Volume 89 1980 > Volume 89, No. 2 > The 'mystery' of Gran Cocal: European discovery and mis-discovery in Tuvalu, by Doug Munro, p 167-198
THE “MYSTERY” OF GRAN COCAL: EUROPEAN DISCOVERY AND MIS-DISCOVERY IN TUVALU
In the century and a half since Europeans discovered the last of the atolls and reef islands of Tuvalu, the central Pacific has been fully charted and the activities of the early explorers fairly well documented. Yet in our research into Tuvalu's early history it has become apparent that there was considerable confusion among early explorers about the identities of the islands they saw. Islands were mis-identified and wrongly located on the charts, so that some explorers did not realise they had made a discovery. The most elusive of all the Tuvalu islands, “Gran Cocal”, was eventually demoted to a shoal, still shown as such on some charts. Even the most recent authoritative works on discoveries in the central Pacific have failed to determine the real identity of Gran Cocal, or to give credit to its discoverer.
In this paper we attempt to untangle the complex sequence of events which led to so much confusion among mariners and chart makers over the northern Tuvalu, and to demonstrate the true identiy of Gran Cocal, as well as to show how it finally came to be known as a shoal, complete with local traditions about it. We draw on original Spanish materials (rather than their published English translations) and make use of original manuscript charts (rather than their published equivalents). We also present, for the first time we believe, an English translation from the Russian of the earliest European contact with Polynesians in the southern Tuvalu islands.
THE ISLAND SETTING
The nine islands comprising the independent nation of Tuvalu lie just south of the equator and to the west of the International Date Line (see Figure 1).- 168 - 169
All are small low coral islands, six of which have lagoons and are true atolls. With the exception of tiny southernmost Niulakita, the islands have been inhabited for many centuries by Polynesians, who probably discovered and settled them in the course of purposeful or drift voyages north from Samoa and Tonga. Nui atoll is known to have been well populated as early as 1568, and genealogical traditions from northernmost Nanumea seem to date settlement there to the 14th century. It is quite likely that archaeological research would push back these dates considerably further.
Before European contact in the early 19th century each Tuvalu island had developed a political and social system well suited to the ecological and territorial limitations of the small atolls. Each was governed by a hereditary chief, or chiefs, elected by a consensus of elders. There seems to have been little inter-island warfare, and the largest political unit was normally a single island. Tuvaluans experienced contact with the Western world relatively late, beginning (with two exceptions) early in the 19th century with sporadic short visits by passing whalers, and continuing as explorers attempted to chart these last unknown parts of the Pacific. Yet they remained essentially isolated and unknown to the West until Samoan pastors trained by the London Missionary Society were established throughout the group during the 1860s and 1870s. At each island churches and schools were founded, and Tuvaluans eagerly mastered the arts of reading and writing. Social changes followed rapidly, and accelerated after 1892 when Tuvalu, then known as the Ellice Islands, came under British protection. In 1916 a Colony, which included the Micronesian Gilbert Islands to the north, was formed. British rule included the introduction of a Western legal system and attempts to replace the hereditary chiefly system with a quasi-parliamentary one. The two groups remained united until 1976, when the Ellice Islanders voted to separate and resume the traditional name of Tuvalu. Tuvalu was granted independence in October 1978. 1
THE EARLIEST EUROPEAN ARRIVALS
With the recent publication of several studies of the exploration and discovery of the Pacific it seemed that the discovery of Tuvalu was a straight-forward and well-understood process (Buck 1953:108-9; Stackpole 1953: 280-1, 342-8; Sharp 1960; Maude 1968:35-133). Of these studies the latest and most authoritative is by H. E. Maude, whose findings are summarised here:- 170
Previously Accepted European Discovery of Tuvalu
Such a list makes it clear that there were two, and perhaps even three, “periods” of discovery in Tuvalu: Mendaña's, Mourelle's, and the whalers'.
The earliest Spanish discoveries in Tuvalu were basically happenstance events, since they took place during the infancy of European trans-Pacific voyaging when navigation techniques were primitive. In late 1567 Alvaro de Mendaña sailed from Peru with two ships and a complement of about 100 men on his first Pacific voyage. 3 His ostensible purpose was “to convert all infidels to Christianity” (Amherst and Thomson 1901, I:v-vi), although more substantial motives included the lure of conquest and possible riches to be found. Fifty-seven days out, on January 16, 1568, the expedition sighted its first land, a low coral atoll. Islanders paddled canoes within earshot of the two ships, but then retreated to shore and could not be induced to return. Because the chief pilot, Gallego, was reluctant to endanger the ships in adverse winds by sending a party ashore, the Spaniards had no further contact with the island. It was undoubtedly fortunate for the islanders that the winds were on their side. Mendaña called the small island Isla de Jesus.
On his second Pacific crossing 27 years later Mendaña discovered another isolated island, apparently uninhabited, and again unsuccessfully tried to land. He called this island La Solitaria (Markham 1904, I:31-2). Maude (1968: 53-64) has carefully and convincingly shown that Isla de Jesus was Nui and La Solitaria was Niulakita, both in Tuvalu. Without chronometers, which were not in use until the late 18th century, longitude could only be reckoned crudely, so these earliest European discoveries in Tuvalu were badly misplaced on the charts. In some early charts “Jesus Island” and Solitaire “Island” are - 171 both shown, the first far to the west, the other far to the east of their real locations (see Figure 2 for “Jesus Island”). Centuries after their Spanish discovery these two islands were rediscovered and renamed, and placed on charts as Nederlandisch Island (Broeze 1975:35-36) and Independence Island (Maude 1968:126) respectively. After these somewhat premature discoveries, there was a hiatus of several centuries before European ships again visited Tuvalu.
MOURELLE'S GRAN COCAL
The next discovery in Tuvalu was also a Spanish one, made by Francisco Antonio Mourelle 4 on a voyage from the Philippines to Mexico. On May 5, 1781, Mourelle sighted a low island which he named El Gran Cocal (or alternatively Isla del Cocal in some manuscripts). Although the expedition's chief pilot, Jose Antonio Vasquez, kept detailed records of the vessel's daily course and noted latitudes and longitudes and other particulars of each island encountered, when the narrative of the voyage was published it was only Mourelle's own summary which was printed, and all details from Vasquez' log were omitted. For our purposes in trying to identify Gran Cocal this was a crucial decision, for Mourelle's narrative merely remarked that the island lay about six degrees south of the equator. This failure on his part to be more specific, particularly in terms of longitude, was to result later in a multitude of contradictory identifications for Mourelle's Gran Cocal.
