina (southern Vietnam), then west around the Paracels and south of Hainan, and on toward Macau.

Ships departing the Mascarenes and proceeding to India before heading to China sailed toward the north end of Madagascar, then toward the Seychelles and on through the Indian Ocean toward the Indian subcontinent and Sri Lanka. After leaving India, they passed through the Strait of Malacca and followed the course from Pulo Timon as noted just above.

The third of the three routes across the Indian Ocean was not affected by monsoons, but the absence of any ports of call, combined with the distances involved, put this route at a disadvantage. Vessels taking this route doubled the Cape of Good Hope and then crossed the southern Indian Ocean along the 37°30’ parallel to the islands of Amsterdam and Saint-Paul. After reaching these islands, they made for the Sunda Strait, following the route noted just above.

As vessels approached the Chinese coast and Macau, they passed among a number of islands. When Macau, studded with hills, came into view, newcomers were impressed by the city’s topography. In 1790, francophone trader Charles de Constant wrote, “The city is charmingly situated, rising on all sides from the sea like an amphitheater on several hills, atop which there are forts and churches.”39 He counted thirteen of the latter as well as three monasteries and one convent, about which he commented, “Most of the churches are large and well-built; Saint-Paul, which belongs to the Jesuits, has a handsome façade with [sets of] three stone columns; they are richly ornamented.”40

At Macau, in accordance with the regulations of China’s Maritime Customs, ships were required to take on a licensed Macau pilot.41 The navigational hazards of the river, with its (p.65) tidal waters, shifting shoals and often shallow depths, made such services necessary for the safe voyage up and down the river.42 The deeper draft of European ships made the possibility of running aground an acknowledged danger. Requiring pilots not only avoided shipping disaster, however. Thanks to the record-keeping that it entailed, it also provided the Chinese authorities with a means of identifying, tracking, and regulating the movement of non-Chinese vessels and people on the river.

While at Macau, in addition to beginning the customs formalities, a captain might find out what other vessels, whether French or not, had gone upriver. “I learned … of the arrival of our two vessels the [Duc de] Choiseul and the Paix, the latter just a week ago, and that a number of vessels from various nations had already gone upriver,” noted Captain Joseph Bouvet of the ship Villevault on September 25, 1765, the day that he anchored at Macau.43 Bouvet was an experienced captain; he had already made eleven voyages for the Compagnie, including one to China in 1752–1754 commanding Duc de Chartres.

The voyage upriver typically took several days, and was made in stops and starts, depending on the winds and the tide. As Captain Bouvet’s log shows, Villevault arrived at Macau and dropped anchor on Wednesday, September 25, 1765. At 7 a.m. the following day, with the pilot on board and the tide favorable, she weighed anchor. In company with her was a second French ship, Beaumont, which had taken on a pilot in Macau at the same time as Villevault.44

From Macau, visitors heading to Canton sailed up the Pearl River estuary to the Bocca Tigris (“Tiger’s Mouth;” 虎門‎ fu2 mun4, Humen “Tiger Gate”), which was known to the French as the Bouche du Tigre and which marked the entrance to the river.45 En route, they spotted fishing stakes, a sight that would often be repeated. Described as “heavy timber spiles—driven into the bottom of the river in long ranges—same as are wooden bridge spiles—at home,” those just above Lintin Island (伶仃島‎) extended from the two sides of the shipping channel to the opposite shores.46 Fishing stakes also often served as navigational aids, as when they marked out the end of a deep channel.

As ships drew near the Bocca Tigris, they often encountered calms.47 While this necessitated the use of boats to keep a ship moving in the right direction, it also offered voyagers an additional opportunity to survey their surroundings.48 The passage, described by different visitors as one-and-a-half to three kilometers (one or two miles) wide, was flanked on both sides by two high points of land. Between them, in the middle of the passage, were two small islands (p.66) called the Wang-tongs (or Wantongs; 橫檔‎).49 Second Bar Pagoda, one of the several nine-story pagodas that served as landmarks during the voyage, could be seen in the distance.

The natural features of the strait tended to attract less attention than did the customs post on the northern Wang-tong, which was described as a “small island with some trees,” and the redoubt and fort at Anunghoy on the eastern side of the channel.50 A little beyond the Wang-tongs lay Tiger Island, “an immense mass of granite covered with short grass” that was estimated to be about 300 feet high.51 In export paintings, the strait was generally depicted from the south, showing ships and small craft approaching or passing through the just-mentioned channel in both directions.52 Like the rest of the river up to Whampoa, the area was also the focus of considerable attention in sailing instructions because of the navigational hazards posed to the deeper-draft East India ships—in this case, especially by shoals and rocks.

