An edited version appeared in Mercator's World, May/June 2000, Volume 5 Number 3. Some information not found in that feature is included on this page.—JW.
“Wee sayled away to the westward to see if wee could find those Islands called the Gallipoloes, which made the Spaniards laugh at us, telling us they were enchanted Islands, and that there was never any but one Capt. Perialto that had ever seen them, but could not come neare them to anchor at them; and that they were but shadows and no reall islands.”
— William Ambrosia Cowley, May 1684
Geologists don't have much trouble dating the various members of the Galápagos Islands group, from the youngsters at its western boundary to the old-timers back east. Map collectors don't have it quite so easy though, even though the oldest known map is still well short of celebrating its 500th birthday. Generally speaking, maps of Galápagos — or of anywhere else for that matter — can be placed in one of three categories when it comes to the dating game. The first takes in the earliest survivors, which often bear neither date nor other specific information. One can only compare such maps to others thought to be of the same era, or look for clues on the map itself. The job is made easier if the cartographer is identified, in which case something might be found in the written historical record to help date the map. Finally, there are the “easy” ones, where publication details are printed on the sheet for all to see. But even here there may be uncertainty if other details on the same map contradict that information. Needless to say, one often finds maps of a favorite place in each of these categories, and although there may not be sufficient information to put a definitive date stamp on every one of them, sometimes they can at least be placed in chronological order.
The Galápagos Islands are no exception. Perhaps their earliest public appearance may be found on a vellum chart fragment — one of two presented to the Library of Congress in the late 1920s by the American philanthropist Edward Stephen Harkness (1874-1940). The Library's complete Harkness Collection of documents from colonial Mexico and Peru spans much of the sixteenth and seventeenth century, but there is nothing in the collection to indicate the date of either chart. The one showing Galápagos covers the Pacific coast of Central and northern South America and measures about 33.5 × 61.5 cm. The appearance of part of a compass rose at the lower border suggests that this surviving fragment is the upper portion of a considerably larger chart. What became of the rest is unknown, but it's probably safe to write if off as a casualty of the centuries, now permanently lost. The chart was tentatively dated in a Report of the Librarian of Congress for the Fiscal Year ending June 1929:
“It appears likely that it was not made until after the year 1561, because it contains the place name Landecho for a village in Guatemala. The village seems to have been named for a president of the Audiencia of Guatemala, named Landecho, who assumed office in 1561.”
The chart's general style though suggests it might be much earlier than that date — a possibility not necessarily ruled out by the Landecho place name. In the absence of reliable information, we may speculate that el Presidente was named after his village, and not vice versa. For example, Fray Tomás de Berlanga (Archbishop of Panamá ca. 1535) is known by the name of his native village of Berlanga in Spain. If the same might be said of Señor Landecho, then the village could have existed long before he took office.
There is some circumstancial evidence to suggest this is indeed the case: The Audiencia (High Court) of Guatemala was created in about 1540, and a Juan Nuñez de Landecho served as its president ca. 1559 or earlier. Also, the Archivo General de Indias contains a reference to a “gobernador Juan Martínez de Landecho de 1563-1568.” In both cases, the “de Landecho” style suggests the men were from the village of the same name, and that would allow the name to appear on a chart drawn before either one of them were in office.
Considering that the chart is relatively unknown, it's not surprising that it does not show up in the literature of Galápagos. What is surprising is that much of that literature credits Abraham Ortelius as the first to name and place the islands on a map in his 1570 Theatrum Orbis Terrarum atlas. Actually, his friend and colleague Gerard Mercator beat him to it by one year: the celebrated “Mercator Projection” of 1569 shows two island groups separated by slightly more than four degrees of longitude, but both labeled “y: de los galopegos.” Ortelius apparently referred to Mercator's map when preparing his atlas, although he did make a few changes. His Typus Orbis Terrarum world map follows Mercator's dual-label example and the longitudinal spacing is about the same. But in the same atlas, his Americae Sive Novi Orbus, Nova Descriptio map of North and South America shows only 2½ degrees of separation between the groups. The Southwest cluster retains Mercator's “galopegos” label, while the Northeast cluster becomes “galepegos.”
Now why did these friends draw two archipelagos separated by several hundred miles yet bearing the same name, why did Ortelius introduce a spelling variation in one label only, and why the reduced spacing on his America Sive sheet? Was the revised second label the sixteenth-century equivalent of a “typo” or was it deliberate, for the sake of differentiating one group from the other? Whatever the answer, the galopegos/galepegos nomenclature survived multiple printings, but was at last fixed — along with the famous bulge in South America — in the 1587 printing of Americae Sive, when Ortelius re-labeled the Northeast cluster as “Ins. de Cocos” and of course that name represents the present Isla del Coco.