Which Tuvalu island, then was Gran Cocal? The position at 6° south limits the field to the three northernmost Tuvalu islands: Nanumea, Nanumanga and Niutao. Most commentators in the 19th century chose Nanumanga as Mourelle's Gran Cocal, and this opinion has been generally accepted (Sharp 1960:150; Maude 1968:93-4). But was Gran Cocal Nanumanga?
We feel there is ample evidence to show that, in fact, Cocal was not Nanumanga but the island of Niutao. If this can be demonstrated, then of course the formerly accepted discoverer of Niutao, whaling captain Obed Starbuck, was not making a new discovery when he sighted that island, and we are left with the further question of who actually discovered Nanumanga. The subsequent sighting of northernmost Nanumea by Mourelle a day after he found Gran Cocal is not in question, however; the name Mourelle gave it, San Augustin, was in use for most of the 19th century. But to clarify matters the story of the discovery of the three northernmost islands of Tuvalu has to be retold.
The Voyage of La Princesa
On November 24, 1780, Mourelle, commanding the Spanish frigate Nuestra Señora del Rosario, more commonly known as La Princesa, sailed from the east coast of Luzon bound for San Blas, Mexico, with secret dispatches for the Viceroy of New Spain. Mourelle set out with considerable misgivings- 172
FIGURE 2- 173
1841 Chart of Tuvalu from the United States Exploring Expedition (Wilkes 1845, Vol. 5: frontispiece)., Uncertainty over location of early Spanish discoveries led to double charting of Nui (Jesus I., and Nederlandisch I. here) and Niulakita (Independence I., here, and also Solitaire I., not shown here, but figured 400 miles eastward, just north of Samoa).
because his orders involved a voyage “altogether new, in latitudes almost unknown”, and meant sailing during a difficult time of year. To sail from the Philippine Islands to Mexico, he wrote,
. . . the proper time of departure is the month of June, when the westerly winds will carry a ship to the east of the Marianne Islands. At any other season it would be vain to hope for a prosperous voyage (Mourelle 1799:197-9).
As Figure 3 makes clear, his worst fears were realised: La Princesa, attempting to make a southern crossing to avoid enemy English ships and unfavourable winds in the northern latitudes, pursued a tortuous course which eventually took her far to the south of Tonga. What the distraught captain described as “millions of cockroaches”, so many that: “It is impossible for a man, without having seen them with his own eyes, to form any idea of their numbers” (Mourelle 1799:229), attacked and devoured most of the ship's insufficient stores. Early in 1781 some refreshments were obtained in Tonga (Langdon 1977:52-4) but Mourelle failed to find the westerly winds he needed and turned north. His route was to take him through the hitherto unknown Tuvalu group.
La Princesa in Tuvalu
Attempting to make all possible speed now, since his crew was in ill-health and provisions were meagre, Mourelle (1799:234) wrote:
I availed myself of every favourable moment, and on the 5th of May, I found myself in the latitude of 6° [south]. We here found a very low island, surrounded by a sandy shore, terminating in an impenetrable ledge of rocks, near which I found no bottom with a line of upwards of fifty fathoms. The island was covered with a thick plantation of cocoatrees. The sight of these was the more pleasing to my crew, as that very day the last of the provision procured . . . [in Tonga] was expended.
“Indians” came off in canoes, negotiating the reef with great difficulty so that they were able to bring only a few coconuts. Their attempt to tow La Princesa to shore failed, and after six hours Mourelle gave up hope of securing supplies and sailed away.
The appearance of the islanders surprised the Spaniards:
They came on board so bedaubed with paint, that we were almost tempted to take them for representatives of demons. The beards of most of them hung down to the breasts. Near the plantation of cocoa-trees, there was such a number of huts, regularly arranged, that the island may be concluded to be extremely populous (Mourelle 1799:235).
Mourelle departed from the island, which came to be called El Gran Cocal- 174
FIGURE 3- 175
Mourelle's wandering course through the south-west Pacific in 1781. Gran Cocal is shown as Cocos I. near the 180° meridian at 5° south. (Portion of chart accompanying Mourelle 1799).
(Great Coconut Plantation), on the evening of May 5. He sailed north-west and the next day,
. . . in the evening, we saw another island, very low but larger than the preceding, and passed to the north of it the distance of 6 leagues. (Mourelle nd.: entry for May 6, 1781)
La Princesa did not approach this island and there was no contact with the inhabitants. Mourelle called it San Augustin. A few weeks later the vessel reached Guam without further incident, except that the crew was badly affected with scurvy. After securing new stores Mourelle set sail again, this time following the more usual northern Pacific route, and reached San Blas in late September, 1781 (Brand 1967:132).
The Identity of Gran Cocal Island
The identity of the first of the two Tuvalu islands Mourelle discovered has been the subject of some speculation during the past two centuries. Since the Spanish captain's published account provided few substantive details, researchers have had to rely on an estimate of the relative positions of the two islands, and on the singular characteristic of the painted faces of the Cocal islanders. Recent writings on Pacific discoveries, as well as earlier accounts and maps, have agreed in identifying Gran Cocal as the northern Tuvalu island of Nanumanga.
There are a number of discrepancies between the information provided in Mourelle's account and this identification, however, and it now seems clear that Gran Cocal must certainly have been Niutao, some 57 miles east of Nanumanga. This conclusion is further supported by new information: original manuscript logs of both Mourelle and his chief pilot, Vasquez. But before examining this material it will be useful to show how other researchers have apparently misread the original Mourelle account.
In his study of discoveries in the Pacific, Andrew Sharp (1960:150-1), relying mainly on the brief description of Nanumanga in the Pacific Islands Pilot, concluded that it was this island that Mourelle had visited briefly on May 5, 1781. H. E. Maude's more thorough study, bolstered by a personal acquaintance with Tuvalu, was in agreement with Sharp's identification:
. . . the description of the Isla del Cocal given by Maurelle indicates that it was a reef island and its position together with the course of the vessel, leaves one in no doubt that it was, in fact, Nanumanga in the Northern Ellice (Maude 1968:93).