Past the Bocca Tigris and Tiger Island, vessels continued on toward Whampoa, a distance of approximately forty kilometers (twenty-five miles).53 Soon the hills gave way to a landscape that few newcomers could resist describing.54 Pierre Poivre, who first arrived at Canton in 1741, wrote:

I entered the river of Canton; it is peopled like the land; its banks lined with ships at anchor; a prodigious number of small craft are continually gliding along in every direction, some with sails, others with oars, vanishing often suddenly from the sight, as they enter the numberless canals, dug with amazing labor, across extensive plains, which they water and fertilize. Immense fields, covered with all the glory of the harvest, with stately villages rising to the eye on every side, adorn the remoter view, whilst mountains, covered with verdure, cut into terrasses [sic], and shaped into amphitheatres, form the back ground [sic] of this noble landscape.55

Past the Bocca Tigris, the land was punctuated by a network of waterways feeding into the river and creating various low islands of irregular size and shape. For the most part, this land was flat, “with occasional and very abrupt hills,” and was devoted to the cultivation of rice.56

(p.67) In the course of the growing cycle, the scene was transformed into an expanse of green. When the water was at its highest, the fields near the river looked more like large lakes than fields.57 Further along the west side, the paddies were interspersed with fields devoted to crops other than rice.

Besides the rice paddies, small villages and “an infinite number of pagodas” could be seen scattered across the landscape.58 The pagodas were not all the same height, though more than one observer pointed out that they always had an uneven number of stories. Ship captain Auguste Duhaut-Cilly, describing two of the tallest ones, observed, “They are of nine stories, and if we allow twenty feet to the story we must conclude that their total height is about 180 feet. To judge by the deterioration of the materials that form them they must be quite old, which shows that they were constructed with as much skill as elegance. Whatever the purpose for which they were destined, it seems that their usefulness has come to an end; for a long time they have been abandoned monuments.”59

Ship captains and their crews valued the pagodas close to the river both as landmarks that alerted them to navigational hazards and as points of reference for sight lines.60 On the way upriver, ships passed first by the Second Bar Pagoda, which was called by the French Tour du Lion after its Chinese name.61 Charles de Constant commented, “It is on the top of a mountain and is solidly built of bricks; it is very high, nine stories, and is not inhabited…. Towers like this can be seen all over China; they are located here and there, and always in view of each other. Every city has one; their height serves to mark the rank of the city; the first-rank ones have nine stories, the second-rank have seven, and the villages, five. The pagodas, which are spread throughout the country, are used as signal towers.”62

The pagodas typically sported an array of plant life. Somewhat more than halfway from Second Bar to the Whampoa anchorage stood First Bar Pagoda (Tour d’Amour). “Large plants, and even small trees are rooted and growing on several of the basements of the stories—to the very topmost of them, the seeds of which were probably planted there by winds and birds. It is evidently an ancient monument, & is a most venerable looking object, standing on conspicuous and elevated ground, surrounded at the base by low buildings—shaded by lofty spreading trees, and having the appearance of religious temples.”63

(p.68) Whampoa Pagoda, which was situated on Whampoa Island, was covered with brushwood and had trees growing in the crevices of the eaves that separated the stories.64 Like the First and Second Bar pagodas, it, too, had nine stories, and its conical roof was said to be nearly an additional story high.65 The pagoda was visible from First Bar, which was five-and-a-half kilometers (almost three-and-a-half miles) from the Whampoa anchorage, and it was used by the French and other European and American navigators as the mark for clearing First Bar’s lower and upper shoals.66

For the East India ships, whose numbers increased over the decades as the trade grew, arrival at the Whampoa anchorage marked the end of the outbound voyage. The anchorage was bounded on the north by Whampoa Island, and on the south by Danes and French islands. Near the southwestern corner of Whampoa Island was the large walled town of Whampoa, described around 1830 as having a number of buildings that were large enough to be seen at a considerable distance, while “numerous flag-staffs belonging to the mandarin houses, and temples, are seen above the trees.”67 The area’s significant population included compradors, stevedores (longshoremen), blacksmiths, and others whose work, either directly or indirectly, was connected to the international shipping.68 Nearby were large boatyards.69

Vessels arriving at Whampoa from Macau approached the anchorage by passing between Junk Island and the northern side of Danes Island. The largest East India ships anchored at the lowest (easternmost) part of the roadstead. It was customary to downrig after arriving, so a number of paintings show only the lower masts remaining in place.70 Wherever the Western ships anchored, however, they constituted only a small part of the river traffic. “Large cargo-boats and junks, some of them highly adorned, are seen winding their way with great skill between the Indiamen, while the whole surface of the water appears covered with an infinity of small craft, paddling about in every direction.”71

Along the shore of Whampoa, Danes, and French islands could be seen the temporary structures that were known as bankshalls.72 Built of bamboo and covered by mats, they served (p.69) as storage-workshop-farmhouse-sickbay facilities for East India ships and their crews, and one of the first tasks to undertake after a vessel anchored at Whampoa was making arrangements to have one erected.73 Colin Campbell, who arrived at Canton in September 1732, noted, “Upon our Arrival at Canton I set about immediately getting a Compradore for the ship (or a Person to provide the ships Company with Provisions while she is at Vampo [Whampoa]) & to aggree [sic] for a piece of Ground to build a Banksaal to unload the ships stores into.”74 Half a century later, Samuel Shaw, the first supercargo of the first American ship to trade with China, provided additional detail, with particular attention to the French:

