Discoveries

ON THE

PACIFIC COAST OF NORTH AMERICA.

BY THE EDITOR.

THE cartographical history of the Pacific coast of North America is one of shadowy and unstable surmise long continued. The views of Columbus and his companions, as best shown in the La Cosa and Ruysch maps, precluded, for a considerable time after the coming of Europeans, the possibility of the very existence of such a coast; since their Asiatic theory of the new-found lands maintained with more or less modification a fitful existence for a full century after Columbus. In many of the earliest maps the question was avoided by cutting off the westerly extension of the new continent by the edge of the sheet; but the confession of that belief was still made sometimes in other ways, as when, in the Portuguese portolano, which is placed between 1516 and 1520, Mahometan flags are placed on the coasts of Venezuela and Nicaragua.

In 1526 a rare book of the monk Franciscus, De orbis situ ac descriptione Francisci epistola, contained a map which represented South America as a huge island disjoined from the Asiatic coast by a strait in the neighborhood of Tehuantepec, with the legend, “Hoc orbis hemisphærium cedit regi Hispaniæ.” A few years later we find two other maps showing this Asiatic connection,—one of which, the Orontius Finæus globe, is well known, and is the earliest engraved map showing a return to the ideas of Columbus. It appeared in the Paris edition of the Novus Orbis of Simon Grynæus, in 1532, and was made the previous year. It is formed on a cordiform projection, and is entitled “Nova et integra universi orbis descriptio.” It is more easily understood by a reference to Mr. Brevoort’s reduction of it to Mercator’s projection, as shown in another volume. The same map, with a change in the inserted type dedication, appeared in the Pomponius Mela of 1540, and it is said also to be found much later in the Geografia of Lafreri published at Rome, 1554-1572.

The other of the two maps already referred to belongs to a manuscript, De Principiis Astronomiæ, preserved in the British Museum among the Sloane manuscripts. It closely resembles the Finæus map. The authorities place it about 1530, or a little later. In 1533, in his Opusculum Geographicum, Schöner maintained that the city of Mexico was the Quinsay of Marco Polo; and about the same time Francis I., in commissioning Cartier for his explorations, calls the St. Lawrence valley a part of Asia.

What is known as the Nancy Globe preserved the same idea, as will be seen by the sketch of it annexed, which follows an engraving published in the Compte Rendu of the Congrès des Américanistes.

The same view is maintained in a manuscript map of Ruscelli, the Italian geographer, preserved in the British Museum. Perhaps the earliest instance of a connection of America and Europe, such as Ruscelli here imagines, is the map of “Schondia,” which Ziegler the Bavarian published in his composite work at Strasburg in 1532, in which it will be observed he makes “Bacallaos” a part of Greenland, preserving the old notion prevailing before Columbus, as shown in the maps of the latter part of the fifteenth century, that Greenland was in fact a prolongation of northwestern Europe, as Ziegler indicates at the top of his map, the western half of which only is here reproduced.

In this feature, as in others, there is a resemblance in these maps of Ziegler and Ruscelli to two maps by Jacopo Gastaldi, “le coryphée des géographes de péninsule italique,” as Lelewel calls him. These maps appeared in the first Italian edition of Ptolemy, published at Venice in 1548.

The first (no. 59), inscribed “Dell’universale nuova,” is an elliptical projection of the globe, showing a union of America and Asia, somewhat different in character of contour from that represented in the other (no. 60), a “Carta Marina Universale,” of which an outline sketch is annexed.

This same map was adopted (as no. 2) by Ruscelli in the edition of Ptolemy which he published at Venice in 1561, though in the “Orbis descriptio” (no. 1) of that edition Ruscelli hesitates to accept the Asiatic theory and indicates a “littus incognitum,” as Gastaldi did in the map which he made for Ramusio in 1550.