Yet in both cases the criteria for supposing that Gran Cocal was Nanumanga can be seen to apply as well or better to the neighbouring island of Niutao. Niutao lies closer to six degrees than does Nanumanga (6° 06′S for Niutao, 6° 18′S for Nanumanga). 5 Niutao, like Nanumanga, is a small reef island- 176
FIGURE 4- 177
Nanumea, Nanumanga, and Niutao, the three northernmost Tuvalu islands, with a reconstruction of Mourelle's course on May 5-6, 1781.
without a true lagoon, and both have a land area of just under three square kilometres. Each appears nearly circular from the sea and is surrounded by an unbroken reef which produces a line of breakers near the shore. The two islands are thickly wooded, mainly with coconut palms, and have densely settled villages on their western, leeward shores. Moreover, they are so similar in physical appearance that Mourelle's published description could apply equally to either. 6
There are two other factors which have been considered in identifying Gran Cocal: its position relative to San Augustin, which was discovered the following day, and the strikingly painted bodies of the Cocal islanders. The relative positions of the three northernmost Tuvalu islands are sketched in Figure 4. Maude has claimed that the positions of Nanumanga and Nanumea fit those of Mourelle's Gran Cocal and San Augustin. Yet, as Mourelle's narrative quoted above makes clear, La Princesa sailed north-west from Gran Cocal and then passed six leagues (20 miles) 7 to the north-east of San Augustin. This would have been patently impossible if Gran Cocal were indeed Nanumanga, as Figure 4 shows. Even if we suppose that the vessel headed more northerly than Mourelle's account indicates, the 18th century frigate would undoubtedly have been carried considerably to the west of San Augustin by the prevailing winds from the east-north-east to north-east, if Gran Cocal were indeed Nanumanga. 8
The distance between Niutao and Nanumea also seems to fit far better with the crossing of 24 hours Mourelle's account indicates. La Princesa had made good an average of about 65 miles a day since leaving Tonga 11 days earlier. The Nanumanga to Nanumea passage is just 37 miles, and quite unlikely to have taken a full 24 hours, whereas the 71 miles from Niutao to Nanumea would have taken that long.
The relative positions of Gran Cocal and San Augustin thus lend credence to the view that Gran Cocal was actually Niutao. But what then of an additional factor used by Sharp and Maude to support their view that Gran Cocal was Nanumanga — namely the presumed unique Nanumanga custom of daubing faces with soot or paint? In correspondence over this matter, Maude (personal communication, August 1975) said he had never heard of the practice on Niutao, and it was this that confirmed his belief that Gran Cocal must have been Nanumanga. Maude's source, an account in the New Bedford Mercury for March 20, 1829, reads:
Captain Plaskett, of the whaling ship Independence, has politely furnished us with the following discoveries made during his last voyage to the Pacific Ocean- 178
Smut Face island lat. 6° 16′S, long. 177° 19′E
So named because the chiefs and men of distinction, as a mark of their superiority, daub their faces with smut.
While Maude takes this to refer to Nanumanga, the longitude given puts Smut-Face Island almost 60 nautical miles east of Nanumanga but within two nautical miles of Niutao's actual longitude.
Nor was Maude the first to mis-identify Smut-Face Island as Nanumanga. In his classic study of early American whaling activity, Edouard A. Stackpole reports that the Independence II left Rotuma in May 1826 and sailed north-north-east. A week later an uncharted island was sighted: “This island (Nanomana [sic]) was an actual discovery. Captain Plaskett named it Smut-Face Island, from the appearance of the natives who daubed their faces with black streaks” (Stackpole 1953:343). 9
What needs to be stressed, however, is that the probably ritual practice of daubing faces with black paint was fairly common in Tuvalu. It is reported definitely on several islands and may have been practised throughout the group. In 1827 Plaskett was at Vaitupu and reported that the king “is a large good looking man, but very dirty being painted from head to foot with a kind of stuff like soot” (Stackpole 1953:347, quoting Colt nd.: entry for August 22, 1827). And in 1853 another whaling captain spent several days ashore at Nanumea, where he noted that “an old priest was smutched over with some black substance”, and that 30 or 40 islanders were “dressed and painted over their faces and limbs in a most grotesque manner”. He also ungallantly referred to one female priest as “old smutty face” (Pease 1854, 1962). In addition, Mourelle, in 1781, and Plaskett, in 1826, noted the practice on the island we identify as Niutao. From the accumulated evidence it is clear that identifying Smut-Face Island largely on the basis of the islanders' bedaubed faces cannot stand. Indeed, Maude (personal communication, December 1978) is now in agreement with our identification of Gran Cocal (and Smut-Face Island) with Niutao. Interestingly, though the practice was widespread within Tuvalu, it is not reported definitely for Nanumanga.
The Mourelle and Vasquez Manuscripts
Much of the uncertainty over the identity of Mourelle's Gran Cocal arose from the omission of any navigational information in the Spaniard's published narrative. Since position co-ordinates were not given for Cocal and San Augustin, and the distance separating them was not indicated, there has necessarily been a certain amount of conjecture involved. The original manuscript logbooks from the voyage have survived, however, and include the account of Mourelle, the log of his chief pilot, Vasquez, and a chart plotting the daily position of La Princesa and islands seen. These provide enough detail to dispel- 179
FIGURE 5- 180
Part of the manuscript chart of the course of Mourelle's La Princesa in 1781. El Gran Cocal island and San Agustin island (written San Augustin in the log book) are at the upper right. From a manuscript in the Museo Naval, Madrid (Vasquez nd.).
any doubts about Gran Cocal's identity.
A portion of the manuscript chart is reproduced in Figure 5, and the positions of Gran Cocal and San Augustin appear to match those of Niutao and Nanumea rather than Nanumanga and Nanumea. Even more important are the position co-ordinates supplied by the pilot's log (Vasquez nd.) and the appended list of islands sighted. As might be expected, the longitude figures for both islands are not particularly accurate, but granting this concession to an age when longitude calculation was still primitive, Vasquez' positions for the two islands are quite precise. After converting the longitudes to Greenwich standard 10 the figures are: Gran Cocal 6° 02′S, 175° 28′E, San Augustin 5° 35′S, 174° 26′E.
Even if the question of the absolute accuracy of these positions is ignored, the location of the two islands relative to each other can be computed. Doing so, we find that by Vasquez' reckoning San Augustin lay 68 nautical miles west-north-west (293° true) of Gran Cocal. On charts today Nanumea lies 71 miles west-north-west (293° true) of Niutao, which leaves no doubt that it was these two islands Vasquez' figures refer to.
The pilot's detailed account of the sighting of the two islands is a useful supplement to Mourelle's published narrative, as the passages for May 4-6, 1781, indicate:
Nightfall [May 4] horizons very still ... wind fair ... tackle at full sail. At 12 [midnight] a low island was discovered, the west point of which bore to N ¼NW, and the south point to N ¼NE.... At 12:30 [a.m.] saw fires on the west point of this island, at 5 with light winds SSE sailing towards it.
Dawned sky and horizon with thick clouds ... the island ... distant about 1 league ... [at] 8 [a.m.] ... we discovered a settlement with many Indians, the whole island being a coconut plantation. And at 8:30 we entered the shelter of it [the island], remaining to the south, running at a distance of one musket shot, without finding bottom, since it is all beach with reef....