All company ships, on coming to Whampoa, have each a banksall [bankshall] on shore, for the reception of their water-casks, spars, sails, and all the lumber of the ship, and containing, besides, apartments for the sick. The French have theirs, separate from the other Europeans, on an island, thence called French island; the others are on the opposite side, and confined to the ground they occupy,—for the remainder being rice-fields, and constantly watered, renders it impossible to go beyond the limits of the banksall; whereas French island is a delightful situation, and the resort of the gentlemen generally, of all nations, who go on and off at pleasure. Except those of the French and Americans, no common sailors are allowed to go there. For the exclusive privilege of this island, every French ship adds one hundred taels extra to the hoppo’s present. The banksalls are large buildings, framed with bamboo reeds, and covered with mats and straw. They are erected by the Chinese, who pull them down immediately on their being left, in order that they may have the advantage of setting up new ones.75

As Shaw observed, the French paid an extra fee for the use of French Island (Ile des Français), which lay just to the west of Danes Island (Ile des Danois) across the narrow French River. In addition to providing space for bankshalls, the two islands were used for exercise and recreation by the crews of the East India ships and as burial grounds for both Chinese and non-Chinese.76 On French Island, indigo, cotton, and Chinese potatoes could be seen growing on the terraces that graced its slopes, while gardens yielded fruit that, as Swedish chaplain and naturalist Pehr (p.70) Osbeck observed, would be grown in Sweden as rarities in hot houses.77 Further down there was a Chinese burial ground, while the graves of several Europeans were visible on the lowest terraces.

For supercargoes, writers, and bakers, all of whom were housed in Canton, as well as for ship captains and officers who went back and forth as needed between Whampoa and Canton, the final part of their voyage was made up of the twenty kilometers (twelve miles) from Whampoa to Canton, and was dominated by boat traffic.78 Newcomers, in particular, were astonished at the sheer number of watercraft they encountered. Charles de Constant wrote, “The closer I got to the city, the more the number of boats increased, and soon our vessel had difficulty moving forward, even though the river was a mile wide.”79 River traffic included “salt junks discharging their cargoes into the canal boats, the vessels from the interior laden with wood, and immense rafts of timber and of bamboos, … revenue cruizers …, thousands of small ferry boats …, immense junks of four or five hundred tons, and even larger, moored in the stream,” and more.80 There was also the floating city, “the stationary boats that comprise another city on the river (divided into streets, squares, and avenues) where circulate constantly a great number of smaller boats that provide the inhabitants with all things needed and desired.”81

Some first-time visitors admitted that they did not believe the variety and profusion of river traffic that they had read about in earlier accounts until they saw it with their own eyes.82 “Although I had read several descriptions of this passage from Wampoa to Canton, and had considered most of them as containing exaggerations not altogether consistent with the truth, yet I must confess that those descriptions fall infinitely short of the surprizing scene through which we passed in a short voyage of about ten to twelve miles.”83

The return voyage began with the descent of the river, which was made, as in the ascent, in several stages, depending on the tides and also on whether the entire cargo could be loaded at Whampoa, or whether an additional stop at Second Bar was required. When the ship arrived at Macau, the pilot disembarked. In the South China Sea, it was advisable to take soundings at Macclesfield Shoal, and essential to sight either Pulo Sapata or Pulo Condore in order to confirm the ship’s position.84 After passing through Sunda Strait, ships took advantage of the general winds that carried them to the Mascarenes or the Cape of Good Hope.85 Most of the French vessels called at Isle de France, which they reached in about four weeks after passing through the Sunda Strait.86 From there they sailed toward the Cape of Good Hope; after doubling the Cape, they typically made for the islands of Saint Helena and/or Ascension, which offered a break (p.71)

The Voyage and Sea Routes

Map 4.2: French sea routes to China via Cape of Good Hope, 1698–1803

during the long voyage homeward. From Saint Helena and Ascension, they followed a course that took them westward of the Cape Verde Islands and the Azores and finally home to France.

Ports of Call

Ports of call provided an opportunity to take on water, provisions, and wood, large quantities of which were always necessary for cooking; to refresh the crew; and to make needed repairs to the ship. For example, when Villevault put in at Cadiz in February 1765 on her outbound voyage, Captain Bouvet noted that in addition to replenishing the supply of water that the ship and her crew had consumed since departing Lorient, he also made several repairs that had not been possible to carry out at sea.87

The principal ports of call for vessels utilizing the Cape Horn route in the early eighteenth century included Concepción, Valparaíso, and Arica in Chile and Pisco, Callao, and Huacho in Peru.88 A few also called at Ilha Grande off the coast of Brazil. For ships sailing via the Cape of Good Hope, stops typically included one or more of the following: Cadiz, Tenerife, Santiago (Cape Verde), the Mascarenes, Pondicherry, and Malacca. Cadiz served especially as a source of silver, and consequently vessels called there en route to China rather than on the return. While a majority of the ships called at Isle de France on either the outbound or return voyage, or both, not all ships made the same number of stops, though the Mascarenes were indeed the most frequent, and most important, port of call. On the return voyage, most ships stopped at Isle de France and, in the Atlantic, at Ascension; some called at Saint Helena in addition to, or occasionally instead of, Ascension.