Wuttke has pointed out two maps preserved in the Palazzo Riccardi at Florence, which belong to about the year 1550, and show a similar Asiatic connection. The map of Gaspar Vopellius, or Vopellio (1556), also extended the California coast to the Ganges. It appeared in connection with Girava’s Dos Libros de Cosmographia, Milan, 1556, but when a new titlepage was given to the same sheets in 1570, it is doubtful if the map was retained, though Sabin says it should have the map. The Italian cartographer, Paulo de Furlani, made a map in 1560, which according to Kohl is preserved in the British Museum. It depicts Chinamen and elephants in the region of the Mississippi Valley.

From Kohl’s sketch, preserved in his manuscript in the library of the American Antiquarian Society, the annexed outline is drawn. Furlani is reported to have received it from a Spanish nobleman, Don Diego Hermano, of Toledo. The connection with Asia is again adhered to in Johannes Myritius’s Opusculum geographicum, where the map is dated 1587, though the book was published at Ingolstadt in 1590. Just at this time Livio Sanuto, in his Geografia distinta (Venice, 1588), was disputing the Asiatic theory on the ground that the Mexicans would not have shown surprise at horses in Cortés’ time, if they had formerly been inhabitants of a continent like Asia, where horses are common. Perhaps the latest use of the type of map shown in the “Carta Marina” of 1548 was just a half century later, in 1598, in an edition of Ortelius, Il Theatro del mondo, published at Brescia. The belief still lingered for many years yet in some quarters; and Thomas Morton in 1636 showed that in New England it was not yet decided whether the continent of America did not border upon the country of the Tartars. Indeed, the last trace of the assumption was not blown away till Behring in 1728 passed from the Pacific to the Arctic seas.

Such is in brief the history of the inception and decline of the belief in the prolongation of Asia over against Spain, as Toscanelli had supposed in 1474, and as had been suspected by geographers at intervals since the time of Eratosthenes. The beginning of the decline of such belief is traced to the movements of Cortés. Balboa in 1513 by his discovery of the South Sea, later to be called the Pacific Ocean, had established the continental form of South America, whose limits southward were fixed by Magellan in 1520; but it was left for Cortés to begin the exploration to the north which Behring consummated.

After the Congress of Badajos had resolved to effect a search for a passage through the American barrier to the South Sea, the news of such a determination was not long in reaching Cortés in Mexico, and we know from his fourth letter, dated Oct. 15, 1524, that it had already reached him, and that he had decided to take part in the quest himself by despatching an expedition towards the Baccalaos on the hither side; while he strove also to connect with the discoveries of Magellan on the side of the South Sea. Cortés had already been led in part by the reports of Balboa’s discovery, and in part by the tidings which were constantly reaching him of a great sea in the direction of Tehuantepec, to establish a foothold on its coast, as the base for future maritime operations. So his explorers had found a fit spot in Zacatula, and thither he had sent colonists and shipwrights to establish a town and build a fleet, the Emperor meanwhile urging him speedily to use the vessels in a search for the coveted strait, which would open a shorter passage than Magellan had found to the Spice Islands. But Cortés’ attention was soon distracted by his Honduras expedition, and nothing was done till he returned from that march, when he wrote to the Emperor, Sept. 3, 1526, offering to conduct his newly built fleet to the Moluccas.

But two other fleets were already on the way thither,—one under Garcia de Loaysa which left Spain in August, 1525, and the other under Sebastian Cabot, who stopped on the way at La Plata, had left in April, 1526. So Cortés finally received orders to join with his fleet that of Loaysa, who had indeed died on his voyage, and of his vessels only one had reached the Moluccas. Another, however, had sought a harbor not far from Zacatula, and had brought Cortés partial tidings at least of the mishaps of Loaysa’s undertaking. What information the rescued crew could give was made use of, and Cortés, bearing the whole expense, for a reimbursement of which he long sued the home Government, sent out his first expedition on the Pacific, under the command of his cousin Alvaro de Saavedra Ceron, armed with letters for Cabot, whose delay at La Plata was not suspected, and with missives for sundry native potentates of the Spice Islands and that region.