Here we remained on board and some canoes came out which brought nothing more than a few coconuts. The Indians came much painted all in black, and spoke almost exactly like those of the other islands. But none wanted to climb aboard, instead their earnest wish was to tow the frigate towards the island with their canoes. They left and went to land where there were more than 300 Indians on its shores, and they brought nothing more than a few coconuts....
At noon [May 5] we continued with full sail, with moderate SE ... at 4:30 [p.m.] we lost sight of the island from below [on the deck] to the SSE....
Dawned [May 6] ... light winds from NNE ... at 10 a light wind - 181 appeared in the 3rd quarter and we sailed under full sail close-hauled.. . . At 3:30 we saw from aloft an island, low and tree covered, not very large, which bore to the SSW O/8 W, distant 5 to 6 leagues, which island I judge to be in lat. 5° 35′ [south] and longitude 48° 33′ [east], appearing to be about 3 leagues in circumference. . . (Vasquez nd., trans. K. Chambers).
Early Charting of Gran Cocal and San Augustin
Although some of the confusion which developed later in the 19th century over whether Gran Cocal was an island or a shoal can be attributed to charting errors, as will be seen, the earliest charts to map Mourelle's discoveries were actually amazingly accurate. Vasquez' own manuscript chart has been reproduced above, as has a portion of the map which appeared with Mourelle's narrative published in 1799. The first navigational chart to show Gran Cocal and San Augustin was apparently Aaron Arrowsmith's Chart of the Pacific Ocean, published in 1798. Arrowsmith may have used the Vasquez manuscript chart or list of island positions, since the relative positions he assigned to the two islands correspond to those determined by Vasquez, although Arrowsmith's longitudes and latitudes are somewhat at variance with those of Vasquez. But in 1810 Arrowsmith issued a revised edition of his Chart, in which he shifted the positions of both islands slightly, moving Gran Cocal to the west so that it lay much closer to San Augustin. 11 In making this change, a response to position reports from a merchant brig which sighted Nanumea in 1809 (discussed below), Arrowsmith initiated the process of identifying Mourelle's Gran Cocal with the island of Nanumanga instead of Niutao.
WHO DISCOVERED NANUMANGA?
If we can conclude that Mourelle's Gran Cocal was Niutao, we are still left with the question of who discovered Nanumanga. Probably the first outsider to sail through the northern Tuvalu after Mourelle was Captain Patterson in the brig Elizabeth, on a trading voyage from Port Jackson to China in 1809. Patterson sighted Nanumea, and, apparently unfamiliar with Mourelle's discovery of San Augustin, he named the north-western islet Taswell's Isle and the south-eastern islet Sherson's Isle (Naval Chronicle, 24(1810):313). It seems that Patterson never sighted Nanumanga. In 1819 Captain De Peyster, an American in command of the British merchant brigantine Rebecca, passed through southern Tuvalu waters and discovered Funafuti and Nukufetau (Maude 1968:115; Paulin 1947), but did not continue on to the north.
The following year a Russian exploring expedition led jointly by Captain M. N. Vassiliev and Captain G. S. Shismarev, and consisting of the sloops Otkrytie (Discovery) and Blagonamerenny (Good Intent), cruised through - 182 Tuvalu waters. On April 17, 1820 the Blagonamerenny, under the command of A. P. Lazarev, encountered an inhabited atoll in the southern Tuvalu and spent a day in contact with the inhabitants and surveyed the island. This atoll was Nukufetau. Our identification here rests on Lazarev's observations at 8° 4′S, the sketch in Lazarev's account which corresponds with Nukufetau's distinctive rectangular shape, and the statement that the island lay 176 miles south-east of the island seen by the Elizabeth in 1809, that is, Taswell's Isle, or Nanumea (Lazarev 1950:163-8). It did not appear on the 1810 edition of Arrowsmith's Chart of the Pacific Ocean and Lazarev took it for a new discovery. Accordingly the expedition gave it the name “Islands of Good Intent” before sailing on to the north. Although we cannot count this an actual discovery. Accordingly, the expedition gave it the name “Islands of Good Intent” the charts the previous year, Lazarev's account of his encounter with the Tuvaluans deserves quoting at some length since it provides some interesting details and describes the first known European contact with the southern Tuvalu islanders (De Peyster had not actually seen anyone at either Nukufetau or Funafuti). The narrative of Lazarev's voyage, available previously only in Russian, notes that:
At 7:30 in the morning [on April 17, 1820] . . . [we] saw . . . some land. These turned out to be eleven low coral islands, with many trees, joined by a reef which extended about 15 miles to their south.. . . Approaching closer, to about 5 miles, we saw a circle of water between the islands, which were linked to one another by coral reefs.
The islands were low, and formed from coral, and we noticed on them a vast number of coconut trees, pandanus and bread-fruit trees. None of the islands was more than a mile long.. . .
Meanwhile, four little boats set off from the beach and headed directly for us, while we began to sail lengthways along the group of islands.
Although the Discovery was laying to, the islanders did not head for it, but headed for us, as we were much nearer to the islands. Wanting to observe the islanders, we struck sails and also hove to. When the islanders had approached to within about 40 metres of us without, however, pulling right up alongside, then each of us began to wave to them with white handkerchiefs or with any green plants which happened to be on the boat, as a substitute for the green branches which on other islands, the savages display as a mark of peace. In this way we tried to persuade them to draw up alongside, but in vain. Captain Shismarev sent an officer on a ship's boat, to meet the savages and give them some hoop iron and mirrors. When he approached them, they did not move away, but nevertheless, they did not seem to want to meet him. However, he came up to one boat in which, as in the other three, there sat four - 183 islanders. They took hold of our oars, and in this way kept close to the ship's boat. It was clear that they were not at all afraid of the sloop, for they reckoned it too far away for any weapons to be used against them, however, they were cautious about the small boat. Their comrades, seeing our peaceful intentions, also approached us, and we finally became acquainted. As a sign of their friendship, the islanders touched the faces of our men with their noses and tormented them with the stink of sweat and coconut oil. The officer invited them on board the sloop, and, after showing them where to board it, began to head back. The islanders at first followed him, apparently very fast, and then suddenly they again stopped, and began to talk among themselves, refusing to continue, even though we threw them a rope. Such is the usual manner of all savages when they see aliens for the first time.