(p.72) In the late 1820s to early 1840s, ports of call reflected the use of sea routes via Cape Horn as well as via the Cape of Good Hope, with one or more of the following (listed in alphabetical order) visited by French merchantmen en route to China: Batavia, Bombay, Calcutta, Java, the Mascarenes, Pondicherry, Samarang (Java), San Blas (Mexico), the Sandwich Islands (Hawai’i), Singapore, the Straits, Surabaya (Java), and Valparaiso. Ports of call on the homeward voyage to France included Batavia, Calcutta, Cochinchina, Java, Manila, San Blas, Singapore, and Sumatra. As with the outbound voyage, not all ships made the same number of stops, though Manila’s trading opportunities made it prominent.

It can be seen, in conclusion, that the voyage to and from China required not only good seamanship but also a knowledge of winds and sea routes, and the ability to deal with navigational hazards at a time when parts of the South China Sea still remained to be accurately charted. The French were at the forefront of eighteenth-century hydrography, with d’Après de Mannevillette’s Neptune oriental the bible of sailing instructions after its initial publication in 1745.

Voyages were timed to take advantage of the monsoons, a fact that was reflected in the dates of the trading season in Canton. During the eighteenth century, most French vessels arrived between July and October, and departed between late November and January. The pattern changed as navigators made increasing use of alternative routes that freed them from the restrictions imposed by the monsoons. In the nineteenth century, vessels arrived in every month except June, while a few departures occurred as late as April to June or even July.

The early eighteenth-century use of the sea route to China via Cape Horn owed its introduction to the inclusion of traders and navigators from Saint-Malo who engaged in trade with Spanish colonies on the Pacific coast of South America. The voyage across the Pacific to China was an extension of this route, as vessels called first in South America before heading for Asia. On the other hand, the chief disadvantage to the route across the Pacific was its length and the scarcity of ports of call, with little opportunity to repair or to re-provision.

In addition to testifying to French navigational skills, both the Cape Horn and Cape of Good Hope routes pointed to the interconnected nature of global trade, demonstrating how the route used for one region of markets, be it in South America or the Indian Ocean, might be expanded to serve for a second, considerably more distant, market as well: that of China.


(1.) The cost of trade at Canton was a deterrent to French country vessels as it required considerable capital, not only because of cargo duties, but also because the port fees levied on a ship were based on her dimensions rather than on her carrying capacity. As a result, the duties for a smaller ship could be prohibitively high relative to her tonnage. For a more detailed explanation, see Dermigny, La Chine et l’Occident, 314–316; Manning, Fortunes, 182–183; Van Dyke, Canton Trade, 26–30.

(3.) Joseph Huddart and Jean-Baptiste-Nicolas-Denis d’Après de Mannevillette, The Oriental Navigator, or New Directions for Sailing to and from the East Indies, China, New Holland, &c. &c. &c. (London: Robert Laurie and James Whittle, 1801), 18.

(4.) Huddart and d’Après de Mannevillette, The Oriental Navigator, 5, note, “There is generally a month or two of unsettled weather about the change of the Monsoon, (near land, especially) subject to heavy gales of wind, which, though of short duration and often of small extent, are dangerous to shipping from the sudden shifts to some other quarter, that makes the sea run cross, and wreck the vessels.”

(5.) These figures include both vessels consigned to China from the outset and those that after reaching Isle de France were reconsigned or independently opted to continue to China. For sources of figures in this and the several following paragraphs, see Appendices 1 and 2.

(6.) D’Après de Mannevillette, Instructions sur la navigation, 524–526; Huddart and d’Après de Mannevillette, The Oriental Navigator, 476: d’Après de Mannevillette recommended departing between mid-November and mid-February, “for although your business would permit you to sail at the beginning of the eastern Monsoon, the winds are still so changeable, that you had better wait till they are a little settled”; Dermigny, La Chine et l’Occident, 259–260. Additional later routes may be found in Van Dyke, “New Sea Routes to Canton in the 18th Century and the Decline of China’s Control over Trade,” in Studies of Maritime History 1 (2010): 57–108.

(10.) Règlement touchant la Marine, 67, Titre 24, Article 2.

(11.) Règlement touchant la Marine, 68, Titre 24, Article 5.

(12.) Information on prizes offered by the Académie des sciences may be found in Ernest Maindron, Les fondations de prix à l’Académie des sciences: les lauréats de l’Académie, 1714–1880 (Paris: Gauthier-Villars, 1881), 14–22. The question of longitude was not taken up immediately despite Rouillé de Meslay’s bequest.

(13.) On Acts of Parliament and awards: “Papers of the Board of Longitude collection, 1714 act under Queen Anne,” http://www.cudl.lib.cam.ac.uk/view/MS-RGO-00014-00001/19.

(14.) The eighteenth century witnessed considerable advances in charting the seas east of the Cape of Good Hope. The correspondence between d’Après de Mannevillette and Scottish hydrographer Alexander Dalrymple offers insights into how these two famous hydrographers went about compiling charts; see Andrew S. Cook, “An exchange of letters between two hydrographers: Alexander Dalrymple and Jean-Baptiste d’Après de Mannevillette,” Les Flottes des Compagnies des Indes, 1600–1857 (Vincennes, Service historique de la Marine, 1996), 173–182. On the history of, and advances made in, the charting of the Cochinchinese coast, see Charlotte Minh Hà Pham, “European Navigation, Nautical Instructions and Charts of the Cochinchinese Coast (16th–19th Centuries),” Moussons 27 (2016), 101–129.