After an experimental trip up the coast, in July, 1527, two larger vessels and a brigantine set sail Oct. 31, 1527. But mishap was in store. Saavedra alone reached the Moluccas, the two other vessels disappearing forever. He found there a remnant of Loaysa’s party, and, loading his ship with cloves, started to return, but died midway, when the crew headed their ship again for the Moluccas, where they fell at last into Portuguese prisons, only eight of them finally reaching Spain in 1534.

It will be remembered that the Portuguese, following in the track of Vasco da Gama, had pushed on beyond the great peninsula of India, and had reached the Moluccas in 1511, where they satisfied themselves, if their longitude was substantially correct, that there was a long space intervening yet before they would confront the Spaniards, pursuing their westerly route. It was not quite so certain, however, whether the line of papal demarcation, which had finally been pushed into the mid-ocean westerly from the Azores, would on this opposite side of the globe give these islands to Spain or to themselves. The voyage of Magellan, as we shall see, seemed to bring the solution near; and if we may believe Scotto, the Genoese geographer, at about the same date (1520) the Portuguese had crossed the Pacific easterly and struck our northwest coast. The mishaps of Loaysa and Saavedra, as well as a new understanding between the rival crowns of the Iberian peninsula, closed the question rather abruptly through a sale in 1529—the treaty of Saragossa—by Spain, for 350,000 ducats, to Portugal of all her rights to the Moluccas under the bull of demarcation.

Cortés, on his return from Spain (1530), resolved to push his discoveries farther up the coast. The Spaniards had now occupied Tehuantepec, Acapulco, and Zacatula on the sea, and other Spaniards were also to be found at Culiacan, just within the Gulf of California on its eastern shore. The political revolutions in Cortés’ absence had caused the suspension of work on a new fleet, and Cortés was obliged to order the construction of another; and the keels of two were laid at Tehuantepec, and two others at Acapulco. In the early part of 1532 they were launched, and in May or June two ships started under Hurtado de Mendoza, with instructions which are preserved to us. It is a matter of doubt just how far he went, and both vessels were lost. Nuño de Guzman, who held the region to the north, obstructed their purpose by closing his harbors to them and refusing succor; and Cortés was thus made to feel the deadliness of his rivalry. The conqueror now himself repaired to Tehuantepec, and superintended in person, working with his men, the construction of two other ships. These, the “San Lazaro” and “Concepcion,” under Diego Becerra, left port on the 29th of October, 1533, and being blown to sea, they first saw land in the latitude of 29° 30´ north on the 18th of December, when, coasting south and east, they developed the lower parts of the Californian peninsula. Mutiny, and attacks of the natives, during one of which the chief pilot Ximenes was killed, were the hapless accompaniments of the undertaking, and during stress of weather the vessels were separated. The “San Lazaro” finally returned to Acapulco, but the “Concepcion” struggled in a crippled condition into a port within Guzman’s province, where the ship was seized. A quarrel ensued before the Audiencia, Cortés seeking to recover his vessel; but he prospered little in his suit, and was driven to undertake another expedition under his own personal lead. Sending three armed vessels up the coast to Chiametla, where Guzman had seized the “Concepcion,” Cortés went overland himself, accompanied by a force which Guzman found it convenient to avoid. Here he joined his vessels and sailed away with a part of his land forces to the west; and on the 1st of May, 1535, he landed at the Bay of Santa Cruz, where Ximenes had been killed. What parts of the lower portion of the Californian peninsula Cortés now coasted we know from his map, preserved in the Spanish Archives, which accompanied the account of his taking possession of the new land of Santa Cruz, “discovered by Cortés, May 3, 1535,” as the paper reads. The point of occupation seems to have been the modern La Paz, called by him Santa Cruz. The notary’s account of the act of possession goes on to say,—