These islanders were all of average height, well shaped, and quite corpulent, and their faces were regular and some even quite handsome, although blacker than those of many islanders of the South Seas, but still not as black as the faces of Africans. The colour of their bodies was dark brown, and glossy from their habit of rubbing themselves with coconut oil. At their waist they had belts made of coconut fibre and multi-colored ribbons of their own manufacture, mostly red. Their hair was tightly bound in a very dark bun, and something which looked rather like an animal's tail hung from it. White and orange flowers were bound into their hair, and almost all of them had white shells tied to their arms, above the elbow. One of these was given to us as a present. Approaching us, the islanders waved a green branch as a sign of peace. It seemed to me that their bodies were not painted 12 but others insisted that they saw several signs on their shoulders. Their ears were pierced and stretched, as with many islanders of the Southern Seas, but there was nothing fixed to the ears. The boats of these islanders were of extremely simple, even rather poor workmanship. They were made from several planks bound together with cords of coconut fibre, and they leaked so much that those sitting in them had to continually bail out the water. They were about 20 feet long and broad enough for one person to sit, that is, about 1° feet. On one side they had an outrigger. The bow and stern were identical and they had a triangular mat sail.
The weapons of these islanders consisted of wooden spears with barbs, of crude workmanship, probably because they do not use them much and rarely fight each other. We concluded this last from their manner with us. On meeting us they displayed no insolence or hostility, and their attitude was extremely friendly. From everything we saw, we concluded that this was an extremely peaceful people. They brought none of their fruit, such as coconuts, with them, nor any examples of their - 184 workmanship, except one mat. Presumably, they came out not to trade with us, but merely to look at us. Both from this and from their lack of interest in the bits of iron which we gave them, one can conclude that they had not yet seen Europeans and were not yet acquainted with iron, that most attractive of all metals for South Sea islanders. On the other hand, mirrors, even the very smallest, attracted all their curiosity. One can also conclude that no Europeans have ever been here from the fact that, even having seen our ship, they apparently had no idea of the needs of sailors, and did not bring us any food, although they apparently have food in abundance.
Seeing that, in spite of all our gestured invitations it was impossible to persuade the islanders to join our ship and come aboard, we gave several of the older men some silver and bronze medallions, and began to sail along the islands in order to survey them. We thought that perhaps the inhabitants would go to the Discovery if we left them, but they returned directly home, and soon the Discovery also put to sail and began to follow us.. . .
We asked the leader of the expedition whether we should regard these islands as new discoveries, and received the answer that “If they have not been seen by anyone before us, then we will name them the ‘Islands of Good Intent’.” We thanked them for the honour, assuring them that these islands are certainly new, for they are not marked on the latest maps of Arrowsmith which we had with us, and indeed there were no islands shown anywhere near them. . . (Lazarev 1950:163-8, trans, David Christian).
The two vessels sailed northward and crossed the equator five days later without having sighted any other Tuvalu islands.
The next record of a European ship in Tuvalu is the Nantucket whaler Independence II under Captain George Barrett, apparently the first whaler to hunt the waters around Tuvalu and the Gilbert Islands. Barrett discovered Nukulaelae, bartering coconuts from the islanders in November 1821; he named the island Mitchell's Group after the owner of his ship. Barrett also rediscovered uninhabited Niulakita (Mendaña's La Solitaria) and added it to the charts in its correct position, calling it Rocky Island (Maude 1968:124-5; Sharp 1960:195). This name was never much used, but Independence Island, after Barrett's ship, was one of several names which came into general use for Niulakita during the 19th century (see Figure 2). Barrett did not record sighting any islands in northern Tuvalu.
But the next (recorded) European vessel in Tuvalu waters apparently, albeit unknowingly, discovered Nanumanga. Louis Isidor Duperrey, captain of the Coquille on a French Government exploring expedition, left Rotuma early in May 1824. On May 9 he sighted an island which he calculated to lie - 185 at 6° 5′ 33″S, 176° 20′E, 13 which he identified as Mourelle's Gran Cocāl (Duperrey nd.: 45; Dunmore 1969:145). It is clear from the fairly accurate position report that the island was Nanumanga. Duperrey seems to have had no contact with the islanders. The Coquille hove to for the night in order to be able to inspect Mourelle's San Augustin in daylight, and the next morning the ship passed along the south-west, or leeward, side of that atoll. Again no contact was made with the inhabitants, although an officer aboard (Lesson 1839: 446) noted that several of them in a canoe tried but were unable to catch up with the fast-moving Coquille. Another officer drew a plan of the island which indicates the ship's course and the location of the village on San Augustin (Duperrey 1827:15, and Plate 21). Duperrey's position for the village was 5° 39′ 08″S, 176° 06′E, almost exactly today's official position for Nanumea.
It is noteworthy that Duperrey had been forced to heave to at night in order not to pass San Augustin. Early the next morning he was off the SW coast of Nanumea, and by 9:30 a.m. had continued on his course north. The contrast with Mourelle's crossing from Gran Cocal to San Augustin, which took a full 24 hours, makes it even more certain that Mourelle's Cocal was Niutao, while Duperrey was, in all likelihood, the actual discoverer of Nanumanga. But Duperrey, like most of his fellow navigators of the 19th century, believed that Nanumanga and Nanumea were the pair of islands named Gran Cocal and San Augustin by Mourelle. Consquently, he never claimed nor was given credit for the discovery of Nanumanga. Otherwise it would probably have been known as Coquille Island or Duperrey Island to 19th century navigators.
To finally correct the record we need to note that since Mourelle discovered Niutao in 1781, the sighting of that island by Captain Obed Starbuck of the Nantucket whaler Loper in 1825 (Stackpole 1953:346; Sharp 1960:204-5; Maude 1968:126) was not a discovery but a rediscovery. We need not remove Starbuck's name from the list of European discoverers of Tuvalu, however, for in late 1825 his sighting of Vaitupu, which he named Tracy Island after his first mate, was apparently the first instance of a European chancing upon that island. 14
A CONFUSION OF ISLANDS
The confusion of identifications and island names in northern Tuvalu continued for much of the 19th century. Patterson, who sighted Nanumea in 1809, was unwittingly the cause of some of the mix-up since he did not realise the island was Mourelle's San Augustin, and renamed the two main islets Taswell's and Sherson's. Taswell's Isle was quickly equated with San Augustin, but Sherson's Isle was more problematical. The difficulty lay in the brief published account of Patterson's sighting which was ambiguous and enigmatic: - 186
Taswell's Isle, west side, lat. 5° 37′S. long by Sun and Moon and chronometer 176° 9′ 34″E.