(15.) A detailed explanation of the instruments and methods used in celestial navigation may be found in Nathaniel Bowditch, The American Practical Navigator, Being an Epitome of Navigation and Nautical Astronomy (Washington, D.C.: Government Printing Office, 1883).

(16.) An explanation of dead reckoning may be found in Chapter 10 of H. H. Shufeldt, G. D. Dunlap, and Bruce A. Bauer, Piloting & Dead Reckoning (Annapolis: Naval Institute Press, 1991).

(18.) French missionaries were also an important source of information, especially for the Cochinchinese coast. See Pham, “European Navigation, Nautical Instructions and Charts of the Cochinchinese Coast,” 8–9.

(19.) James Horsburgh, India Directory, or, Directions for sailing to and from the East Indies, China, New Holland, Cape of Good Hope, Brazil, and the interjacent ports: compiled chiefly from original journals at the East India House, and from Observations and Remarks, made during twenty-one years’ experience navigating in those seas (London: printed for the author, 1827), 2:122.

(21.) Bryant P. Tilden’s journal [transcript, pp. 35–36] of his first voyage to China in 1815–1816, Bryant P. Tilden Papers, MH 219, Box 3 Folder 2, Phillips Library, Peabody Essex Museum, Salem, Mass.

(22.) Lespagnol, Messieurs de Saint-Malo, 830. Brillant was captured at Arica (Chile) on the homeward voyage.

(23.) Lespagnol, Messieurs de Saint-Malo, 585–588, 590; Dahlgren, Les relations commerciales et maritimes; Dahlgren, “Voyages français à destination de la mer du Sud avant Bougainville (1695–1749),” Nouvelles archives des missions scientifiques et littéraires 14 (Paris: Imprimerie nationale, 1907). Lespagnol notes that the Canaries (or after the end of the War of the Spanish Succession, the Cape Verde Islands) and one of the islands along the Brazilian coast were the two minimum stops for reprovisioning on the voyage from France to the Pacific coast of South America.

(24.) Willem Cornelisz Schouten, Iovrnal ou description du merveilleux voyage de Gvillavme Schovten, Hollandois natif de Hoorn, fait [l]es années 1615, 1616, & 1617 (Amsterdam: Guillaume Ianson, 1618), 22–24.

(25.) La Barbinais Le Gentil, de, Nouveau voyage au tour du monde enrichi de plusieurs Plans, Vuës & Perspectives des principales Villes & Ports du Pérou, Chily, Bresil, & de la Chine: avec une Description de l’Empire de la Chine (Amsterdam: Pierre Mortier, 1728), 1:19–22.

(28.) La Barbinais Le Gentil departed Guacho (Huacho), Peru, on March 4, 1716, and arrived at Guam on May 30, 1716. La Barbinais Le Gentil, Nouveau voyage, 1:148. Four thousand leagues: 19,312 kilometers; 12,000 miles; 10,427 nautical miles.

(30.) Lespagnol, Messieurs de Saint-Malo, 591–593. The ports of call in France’s American colonies were Cayenne (French Guiana), Martinique, and occasionally Plaisance (Placentia, Newfoundland).

(31.) François was sold in Pondicherry on her way home, having been judged unfit to make the entire voyage. Dahlgren, “Voyages français,” 493–494; Veyssière, “Les voyages français à la Chine,” 82.

(32.) Madrolle, Les premiers voyages français à la Chine; La Compagnie de la Chine, 1698–1719 (Paris: A. Challamel, 1901); Dahlgren, Les relations commerciales et maritimes; Lespagnol, Messieurs de Saint-Malo, 830; Rodrigue Levesque, “French Ships at Guam, 1708–1717: Introduction to a Little-Known Period in Pacific History,” The Journal of Pacific History 33, no. 1 (June 1998), 105–110; Susan E. Schopp, “The French in the Pearl River Delta: A Topical Case Study of Sino-European Exchanges in the Canton Trade, 1698–1840” (PhD diss., University of Macau, 2015), 242–264; Veyssière, “Les voyages français à la Chine,” 81–84. Some care should be used with Madrolle; Paul Pelliot, Le premier voyage de “l’Amphitrite” en Chine (Paris: P. Geuthner, 1930), is particularly helpful in correcting errors relating to the first voyage of Amphitrite.

(34.) Detailed sailing instructions for the sea routes to China via the Cape of Good Hope may be found in d’Après de Mannevillette, Instructions sur la navigation, and d’Après de Mannevillette, Le Neptune oriental ([Paris]: [Chez Demonville], 1775). For sailing instructions in English, see Huddart and d’Après de Mannevillette, The Oriental Navigator; and Horsburgh, India Directory. Horsburgh has an index containing an alphabetical list of all places, including islands and shoals, covered in the book. Logbooks of eighteenth-century voyages by French merchantmen to China indicate the routes and ports of call, and may be found in the series 4JJ 129–141 and 144 at the Archives Nationales, Paris. For secondary sources, see Haudrère, La Compagnie française des Indes, 446–464, and Veyssière, “Les voyages français à la Chine,” 159–164. A brief summary in English may be found in the chapter by Philippe Haudrère in Ships, Sailors, and Spices: East India Companies and Their Shipping in the 16th, 17th, and 18th Centuries, ed. Jaap R. Bruijn and Femme S. Gaastra (Amsterdam: NEHA, 1983), 81–84.