“On the third day of May, in the year of our Lord 1535, on the said day, it may be at the hour of noon, be the same less or more, the very illustrious Lord don Hernando Cortés, Marquis of the Valley of Guaxaca, Captain-general of New Spain and of the Southern Sea for his Majesty, etc., arrived in a port and bay of a country newly discovered in the same Southern Sea, with a ship and armament of the said Lord Marquis, at which said port his Lordship arrived with ships and men, and landed on the earth with his people and horses; and standing on the shore of the sea there, in presence of me Martin de Castro, notary of their Majesties and notary of the Administration of the said Lord Marquis, and in presence of the required witnesses, the said Lord Marquis spoke aloud and said that he, in the name of His Majesty, and in virtue of his royal provision, and in fulfilment of His Majesty’s instructions regarding discovery in the said Southern Sea, had discovered with his ship and armament the said land, and that he had come with his armament and people to take possession of it.”

Finding his men and horses insufficient for the purposes of the colony which he intended to establish, Cortés despatched orders to the main for assistance, and, pending its arrival, coursed up the easterly side of the gulf, and opportunely fell in with one of his vessels, much superior to his own brigantine. So he transferred his flag, and, returning to Santa Cruz, brought relief to an already famishing colony.

News reaching him of the appointment of Mendoza as viceroy, Cortés felt he had greater stake in Mexico, and hurriedly returned. Not despairing of better success in another trial, and spurred on by indications that the new viceroy would try to anticipate him, he got other vessels, and, putting Francisco de Ulloa in charge, despatched them (July 8, 1539) before Guzman’s plan for their detention could be put into execution. Ulloa proceeded up the gulf nearly to its head, and satisfied himself that no practicable water passage, at least, could bring him to the ocean in that direction, as Cortés had supposed. Ulloa now turned south, and following the easterly coast of the peninsula rounded its extremity, and coursed it northerly to about 28° north latitude, without finding any cut-off on that side. So he argued for its connection with the main.

And here Cortés’ connection with discoveries on the Pacific ends; for Mendoza, who had visions of his own, thwarted him in all subsequent attempts, till finally Cortés himself went to Spain. The name which his captains gave to the gulf, the Sea of Cortés, failed to abide. It grew to be generally called the Red Sea, out of some fancied resemblance, as Wytfliet says, to the Red Sea of the Old World. This appellation was supplanted in turn by the name of California, which, it is contended, was given to the peninsula by Cortés himself.

The oldest map which we were supposed to possess of these explorations about the gulf, before Dr. Hale brought the one, already mentioned, from Spain, was that of Castillo, of which a fac-simile is herewith given as published by Lorenzana in 1770, at Mexico, in his Historia de Nueva España. Castillo was the pilot of the expedition, sent by Mendoza to co-operate by sea with the famous expedition of Coronado, and which the viceroy put under the command of Hernando d’Alarcon. The fleet, sailing in May, 1540, reached the head of the gulf, and Alarcon ascended the Colorado in boats; but Marcou thinks he could not have gone up to the great cañon, which however he must have reached if his supposed latitude of 36° is correct. He failed to open communication with Coronado, but buried some letters under a cross, which one of that leader’s lieutenants subsequently found.

In 1542 and 1543 an expedition which started under Juan Rodriguez Cabrillo, a Portuguese in the Spanish service, explored the coast as far as 44° north, reaching that point by coasting from 33°, where he struck the land. He made a port which he calls San Miguel, which Bancroft is inclined to believe is San Diego; but the accounts are too confused to track him confidently, and it is probable that Cabrillo’s own vessel did not get above 38°, for Cabrillo himself died Jan. 3, 1543, his chief pilot, Ferrelo (or Ferrer), continuing the explorations. Bancroft does not think that the pilot passed north of Cape Mendocino in 40° 26´.

Thus from the time when Balboa discovered the South Sea, the Spanish had taken thirty years to develop the coast northerly, to the latitude of Oregon. In this distance they had found nothing of the Straits of Anian, which, if Humboldt is correct, had begun to take form in people’s minds ever since Cortereal, in 1500, had supposed Hudson’s Straits to be the easterly entrance of a westerly passage.