Sherson's Isle, about S.S.E. of the above four or five leagues, more extensive. These islands appear well wooded, very low, and cannot be seen above six or seven leagues in the clearest weather from the mast-head: they lie in a N.W. and S.E. direction (Naval Chronicle, 24 : 313). 15
It is easy to see how cartographers and mariners of the time could interpret this report as a reference to two separate islands (they did so, as we shall see), since they are given different names and said to lie four or five leagues apart (twelve to fifteen miles). Maude (1968:107) feels, as we do, that Patterson was referring to Lakena and Nanumea islets of Nanumea atoll, which are separated by only about 2.5 miles of reef and lagoon. At high tide the reef connecting Lakena islet with the rest of Nanumea is submerged and from the deck of a ship the atoll could therefore easily have been mistaken for two separate reef islands. This interpretation is supported by the common 19th century practice of referring to atolls in the plural, as “Groups” or clusters of islets around a lagoon, instead of as single entities. It would also be correct to say of Lakena and Nanumea that “they lie in a N.W. and S.E. direction”, as Patterson does, and to note that the island to the south-east was “more extensive”, as Nanumea islet clearly is. But one could hardly speak of Nanumea and Nanumanga, separated by 37 miles of open ocean, as lying “in a N.W. and S.E. direction”. Similarly, if the island to the south-east had been Nanumanga, Patterson would not have called it “more extensive”, since it is considerably smaller than Nanumea.
Probably the most troublesome part of Patterson's description, and the main cause of the later confusion, is his statement that four or five leagues separated Taswell's and Sherson's. What would seem to have happened, however, is that the “four or five leagues” of the published account should have read “four or five miles”, but was misprinted. If we accept this as the likely explanation, then the enigma is cleared up, and Patterson's Taswell's and Sherson's are clearly the two large islets of Nanumea atoll. 16
But the ambiguity in the published account of Patterson's sighting led map-makers, navigators and sailing directory compilers to mis-identify Sherson's Isle as first one, then the other, of the remaining two islands in northern Tuvalu. Thus, when Purdy (1814:153) and Krusenstem (1824:23-4), who worked solely from charts, equated Mourelle's Gran Cocal with Sherson's Isle, they were arguing that Cocal was Niutao (since this was the island actually sighted by the Spaniard on May 5, 1781). But when Duperrey (nd.:45), Lesson (1839:446), Dumont d'Urville (1834-5, II:440-1), and Findlay (1851:998, 1863:598-9, 1871:668) also identified Cocal as Sherson's Isle, the Cocal they meant was Duperrey's unwitting discovery, Nanumanga.- 187
HUDSON ISLAND (NANUMANGA)
In 1841 the nomenclatural confusion which had arisen over the names and identities of Nanumea, Nanumanga and Niutao was simplified when the Peacock and the Flying Fish, two vessels of the United States Exploring Expedition, came upon Nanumanga. Lieutenant William Hudson in command of the Peacock carefully surveyed the island (without landing), and, not finding it on his charts, accepted it as a new discovery. It is ironic that a complex situation of mis-identifications could be simplified by yet another misidentification, but this is what happened. On the completion of surveying activities on March 24, 1841 Hudson (nd.:168) noted in his journal:
This Island is a Discovery — and lies 20 miles to the Northward [sic] of the Cocal of Maurelle — in 1781 — and the Sherson of Brig Elizabeth 1809 — as laid down on Arrowsmith's chart. 17
Hudson's calculations placed the centre of the island at 6° 19′ 30″S, 176° 25′ 30″E, which is just five miles east of the present official position for Nanumanga.
A look at Arrowsmith's last chart of the Pacific, the 1832 edition, clarifies Hudson's statement that the new “discovery” lay 20 miles southward of Cocal and Sherson “as laid down on Arrowsmith's chart.” Arrowsmith had placed Sherson (and Gran Cocal) at 5° 57′S, 176° 30′E, much too close to St. Augustine; this was north of Duperrey's reported position by some 8 miles, and fully 15 miles further north than the position reported by the Russian Khromchenko some years later. So with the chart showing only empty sea, Hudson was perhaps justified in feeling he had made a new discovery. Charles Wilkes, leader of the United States Exploring Expedition, later named the island after Lieutenant Hudson (Wilkes 1845, V:44).
Following this latter-day rediscovery, the name Hudson Island for Nanumanga gradually came into general use by chart-makers and mariners, replacing the previous Gran Cocal, Isla del Cocal, and Sherson's Isle. Hudson's exacting survey was definitive, and established without doubt the position of Hudson Island on future charts. In the whaling logs Nanumanga was still often referred to as Sherson's Isle, and Nanumea as Taswell's, 18 but for the most part the two islands were known to sailors throughout the rest of the 19th century as Hudson and St. Augustine, the anglicised form of Mourelle's San Augustin.
The Wilkes expedition bestowed another name on a Tuvalu island which also endured for much of the 19th century. The day previous to surveying Hudson Island the Peacock and Flying Fish had come upon another island in northern Tuvalu, and although it appeared on the charts, it was unnamed. Wilkes (1845, V:44) notes:
In latitude 6° 10′S, and longitude 177° 41′E they passed a small island - 188 which has no lagoon . . . and although it appears to have been seen before, yet as the charts only designate it as an island, I have bestowed upon it the name of Speiden, after the purser of the Peacock.. . .
This was the island of Niutao which, as we have seen, was discovered by Mourelle in 1781, and again by Starbuck in the Loper in 1825.
Niutao, in fact, had other names besides Speiden Island and Loper Island. In 1835 it was seen by another whaler, the L.C. Richmond, and placed in 6° 18′S, 177° 45′E. Not finding it on his chart, the captain thought he had made a new discovery and named the island after his ship (Anon nd.: entry for August 17, 1835). Niutao was also sometimes called Lynx Island but we are unable to explain the origin of this name.
GRAN COCAL — FROM ISLAND TO SHOAL
The name Gran Cocal, however, continued to appear on the charts for the rest of the century. Instead of marking an island, though, Gran Cocal came to signify an elusive shoal that was thought to lie half-way between Nanumanga and Nanumea. Just how this non-existent shoal came to be marked on the charts has been somewhat of a mystery, but it is now possible to trace the series of events that led to its creation.
The Pacific Islands Pilot (1943:433) gives one clue to the evolution of Gran Cocal shoal (or “Grand” Cocal as it was sometimes written) out of Gran Cocal island:
In 1781, Maurelle is reported to have sighted an island in lat. 6° 05′S., long. 176° 13′E., naming it Gran Cocal; it was afterwards reported by Captain Chramtschenko [sic] to be in lat. 6° 12°′S., long. 176° 13′E. The neighbourhood has frequently been passed over since, but it has not been seen.
Mourelle's supposed position cited here is in reality that reported by Nanumanga's discoverer Duperrey, since Mourelle's published narrative did not give co-ordinates for either Tuvalu island he discovered. Captain V. S. Chromchenko, in command of the Russian American Company's Elena, sighted what he took to be Mourelle's Gran Cocal in 1829 and his position, fixed with the aid of no fewer than eight chronometers, is slightly closer than that of Duperrey, but still some eight miles north-west of the actual location of Nanumanga (Krusenstern 1835:8; Meinicke 1868:125; Lebedev and Grekov 1967:198). Still, this is a fairly accurate fix for that early date, and should not in itself have led to a search later for a non-existent island.