Presumably Ortelius used Mercator as his original source for the Galápagos name, but Mercator's own source remains unknown. Both cartographers referred to Spanish, Portuguese and other resources, but probably not to the vellum chart now at the Library of Congress, where a smaller cluster north and slightly west of Galápagos is absent on their maps. But then, from what source did Mercator take his Galápagos? Did he see an actual chart, or hear a tale told by a Pacific voyager? Until the question can be answered — which may be never — the Mercator Projection retains its second-place position, followed closely by Ortelius. As for the “Cocos” correction which is unique to Ortelius, it probably derives from a resource which came to his attention in time to be incorporated into the 1587 edition of Americae Sive, but not available before that. The actual longitudinal distance between Galápagos and Coco is about 3½ degrees, and the former is 10 degrees from mainland South America. Both Mercator and Ortelius overestimated that distance at about 15 degrees.
Noted Ortelius scholar Marcel van den Broecke (see “Abraham Ortelius” in the May/June 1997 Mercator's World) finds no explanation — or even commentary — in the Ortelius literature on these little mysteries, and these name changes are generally ignored when dating Ortelius maps. But perhaps that's all understandable: whatever the spelling and wherever the placement, these islands were insignificant specks on a mysterious surface full of more interesting places still to be explored. So this little bit of terra incognita would have to wait another hundred or more years for closer attention. During that century, the Galápagos Islands moved around quite a bit, assisted by strong currents and weak navigation. In fact, the local Spaniards assured British buccaneer William Ambrosia Cowley in 1684 that they were “las Encantadas” or enchanted islands, seen by only one person who could not come near them. Cowley, who wasn't having any of this enchanted nonsense, promptly sailed for the islands and drew the first-known detailed chart — quaint today, but at the time the best there was. But despite some accounts to the contrary, neither he nor Ortelius ever labeled the islands “Enchanted” or “las Encantadas,” and it took another century until Robert Sayer's 1775 Map of South America showed a “Galapagos or Inchanted Is.” group. Cowley did however honor himself with a little Galápagos islet which he dubbed “Cowley's Enchanted Island.” Was the buccaneer poking fun at his Spanish informants? He doesn't say.
Cowley named several other islands, and therein begins another Galápagos mystery. At the time of his 1684 visit, Charles II sat on the British throne, and Cowley dubbed his first landfall King Charles Island. Shortly thereafter he named another island the Duke of York's in honor of the king's younger brother James. A few years earlier, Bartholomew Sharp had removed a derroterro — a volume of South American sailing directions — from a captured Spanish vessel. Now back in England and in search of a pardon for acts committed against England's on again/off again Iberian ally, he offered his little prize to King Charles. His Highness immediately recognized its strategic importance, and the buccaneer may have avoided a guest appearance at the gallows by placing the document in royal hands (see “Once a Pirate” in the May/June 1998 Mercator's World). Basil Ringrose — another buccaneer — translated the work into English, then ran into some bad luck during a subsequent “visit” to South America. On February 19th, 1686, an invasion plan went sour, and Ringrose and others were killed before they could retreat to the safety of their ship. At about the same time, Cowley and company were on the other side of the Pacific and he eventually reached England in October of the same year. During his long voyage, Cowley learned of the death of King Charles, and accordingly changed his Duke of York's Island to King James as he prepared his journal for the press.
And now to the mystery: There is a crude Galápagos chart on the last page of Ringrose's English translation of the derroterro presented to King Charles. The islands are all named. One of the names is King James Isle. How did this monarch's name find its way into a volume created when James was not yet King? Given its placement in the volume and very different style from the other illustrations, one doesn't need to see his name to recognize the chart as a later addition. But, a later addition by whom? The presence of that name rules out Ringrose, who had left England before James became King, and left this world before Cowley returned with his collection of island names. Like the source of Mercator's Galápagos information, the originator of this map remains unknown.
It's also uncertain why the unidentified artist would have retrieved the Ringrose work and gone to the trouble of sketching the islands on its last page. But here we can indulge in a little guesswork. In 1699, James Knapton published A Collection of Voyages, a slim volume of four sailing accounts, including “Captain Sharp's Journey Over the Isthmus of Darien” and “Capt. Cowley's Voyage Round the Globe.” A manuscript at the Morgan Library in New York City (ms. 3310) contains the Sharp and Cowley texts from which the Collection was produced. The Cowley section includes a Galápagos chart — nicely drawn, and close in style to the eventual chart inserted in Cowley's printed account. Perhaps at some point in the production process, the Ringrose translation was made available as background material for the Sharp account, and its blank final sheet was a convenient place to draw a preliminary sketch for Cowley's chart. The sketched placement of the islands on a grid of latitude and longitude lines is closely matched by their position in the Morgan manuscript chart, so it's quite possible that these are the very first and then second versions of this chart, followed by Herman Moll's engraved version in the printed book.
As a literary work, the Collection of Voyages did not share the fame of A New Voyage round the World written by Cowley's one-time shipmate William Dampier and also published by James Knapton. However, the one William had a Galápagos map and the other one didn't, and when the second edition of John Harris's massive Navigantium atque Itinerantium Bibliotheca was published in 1744, it contained yet another version: a chart by Emanuel Bowen derived from Moll's chart, but with added text commentary based on a subsequent voyage of Woodes Rogers. The same chart appeared in Harris's third edition published twenty years later.