There seems to have been a general agreement among cartographers for some years yet to consider the newly discovered California as a peninsula, growing out of the concurrent testimony of those who, subsequent to Cortés’ own expedition, had tracked both the gulf and the outer coast. The Portuguese map given by Kunstmann shows it as such, though the map cannot be so early as that geographer places its anterior limit (1530), since the development of the gulf could not have been made earlier than 1535; unless by chance there were explorations from the Moluccas, of which we have no record. The map in this part bears a close resemblance to a manuscript chart in the British Museum, placed about 1536, and it seems probable that this is the approximate date of that in Kunstmann. The California peninsula is shown in much the same way in a map which Major ascribes to Baptista Agnese, and places under 1539. It belongs (pl. iv.) to what has been sometimes spoken of as an atlas of Philip II. inscribed to Charles V., but in fact it was given to Philip by Charles. Its essential features were almost exactly reproduced in a draft of the New World (preserved in the British Museum) assigned to about 1540, and held to be the work of the Portuguese hydrographer Homem.

Apian and Münster in 1540, and Mercator in 1541, while boldly delineating a coast which extends farther north than Cabrillo had reached in 1542, wholly ignore this important feature. Not so, however, Sebastian Cabot in his famous Mappemonde of 1544, as will be seen by the annexed sketch. The idea of Münster, as embodied in his edition of Ptolemy in 1540, already referred to, was continued without essential change in the Basle edition of Ptolemy in 1545. In 1548 the “carta marina” of Gastaldi as shown on a previous page, clearly defined the peninsula, while merging the coast line above into that of Asia. The peninsula was also definitely marked in several of the maps preserved in the Riccardi palace at Florence, which are supposed to be of about the middle of the sixteenth century.

In the map of Juan Freire, 1546, we have a development of the coast northward from the peninsula, for which it is not easy to account; and the map is peculiar in other respects. The annexed sketch of it follows Kohl’s drawing of an old portolano, which he took from the original while it was in the possession of Santarem. Freire, who was a Portuguese hydrographer, calls it a map of the Antipodes, a country discovered by Columbus, the Genoese. It will be observed that about the upper lake we have the name “Bimini regio,” applied to Florida after the discovery of Ponce de Leon, because of the supposition that the fountain of youth existed thereabout. The coasts on both sides of the gulf are described as the discovery of Cortés. There seems to be internal evidence that Freire was acquainted with the reports of Ulloa and Alarcon, and the chart of Castillo; but it is not so clear whence he got the material for his draft of the more westerly portions of the coast, which, it will be observed, are given much too great a westerly trend. The names upon it do not indicate any use of Cabrillo’s reports; though from an inscription upon this upper coast Freire credits its discovery to the Spaniards, under orders from the emperor, conducted by one Villalobos. Kohl could not find any mention of such an explorer, but conjectured he was perhaps the one who before Cabrillo, as Herrera mentions, had named a river somewhere near 30° north latitude “Rio de Nuestra Señora,” and which Cabrillo sought. Kohl also observes that though the coast line is continuous, there are places upon it marked “land not seen,” with notes of its being again seen west of such places; and from this he argues that the expedition went up and not down the coast. It not unlikely had some connection with the fleet which Ruy Lopez de Villalobos conducted under Mendoza’s orders, in November, 1542, across the Pacific to the islands on the Asiatic coast.

In 1554 Agnese again depicts the gulf, but does not venture upon drawing the coast above the peninsula, which in turn in the Vopellio map of 1556, and in that in Ramusio the same year, is made much broader, the gulf indenting more nearly at a right angle. The Homem map of 1558, preserved in the British Museum, returns to the more distinctive peninsula, though it is again somewhat broadened in the Martines map of about the same date, which also is of interest as establishing a type of map for the shores of the northern Pacific, and for prefiguring Behring’s Straits, which we shall later frequently meet. Mention has already been made of the Furlani map of 1560 for its Asiatic connections, while it still clearly defined the California peninsula. The Ruscelli map in the Ptolemy of 1561 again preserves the peninsula, while marking the more northerly coasts with a dotted line, in its general map of the New World; but the “Mar Vermeio” in its map of “Nueva Hispania” is the type of the gulf given in the 1548 edition. The Martines type again appears in the Zaltieri map of 1566, which is thought to be the earliest engraved map to show the Straits of Anian.