(38.) Félix Renouard de Sainte-Croix, Voyage commercial et politique aux Indes orientales, aux îles Philippines, à la Chine: avec des notions sur la Cochinchine et le Tonquin (Paris: Clament frères, 1810), 3:145.

(39.) Constant and Dermigny, Les mémoires de Charles de Constant, 150, 153; 153 shows a map of the city and port. An early view of Macau in which its mountains and forts are prominent may be found in Recueil des voiages [sic] qui ont servi à l’établissement et aux progrès de la Compagnie des Indes orientales, Formée dans les Provinces Unies des Paīs Bas (Amsterdam, 1706), reproduced in La soie & le canon: France–Chine, 1700–1860 ([Paris]: Gallimard, 2010), 26. A view of Macau showing many of its hills is included in Auguste Borget, Sketches of China and the Chinese: From Drawings by Auguste Borget (London: Tilt and Bogue, 1842), Pl. XI, General view of Macao.

(40.) Constant and Dermigny, Les mémoires de Charles de Constant, 150. A century earlier in 1698, François Froger noted ten churches and the presence of the Jesuit, Dominican, Franciscan, and Augustinian orders, as well as the nuns of the Order of St. Claire: Froger and Voretzsch, Relation du premier voyage, 68. Constant, like other visitors to Macau prior to 1835, saw St. Paul’s before most of the building was destroyed by fire.

(41.) For a detailed explanation of the Canton customs procedures on the river and at Whampoa, see Van Dyke, The Canton Trade, Chapter 2.

(42.) The pilots possessed a sound knowledge of the river and its conditions. Bryant P. Tilden wrote of them that they “are regularly educated in all that concerns their profession, and thoroughly know their business, long before they are permitted to act as such. They all undergo a strict examination previous to obtaining a license, and becoming members of Co-Hong—or Company of government pilots. Imm[ediatel]y on their coming on board they assume charge of the ship, give orders etc etc.” Bryant P. Tilden’s journal [transcript, p. 132] of his second voyage to China in 1816–1817, Bryant P. Tilden Papers, MH 219, Box 3 Folder 3, Phillips Library, Peabody Essex Museum, Salem, Mass.

(43.) ANOM: COL C.1.10, ff° 147r–v; Bouvet and Cordier, Le voyage à la Chine, 26.

(44.) ANOM: COL C.1.10, f° 147v; Bouvet and Cordier, Le voyage à la Chine, 26.

(45.) The romanization of place names in the Pearl River Delta reflects their Cantonese name.

(46.) Bryant P. Tilden’s journal [transcript, p. 31] of his first voyage to China in 1815–1816, Bryant P. Tilden Papers, MH 219, Box 3 Folder 2, Phillips Library, Peabody Essex Museum, Salem, Mass.

(47.) Captain Bouvet’s log shows that Villevault passed the Bocca Tigris at 8 in the evening on September 28 (1765). But an hour later, the pilot had him drop anchor because of the tide. At 6 o’clock the following morning—Sunday, September 29—the tide was now sufficiently high, but there was no wind. “With the help of sampans (little local boats), we raised anchor at noon,” Bouvet noted in his log. Eventually the wind came up and his pilot guided the 900-ton ship to within sight of the vessels in the Whampoa roadstead; he anchored that evening. ANOM: COL C.1.10, f° 147v; Bouvet and Cordier, Le voyage à la Chine, 26–27.

(48.) See C. Toogood Downing, The Fan-qui in China, in 1836–7 (London: H. Colburn, 1838), 61–62 for an explanation of the use of boats to tow.

(50.) Horsburgh, India Directory, 330. Huddart and d’Après de Mannevillette, The Oriental Navigator, 451, notes, “The Wantongs are two islands situated near midway between Anunghoy and the western shore; on the principal or northernmost one there is a fort from which a Mandarin comes off to examine your ship, and whether you have a proper Chop which the pilot is accountable for.”

(52.) Downing, Fan-qui, 1:58. Downing described the Bocca Tigris as forming the center of “a very pretty landscape.” Patrick Conner, The Hongs of Canton: Western merchants in south China 1700–1900, as seen in Chinese export paintings (London: English Art Books, 2009), 18, notes, “Western reports tended to belittle not only the forts but the straits themselves…. [T]he painters of the straits, both Chinese and Western, almost invariably reduced the width of the straits and increased the height of the hill-forts for dramatic effect. (In fact, the islands rise at a gentle incline, and the suspension bridge completed here in 1997 has a central span of half a mile.)”

(53.) The route described in this chapter is the one that was taken by the East India ships. A description of the inside passage to Macau, the route used by boats hired to transport Westerners traveling between Canton or Whampoa and Macau (independently of their East India ships), is given in Wood, Sketches of China, 31–34.