In actual fact, a nearly correct explanation for the charting of an island between Nanumea and Nanumanga was given as early as 1868. Writing of the exploration of the “Lagoon Islands” (Tuvalu), Meinicke (1868:125n.) astutely noted that the problem arose in this manner: - 189
Duperrey located [Nanumanga] at 6° 5′ south latitude and the same longitude [as Khromchenko's position, namely 183° 47′E], Hudson [placed it at] 6° 19′ south latitude and 183° 37′ east longitude.
Therefore many maps have a separate Hudson Island to the south of Gran Cocal, which cannot exist, inasmuch as all three discoverers saw only one island here.
Meinicke is essentially right but goes astray in two particulars: he gives too high a figure for Hudson's longitude (all of Meinicke's longitudes are 6-7° too high, perhaps to adjust them to a standard other than Greenwich), and it is not Hudson Island which therefore cannot exist, but Gran Cocal, which was now left on the charts north of Hudson Island and south of St. Augustin. Gran Cocal island thus remained in essentially the position it occupied on Arrowsmith's authoritative 1832 chart, its location an amalgam of that reported by Duperrey and Khromchenko for Nanumanga, and apparently moved slightly north to take account of Patterson's ambiguous position report for Sherson's Isle. With the publication of the Wilkes' Expedition chart in 1845 (Figure 2) Gran Cocal Island had shifted from the realm of reality (albeit elusive reality) to that of imagination. While Meinicke had been basically correct in pointing out this error as early as 1868, his explanation passed unnoticed, and charts of the northern Tuvalu continued to show an extra island between Nanumea and Nanumanga. This error began to be phased out of the charts only after 1877 when a similar correction appeared in the 4th edition of Findlay's standard sailing directory (Findlay 1877:754), the predecessor of the Pacific Islands Pilot.
In the meanwhile, Gran Cocal attracted the attention of the Royal Navy and became the object of several fruitless searches, beginning in 1872 when Captain Moresby of H.M.S. Basilisk was instructed to look for Gran Cocal to the north of Nanumanga. Moresby (1872), of course, found no new island, but he did pick up some new information about a shoal. According to Findlay (1877:754) and the Pacific Islands Pilot (1891:236), Moresby heard from an English trader living on Nanumea about “an extensive shoal 3 or 4 miles in circumference [located] somewhere between Nanumanga and Nanumea with depths of from 5 to 7 fathoms on which the sea breaks in heavy weather”. This trader, whose name was Tom Day (Hayter nd.: entry for July 21, 1872), also said “that he had frequently sailed over the shoal and seen the bottom”.
When Moresby reported this information to the British Admiralty, the position previously allotted to Gran Cocal island was now assigned to a shoal. If Mourelle's Gran Cocal island could not be found, perhaps Gran Cocal shoal could be. In 1886 H.M.S. Miranda passed between Nanumanga and Nanumea and looked for the shoal. Commander Rooke (1886:4) reported:
I passed over the supposed position of the eastern end of the Gran Cocal - 190 Shoal, but got no soundings, nor saw any discolored water. The natives of Hudson Island informed me that they got 10 fathoms on the shoal.
With even the natives of the neighbouring island now apparently confirming its existence, Admiralty concern for Gran Cocal shoal went on. Visiting Tuvalu in 1892 to declare a British Protectorate, Captain Gibson (1892) of H.M.S. Curacoa could find no shoal or shallow water between the two islands. When H.M.S. Torch reported no sign of it in 1904, matters apparently rested there for the time being and a later edition of the Pacific Islands Pilot (1943:434) noted that “the natives of Nanumanga and Nanumea islands deny its existence, which is therefore considered doubtful.” This should perhaps have been the end of the rather lively career of non-existent Gran Cocal shoal, but in fact Royal Navy vessels have again searched for it several times since 1956, though they found nothing (E. V. Ward, personal communication, January 1976).
What then of this Gran Cocal shoal, or at least of the bank reported in some detail to Moresby by the English trader on Nanumea, and which was also mentioned by islanders when the Torch visited the vicinity in 1904? In fact, there is such a bank, although it lies very close to Nanumanga and not between the two islands. As described in Ward's Sailing Directions of the Gilbert and Ellice Islands:
Te Akau a Talie, with a least depth of six fathoms lies 015° 2.5 miles from the northern end of Nanumanga. The bank is 6 cables long by 2 cables wide ca. 3600 ft. and easily sighted in usual conditions. It is a useful anchorage for vessels working Nanumanga and awaiting high tide (Ward 1967:51).
Maude (1968:94) is undoubtedly right that it was this bank of which Moresby and others heard and mistakenly thought to be the missing Gran Cocal. The bank, a traditional fishing ground for Nanumanga canoes, is commemorated in several of Louis Becke's stories, notably “The Rangers of the Tia Kau”, which vividly portrays the dangers of fishing there (Becke 1894:56-65, 1897: 29-40, 1909:80-1).
To mariners an uncharted shoal is a serious risk, of course, so it is perhaps natural that Gran Cocal, a shoal so much written about in the pilots, speculated over, and searched for by naval vessels, is not easily consigned to oblivion. Thus Gran Cocal, the shoal which never existed, is still mentioned in a recent sailing guide (Ward 1967:51-2), which speculates that, even if it cannot be found today, it may perhaps have existed at some time:
There is a strong tradition amongst the people of Nanumea and Nanumanga that a large four fathom shoal lies midway between the two islands. In the last century this shoal was reported by mariners and traders and until recently was marked on the chart and described in the - 191 pilot.. . . No seamen at present in the group have seen this bank.. . . If the bank did exist it is now gone and the area between Nanumea and Nanumanga is clear of shoals.
But we have seen how the “native traditions” and the Pacific Islands Pilot descriptions originated. The evidence presented shows clearly that Gran Cocal shoal was actually created by a series of European navigational and charting errors which became enshrined on the charts. The belief that local traditions spoke of such a shoal seems likewise to have been inadvertently fostered by published accounts which appeared in an early edition of the pilot. There seems to be no substantial basis to these reports, and recent research on Nanumea (by K. Chambers) has failed to turn up any local knowledge or tradition of such a shoal, thus confirming the pilot's later note that the natives of the two islands denied the shoal's existence.
Gran Cocal shoal, its origin obscured in a long series of minor miscalculations, thus lingered on into the 20th century as a navigator's enigma, while the nine more substantial islands of Tuvalu found their way calmly on to the charts. For the most part, in fact, the charting of Tuvalu had been accomplished, with considerable accuracy, by the mid-1800s (although as late as 1872 Moresby complained that Niutao was shown fully 40 miles out of position on his charts). The following table provides a revised assessment of the European discoverers of Tuvalu.