MARTINES, 155-(?).

This sketch follows a copy by Kohl (Washington Collection) of the general map of the world, contained in a manuscript vellum atlas in the British Museum (no. 9,814), from the collection of the Duke de Cassano Serra. It is elaborately executed with miniatures and figures. The language of the map is chiefly Italian, with some Spanish traces. Kohl believes it to be the work of Joannes Martines, the same whose atlas of 1578 is also in the Museum, and whose general map (1578) agrees in latitudes and other particulars with this. The present one lacks degrees of longitude, which the 1578 map has, as well as the name “America,” wanting also in this. Kohl places it not long after the middle of the sixteenth century. In the Catalogue of Manuscript Maps, i. 29, the atlas of 1578 is mentioned as containing the following numbers relating to America: 1. The world. 2. The two hemispheres. 3. The world in gores. 10. West coast of America. 11. Coast of Mexico. 12-13. South America. 14. Gulf of Mexico. 15. Part of the east coast of North America.

In the Museum manuscripts, no. 22,018, is a portolano by Martines, dated 1579; and another, of date 1582, is entered in the 1844 edition of the Catalogue of Manuscript Maps, i. 31. Kohl’s Washington Collection includes two Martines maps of 1578.

The manuscript map of Diegus (Homem) of 1568, in the Royal Library in Dresden, gives the peninsula, but turns the more northerly coast abruptly to the east, connecting it with the archipelago, which stands for the St. Lawrence in his map of 1558.

The great Mappemonde of Mercator, published at Duisburg in 1569, in which he introduced his new projection, as will be seen by the annexed sketch, keeps to the Martines type; and while it depicts the Straits of Anian, it renders uncertain, by interposing a vignette, the passage by the north from the Atlantic to the Pacific. The next year Ortelius followed the same type in his Theatrum orbis terrarum,—the prototype of the modern atlas.

A similar western coast is defined by Porcacchi, in his L’ isole piu famose del mondo, issued at Venice in 1572.

The peninsula of California, but nothing north of it, is again delineated in a Spanish mappemonde of 1573, shown in Lelewel. The Mercator type is followed in the maps which are dated 1574, but which appeared in the Theatri orbis terrarum enchiridion of Philippus Gallæus, published at Antwerp in 1585.

In the same year the Italian cartographer Furlani, or Forlani, showed how he had advanced from the views which he held in 1560, in a map of the northern Pacific, which is annexed. It is the earliest map in which Japan has been noted as having its greatest length east and west; for Ortelius and others always give it an extension on the line of the meridian.

Sir Humphrey Gilbert’s map in 1576 gives the straits, but he puts “Anian” on the Asiatic side, and does not indicate the Gulf of California, unless a forked bay in 35° stands for it. The map in Best’s Frobisher makes the Straits of Anian connect with “Frobisher’s straightes” to give a through passage from ocean to ocean, and depicts a distorted California peninsula.

Mention has already been made on a previous page of a Martines map of 1578. It has a similar configuration to that already shown as probably the earliest instance of its type.

Of the explorations of Francis Drake in 1579 we have no cartographical record, except as it may be embodied in the globe of Molineaux, preserved in the Middle Temple, London, which is dated 1592, and in the map of the same cartographer, dated 1600. Molineaux seemingly made use of the results of Cabrillo’s voyage, as indicated by the Spanish names placed along the coast. It was one of the results of Drake’s voyage that the coast line of upper California took a more northerly trend. The map of Dr. Dee (1580) evidently embodied the views of the Spanish hydrographers.