(54.) Whether written in the 1740s or the 1830s, Westerners’ descriptions of the natural landscape as seen from the river differ little over the decades, with the chief exception of the degree of deforestation of the mountains—a process that began long before the era of the Canton trade. François Froger described the hills in 1698 as covered with trees (Froger, Relation du premier voyage, 75), while in 1779, Charles de Constant called the mountains “bare, without trees” ( Charles de Constant and Philippe de Vargas, Récit de trois voyages à la Chine (1779–1793) (Pékin, 1939), 6; Constant and Dermigny, Les mémoires de Charles de Constant, 138). Modern studies and Canton trade export paintings show that South China suffered significant deforestation and loss of vegetation over the centuries. See Walter E. Parham, “Art and the Pearl River Delta Environment” (paper presented at the International Symposium on Environment and Society in Chinese History, Center for Chinese Social History, Nankai University, Tianjin, China, August 17–19, 2005, https://southchinaenvir.com/historical-perspectives/art-and-the-pearl-river-delta-environment.

(55.) Pierre Poivre, Travels of a Philosopher: or, Observations on the Manners and Arts of Various Nations in Africa and Asia (Dublin, 1770), 138–139. The “river of Canton” (la rivière de Canton) is another name for the Pearl River.

(57.) Carl Gustaf Ekeberg, Précis historique de l’économie rurale des Chinois (Milan: Frères Reycends, 1771), 7–8. Ekeberg was writing on the effect of the tides on cultivation. See also Wood, Sketches of China, 60: “These tracts, which are overflowed by every spring-tide, except where the embankments protect them, are admirably adapted to the cultivation of rice, and are devoted chiefly to that purpose.”

(60.) According to hydrographers Huddart and d’Après de Mannevillette, the lower shoal of the First Bar had “not more water upon it than 2 fathoms [twelve feet]” at low tide. See Huddart and d’Après de Mannevillette, The Oriental Navigator, 452. Twelve feet was hardly enough for ships drawing as much as twenty to twenty-five feet when loaded. Auguste Duhaut-Cilly mentioned that his ship, Héros, required sixteen sampans to get her safely across Second and First bars: six “to mark out the rocks and ten others to tow the ship.” Duhaut-Cilly, Frugé, and Harlow, A Voyage to California, 233, 234.

With regard to sight lines, the India Directory, for example, referenced both the Second Bar and Whampoa pagodas in explaining the position of Brunswick Rock, a submerged hazard shortly above the First Bar: “When upon the rock, the Second Bar pagoda bore S. by E. 3/4 E., Clump of Trees or Chop House S.S.E. 1/4 E., Whampoa Pagoda W. ½ N.” Horsburgh, India Directory, 334.

(61.) Shizita, 獅子塔‎.

(63.) Bryant P. Tilden’s journal [transcript, pp. 33–34] of his first voyage to China in 1815–1816, Bryant P. Tilden Papers, MH 219, Box 3 Folder 2, Phillips Library, Peabody Essex Museum, Salem, Mass. Pehr Osbeck, who was one of Carl Linnaeus’s “Apostles” and who kept a detailed record of virtually every plant he saw, commented that the pagodas he could see in the countryside were too far away for him to be able to identify the exact species of trees and plants growing on them. Pehr Osbeck, A Voyage to China and the East Indies, transl. John Reinhold Forster, 2 vols. (London: Benjamin White, 1771), 199.

(66.) Several smaller pagodas served similar purposes; for example, North Shore Pagoda to the east of Junk Island was the leading mark for clearing Brunswick Rock. “This survey of Canton River from the second bar creek to the upper part of Whampoa Reach” ([London]: Published by James Horsburgh, hydrographer to the Honble. E.I. Company, Feby. 2d, 1818), https://www.loc.gov/resource/g7822p.ct005539/?r=0.184,0.049,0.112,0.085,0.

The artist Borget described his view of the pagodas of the city from a location beyond Canton: “We ascended on a large and magnificent plateau. The view from this spot was very pleasing. We were behind Canton, and could see its pagodas through amongst the trees, and sometimes overtopping them.” See Borget, Sketches of China and the Chinese, excerpt on unnumbered page preceding Pl. XXXII, Commercial junk and mountains behind Canton.

(68.) Hunter, TheFan-kwae,” 13; Wood, Sketches of China, 47. Hunter estimated the population as “many thousand inhabitants.”

(69.) Charles de Constant, in Vienne, La Chine au déclin des Lumières, 146. Bryant P. Tilden observed that “thousands of both sexes were busy at various work” in what he described as “the pretty, populous village town of Whampoa.” Elsewhere he wrote, “The pretty town, or rather the extended island village of Whampoa … is built on mud-banked land; and is always a busy lively place, being inhabited by shipwrights, all sorts of mechanics, fishermen, and rice growers.” Bryant P. Tilden’s journal [transcript, pp. 35, 100] of his first voyage to China in 1815–1816, Bryant P. Tilden Papers, MH 219, Box 3 Folder 2, Phillips Library, Peabody Essex Museum, Salem, Mass. The island of Whampoa was bounded on the north by the river; on the east, by Junk River; on the west, by Fiddler’s Reach and Honam (Henan) Island; and on the south, by the anchorage, which was in turn bounded on its south side by Danes Island (Ile des Danois) and French Island (Ile des Français). Wood, Sketches of China, 47–49 offers a detailed description of the island, the town, and the pagoda.