The fifth and final version of Cowley's Galápagos chart appears in volume 4 of James Burney's Chronological History of the Discoveries in the South Sea or Pacific Ocean, and again in the author's History of the Buccaneers of America, both published in 1816. The chart is, in effect, a “carbon copy” (perhaps, a tracing) of the chart in Knapton's Collection, which Burney inserted to clarify a post-Cowley mix-up. James Colnett's Voyage to the South Atlantic and Round Cape Horn into the Pacific Ocean included a description of his visit to Galápagos in 1793. While there, he thought that an island not seen by Cowley was in fact the buccaneer's King Charles' Island. Since Colnett's mistake found its way onto a chart by Aaron Arrowsmith, Burney sketched the outline of Colnett's discovery in its correct position on Cowley's chart to show that it was in fact a new island, and not the one Colnett identified as King Charles' Island.
To return to Aaron Arrowsmith, his Galápagos chart was one of several he created for Colnett's book, all showing a date of 1st January, 1798 in the bottom margin. Although the same date appears on subsequent editions of the Galápagos chart, other evidence on the same charts indicates they were actually produced over an interval of more than twenty years, with the one included in the book as the first in the series. The outline of Charles Isle is incomplete on this chart, only a small fragment of Norfolk Isle is shown, and Arrowsmith's address is given as Charles Street, Soho.
The second edition of this chart displays no changes other than Arrowsmith's address, which is now given as Rathbone Place. Other Arrowsmith maps showing this address (often, 24 Rathbone Place) are dated in the early 1800s, and therefore this chart is assigned a date of 1805 pending additional research.
The Charles and Norfolk Isle outlines remain incomplete on the third edition of the chart, but “Careening Place. Water and plenty of Wood” is added at the North end of Charles. James Burney states that Arrowsmith added this and other details based on information from another source that remains unidentified. This is likely, since there is no evidence that Colnett actually visited this general area. In fact the track of his ship, illustrated on the four published versions of the chart, shows he passed well to the South of this island. As already noted, the chart bears the same date as the one in the book, but the address is now given as 10 Soho Square. It is known that Arrowsmith moved to this location in 1808.
Next in line is a draft version presumably drawn in advance of the fourth edition. Its title is written in a plain hand, several new details are added and the publisher and date are omitted. The draft shows a Dower's island — the present Genovesa — for the first time, and the incomplete Norfolk fragment is replaced by a rectangular-shaped Indefatigable. The new shape closely resembles a Porter's Isle on a chart drawn by John Fyffe, captain of HMS Indefatigable which visited Galápagos in 1815. The legend “Post Office” is added to Charles Isle, and this marks the first known appearance of the now-famous Galápagos landmark on a map. Captain David Porter, namesake of Porter's Isle and captain of the United States frigate Essex, did mention it in his own Journal of a Cruise (1815), where he refers to it as “… a box nailed to a post, over which was a black sign, on which was painted Hathaway's Postoffice.” He did not however show it on his own map of the island, nor did he speculate on Hathaway's identity. Captain Colnett is sometimes cited as the originator of the Post Office, but since Arrowsmith's charts and Colnett's own text rule out this possibility, about all we can say is that Hathaway probably set up his post office on a post a year or so before Porter found it, for it is unlikely that the painted sign would have survived much longer than that.
The fourth and final edition of Arrowsmith's Galápagos chart incorporates all the changes found on the draft edition. The cartouche includes “Additions & Corrections to 1817” which suggests the draft was prepared in that year or shortly thereafter. The address is again given as 10 Soho Square, but this time it is followed by “Hydrographer to His Majesty.” Aaron Arrowsmith was appointed Hydrographer in 1820. From this evidence it would seem that Arrowsmith's first Galápagos chart was indeed published in 1798, but the final version was not issued until 1820 or slightly thereafter. The explanation for the unvarying date despite the passage of almost a quarter-century is the same as that given for Mercator's Galápagos informant, and for the preliminary sketch in the Ringrose volume: “unknown.”
No doubt other maps await our discovery, and sooner or later Messrs. Mercator and Ortelius may have to surrender the second- and third-place honors they hold in this little chronology. Perhaps there's even another chart to pre-date the one at the Library of Congress. As for the others mentioned here, so many maps were produced in the interval between the last Ortelius and the first Arrowsmith that these can't be given a numerical place. But that certainly doesn't make the effort to resolve the mysteries surrounding Galápagos Islands cartography any less fascinating. Perhaps the Spanish had it right when they called this place enchanted.
NOTE: A 1524 globe by Johannes Schöner displays a group of four islands labeled “Insuls Gemmarum.” Although that name is not otherwise associated with Galápagos, the placement of the group suggests that it may indeed represent the first appearance of the Galápagos Islands. Or it may not, but that's part of the enchantment.