In 1582 Popellinière repeated the views of Mercator and Ortelius; but in England Michael Lok in this same year began to indicate the incoming of more erroneous views. The California gulf is carried north to 45°, where a narrow strip separates it from a vague northern sea, the western extension of the sea of Verrazano.

After the Spaniards had succeeded, in opposition to the Portuguese, in establishing a regular commerce between Acapulco and Manilla (Philippine Islands), the trade-winds conduced to bring upper California into better knowledge. The easterly trades carried their outward-bound vessels directly west; but they compelled them to make a détour northward on their return, by which they also utilized the same Japanese current which brought the Chinese to Fusang many centuries before. An expedition which Don Luis de Velasco had sent in 1564, by direction of Philip II., accompanied by Andres de Urdaneta, who had been in those seas before with Loaysa in 1525, had succeeded in making a permanent occupation of the Philippines for Spain in 1564. It became now important to find a practicable return route, and under Urdaneta’s counsel it was determined to try to find it by the north. One of the galleons deserted, and bearing northerly struck the California coast near Cape Mendocino, and arrived safe at Acapulco three months before Urdaneta himself had proved the value of his theory. The latter’s course was to skirt the coast of Japan till under 38°, when he steered southerly; and after a hard voyage, in which he saw no land and most of his crew died, he reached Acapulco in October. Other voyages were made in succeeding years, but the next of which we have particular account was that of Francisco Gali, who, returning from Macao in 1584, struck the California coast in 37° 30´, and marked a track which other navigators later followed.

The map (1587) in Hakluyt’s Paris edition of Peter Martyr conformed more nearly to the Mercator type; and Hakluyt, as well as Lok, records Drake’s discovery, both of them putting it, however, in 1580.

With the year 1588 is associated a controversy over what purports to be a memoir setting forth the passage of the ship of a Spanish navigator, Lorenzo Ferrer de Maldonado, from the Atlantic to the Pacific, through a strait a quarter of a league wide. The passage took him as high as 75°; but he reached the Pacific under the sixtieth parallel. The opening was identified by him with the long-sought Straits of Anian. The belief in this story had at one time some strong advocates, but later geographical discoveries have of course pushed it into the limbo of forgotten things; for it seems hardly possible to identify, as was done by Amoretti, the narrow passage of Maldonado, under 60°, with that which Behring discovered, sixteen leagues wide, under 65°.

In 1592 we have the alleged voyage of De Fuca, of which he spoke in 1596, in Venice, to Michael Lok, who told Purchas; and he in turn included it in his Pilgrims. He told Lok that he had been captured and plundered on the California coast by Cavendish,—a statement which some have thought confirmed by Cavendish’s own avowal of his taking a pilot on that coast,—and that at the north he had entered a strait a hundred miles wide, under 47° and 48°, which had a pinnacle rock at the entrance; and that within the strait he had found the coast trending northeast, bordering a sea upon which he had sailed for twenty days. This story, despite its exaggerations, and though discarded formerly, has gained some credence with later investigators; and the application of his name to the passage which leads to Puget Sound seems to have been the result of a vague and general concurrence, in the belief of some at least, that this passage must be identified with the strait which De Fuca claimed to have passed.

With the close of the sixteenth century, the maps became numerous, and are mostly of the Mercator type. Such are those of Cornelius de Judæis in 1589 and in 1593, the draughts of 1587 and 1589 included in the Ortelius of 1592, the map of 1593 in the Historiarum indicarum libri XVI. of Maffeius, and those of Plancius and De Bry. The type is varied a little in the 1592 globe of Molineaux, as already shown, and in the 1587 map of Myritius we have the Asiatic connection of the upper coast as before mentioned; but in the Ptolemy of 1597 the contour of Mercator is still essentially followed. In this same year (1597) the earliest distinctively American atlas was published in the Descriptionis Ptolemaicæ Augmentum of Cornelius Wytfliet, of which an account is given in another place. Fac-similes of the maps of the Gulf of California and of the New World are annexed, to indicate the full extent of geographical knowledge then current with the best cartographers. The Mercator type for the two Americas and the great Antarctic Continent common to most maps of this period are the distinguishing features of the new hemisphere. The same characteristics pertain also to the mappemondes in the original Dutch edition of Linschoten’s Itinerario, published in two editions at Amsterdam in 1596, in Münster’s Cosmographia, 1598, and in the Brescia edition (1598) of Ortelius.