The gradual process by which Gran Cocal was demoted from an island to an elusive shoal in fact epitomises the period of early European contact with Tuvalu. At first there was some confusion as to which islands were which, a confusion made worse by the lack of contact with the islanders themselves and by the bestowing of European names on the islands. It was quite possible to confuse Niutao with Nanumanga if one relied only upon previous records of position, or on general descriptions of the islands as seen from the deck of a ship. If the discoverers had merely been able to ask the Tuvaluans what they called the islands then much of the confusion would have been avoided. This is, of course, what happened later. in the mid-1860s foreign traders took up residence on the individual islands and missionaries of the London Missionary Society began to be established in Tuvalu, and before long the Polynesian names for the islands were in fairly regular use among European visitors and in publications. Once this occurred it was clear that there were just nine islands and it was no longer possible to confuse one with another.
It has been apparent throughout this paper that we have followed and built on the work of H. E. Maude, whose painstaking research into early European contacts in the Pacific is a landmark of careful scholarship. We are particularly indebted to Professor- 192
Revised List of European Discoverers of Tuvalu
Maude for directing our attention to new sources and making some of them available from his own library, and also for comments on an earlier draft of this paper.
Many others have generously assisted us. We thank David Christian for translating Russian sources and Beverley Clark for translating the French ones. Robert Langdon assisted in locating microfilm copies of the Spanish manuscript sources. He also commented on an earlier draft of this paper, as did Frank Broeze, O. H. K. Spate and Anne Chambers. Gudmund Stang and Ralph Bolton assisted in translating obscure passages in the Spanish manuscripts, and Dag Sagafos provided the drawings for Figures 1 and 4. Captain E. V. Ward was generous with comments and opinions on Mourelle's Tuvalu visit. We also express our gratitude to Sr. Don Roberto Barreiro-Meiro and to Sra. Rosario Parra for assistance with Spanish archive sources, and Sr. Don Amancio Landin Carrasco for encouragement and constructive comments on an earlier draft.
Keith Chambers wishes to thank the Institute for Social Anthropology, University of Trondheim, Norway, for support facilities during much of the research and writing of this paper. Doug Munro wishes to acknowledge a general debt to Stewart Firth, his Ph.D. supervisor.- 194
1 The growing body of literature on Tuvalu history and culture includes Roberts 1958, Macdonald 1975a and 1975b, Brady 1975, A. Chambers 1975, K. Chambers et al. 1978, Munro 1978, [Wilson] 1978.
2 While strictly speaking a rediscovery, these islands were not placed accurately on the charts until their 19th century “discovery”.
3 Detailed and highly readable accounts of Mendaña's two voyages and the personalities involved are in Beaglehole 1966:39-80 and Spate 1979:119-32.
4 We have followed the original manuscript sources and the recent biography of the Spanish navigator (Landin Carrasco 1971) in rendering his name “Mourelle”, instead of the “Maurelle” given in his published narrative and most subsequent secondary sources.
5 All official positions mentioned in this paper are drawn from United States Department of the Interior 1956.
6 Both authors have lived in Tuvalu for extended periods and visited all nine islands. From the deck of a ship Niutao and Nanumanga are remarkably similar in appearance.
7 A league was commonly three nautical miles at sea, but the length varied at different times in different countries. Both Sharp (1960:3) and Brand (1967:111) note that in early Spanish accounts a league was about 3.43 nautical miles (17° leagues to one degree of longitude). The six leagues in Mourelle's narrative, some 20 miles, would have been about the maximum distance a low island was visible from the masthead.
8 Despite Mourelle's statement that he passed to the north-east of Nanumea, one reconstruction (Maude 1968: map on p.125) shows the ship passing south-west of the island — an inadvertent error but ironically correct if the frigate had been coming from Nanumanga.
9 Though a very useful compendium, Stackpole's work is not always accurate in its identifications. Corrections concerning Tuvalu can be found in Maude 1968:121-33.
10 Vasquez nd. (Museo Naval version), entries for May 4-5 and 5-6, 1781, and appended list of islands seen on the voyage. Also list appended to Mourelle nd., Sevilla Estado version. In the Vasquez list longitudes have been corrected to read east of Paris, while in Mourelle's they are given east of Manila. To further complicate matters, the pilot used as his standard the meridian of San Bernardino, near Manila, and his log positions are thus given east of that line (which the log title page says lies 3° 40′ east of Manila). We converted the Mourelle list of readings to longitude east of Greenwich as follows: Manila lies at 120° 58′ E. Thus Gran Cocal's longitude would be:
120° 58′ + list reading 54° 30′ = 175° 28′ E.
San Augustin's longitude was computed in the same manner (from the Mourelle list reading of 53° 28′ E).
11 In Arrowsmith 1978, Gran Cocal was located at approximately 6° 06′ S, 178° 42′ E, and San Augustin at 5° 42′ S, 177° 36′ E. In the 1810 edition Arrowsmith moved Gran Cocal to approximately 5° 54′ S., 176° 36′ E, and San Augustin to 5° 42′ S, 176° 18′ E, thus reducing the 70 miles separating the two islands on the earlier chart to just 22 miles.
12 Although the Russian text says “painted”, Lazarev seems to be referring to tatooing.
13 Since Duperrey's longitudes were calculated east of Paris, 2° 20′ has been added here to bring them into accord with Greenwich-based positions.
14 The log of the Loper (McCleave nd.) does not provide firm evidence for the sighting of Niutao, although this probably happened late in November 1825. We are grateful to H. E. Maude for allowing us to consult his microfilm copy of the Loper log.
15 Originally published in the Bombay Courier, August 12, 1809, and reprinted in Purdy 1814:153. Patterson's description of Sherson's Isle as SSE of Taswell's was misprinted to read NSE in the Naval Chronicle.
16 Interestingly, a sailing directory published a century ago (Findlay 1877:755 and note) arrived at a similar conclusion, equating Sherson's Isle and Taswell's Isle with Nanumea islet and Lakena respectively. We had not seen this source at the time this section of the article was written.
17 Hudson clearly meant southward of Cocal, not northward, as his position co-ordinates and the published chart of this portion of the voyage (Figure 2) make clear.
18 The whalers were, of course, using names found on their charts. A chart of the South Pacific Ocean published in London in 1849 by James Imray, for example, shows Nanumea as Taswell's and Nanumanga as Sherson's. A portion of the chart is reproduced in Pease 1962:6.
19 While strictly speaking a rediscovery, these islands were not placed accurately on the charts until their 19th century “discovery”.