(72.) Bankshall (bancassal/-aux in French). See Pl. 1.11 and Pl. 1.12, Chinese artist, Whampoa: the anchorage (late eighteenth century), in Conner, The Hongs of Canton, 20–22, for a view of the bankshalls along the southern shore of Whampoa Island. Sometimes referred to in English as a “shed” (un hangar), the bankshall was considerably larger than what the word “shed” often implies. William Hickey described them as being “from sixty to one hundred feet,” or approximately 18 to 30.5 meters, in length. William Hickey and Alfred Spencer, Memoirs of William Hickey, 4th ed. (London: Hurst & Blackett, 1913), 1:197.

(73.) Farmhouse: Osbeck, A Voyage to China, 185–186, notes that they kept “chickens, hogs, &c.” in the bankshalls, too.

(74.) Colin Campbell, Paul Hallberg, and Christian Koninckx, A Passage to China: Colin Campbell’s Diary of the First Swedish East India Company Expedition to Canton, 1732–33 (Göteborg: Royal Society of Arts and Sciences, 1996), 96. Campbell was the first supercargo of Fredericus Rex Sueciæ, the first ship that the Swedish East India Company sent to China.

(75.) Shaw and Quincy, The Journals, 175–176. Shaw, who sailed on the Empress of China, described the bankshalls as he saw them in 1784. The bankshalls also provided workshop space for repairing the ship and her rigging, a necessity after a lengthy voyage; see Constant and Dermigny, Les mémoires de Charles de Constant, 139–140. The islands derived their names from the nationalities that originally obtained the privilege of having their bankshalls on them (see Hunter, TheFan-kwae,” 13). The bankshall also often merits entry in logbooks, as Captain Bouvet’s shows. As Villevault was being positioned at Whampoa on Monday, September 30, Bouvet wrote in his log, “Very early in the morning, with the tide and with the help of sampans, I sailed between the land and the ships in the harbor and positioned us as close as possible to the spot where we’ll set up our bankshalls (or sheds for items in the ships in port).” A few hours later, at 10 a.m., Bouvet sent the ship’s boat upriver with Villevault’s supercargo, who carried the instructions for the Canton comptoir that Captain Bouvet had been given before departing Lorient. Bouvet spent the afternoon securing his mooring; the following day he began downrigging his ship and, together with Captain de La Pallière of the Beaumont, made arrangements for their bankshalls. ANOM: COL C.1.10, f° 147v–148r; Bouvet and Cordier, Le voyage à la Chine, 27–28.

(76.) As burial places, Bryant P. Tilden and others noted: “Foreigners who unhappily may die at Canton and Whampoa anchorage are buried on this [Danes Island], & another named French Island, by officers, ship’s crews, and friends from Canton.” Bryant P. Tilden’s journal [transcript, p. 100] of his first voyage to China in 1815–1816, Bryant P. Tilden Papers, MH 219, Box 3 Folder 2, Phillips Library, Peabody Essex Museum, Salem, Mass. William C. Hunter, writing of the period between 1825 and 1844, stated that all vestiges of the bankshalls on French and Danes islands had long since disappeared, “but numerous decaying tombstones, half buried beneath earth and weeds, still tell the tale.” Hunter, TheFan-kwae,” 13.

(77.) Osbeck, A Voyage to China, 347–348; fruit trees, 188. Chinese potatoes: Osbeck believed these to be “the Spanish potatoe, Convolvulus batatas.”

(78.) Most of the ship’s crews remained with their ships at Whampoa, visiting Canton on only a few days of liberty, but traders and ship’s captains (and occasionally other officers) went all the way to Canton.

(80.) Wood, Sketches of China, 54–55. Wood was writing of the late 1820s. William C. Hunter stated that in 1825, the Pearl River “presented a vastly different appearance … from what it did twenty years later. It was then crowded with native vessels, including those immense coasting junks which have now almost entirely disappeared.” See Hunter, TheFan-kwae,” 13–14.

(82.) Two examples of paintings that show the crowding of boats on the Canton riverfront near the international quarter are reproduced in Hong Kong Museum of Art and Peabody Essex Museum, Views of the Pearl River Delta: Macau, Canton and Hong Kong (Hong Kong: Urban Council of Hong Kong, 1996), and identified as follows: View of Canton, c. 1800, by an anonymous Chinese artist (pp 162–163, N°. 50, E79,708), and William Daniell’s View of the Canton Factories, 1805–1806 (pp. 160–161, N°. 49, AH64.24).

(83.) James Wathen, Journal of a Voyage in 1811 and 1812, to Madras and China (London: J. Nichols, Son, and Bentley, 1814), 182.

(87.) ANOM: COL C.1.10, f° 140r; Bouvet and Cordier, Le voyage à la Chine, 9, entries for February 13 and 14, 1765.

(88.) Dahlgren, “Voyages français;” Lespagnol, Messieurs de Saint-Malo, 830.