In 1600 Metullus in his America sive novus orbis, published at Cologne, simply followed Wytfliet. From the map of Molineaux, likewise of 1600, a sketch of the California peninsula is given elsewhere. A contour of the coast more like that of the Molineaux globe figured on a preceding page belongs to the map given in the Herrera of 1601, but it also introduces views which held to a much wider separation of the shores of the north Pacific than had been maintained by the school of Mercator..

An important voyage in both furthering and confusing the knowledge of the California coast was that of Sebastian Viscaino. This navigator, it is sometimes said, had been in a Manilla galleon which Cavendish had captured near Cape St. Lucas in 1587, when the English freebooter burned the vessel and landed her crew. He is known to have had much opportunity for acquiring familiarity with the coast; and in 1597 he had conducted an expedition to the coast of the California peninsula which had failed of success.

In 1602 (May 5) he was again despatched from Acapulco with three vessels, for the same purpose of discovering some harbor up the coast which returning vessels from the Philippines could enter for safety or repairs, and of finding the mysterious strait which led to the Atlantic. He was absent ten months. He himself went up to 42°, but one of his vessels under Martin Aguilar proceeded to 43°, where he reported that he found the entrance of a river or strait, not far from Cape Blanco; and for a long period afterwards the entrance and Aguilar’s name stood together on the maps. Buache, in his Considérations géographiques et physiques, says that it was the reports brought back from this expedition, describing an easterly trend of the coast above the 43°, which gave rise to the notion that the waters of the Gulf of California found a passage to the ocean in two ways, making an island of the peninsula. The official recorder of the expedition (Ascension) is known to have held this view. We shall see how fixed this impression later became.

Meanwhile the peninsular shape was still maintained in the map in Botero’s Relaciones Universales del mundo, published at Valladolid in 1603; in the Spanish map of 1604, made at Florence by Mathieu Neron Pecciolen (engraved for Buache in 1754); in that of Cespedes’ Regimiento de Navigacion (1606), and in that published in connection with Ferdinand de Quir’s narrative in the Detectionis Freti (1613) of Hudson’s voyage.

A map of Jodocus Hondius of about this time first gave indication of the growing uncertainty which led finally to a prevailing error regarding the head of the gulf. The map was inscribed “Vera totius expeditionis nauticæ Descriptio D. Franc. Draci,” etc., and illustrated Hondius’s edition of Drake and Cavendish’s voyages, and has been reproduced in the Hakluyt Society’s edition of The World Encompassed. The gulf is made to divide about an island at its northern end, producing two arms whose prolongation is left undecided. The circumpolar map of Hondius which appeared in Pontanus’s Amsterdam in 1611, and is given in fac-simile in Asher’s Henry Hudson, shows the Straits of Anian, but nothing more. Another Hondius map in the Mercator of 1613 turns the coast easterly, where the Straits of Anian separate it from Asia. The same atlas of 1613 contains also the America of Michael Mercator, which is of the usual Gerard Mercator type, with the enclosed northern sea contracted to narrow limits and called “Mare dulce.” A similar western coast is drawn in the America of Johannes Oliva of Marseilles, preserved in the British Museum.

In Kasper van Baerle’s edition of Herrera, published at Amsterdam in 1622, we get—as far as has been observed—the earliest insularizing of the California peninsula, and this only by a narrow thread of water connecting a large gulf below and a smaller one above. And even this attempt was neutralized by a second map in the same book, in which these two gulfs were not made to mingle their waters. A bolder and less equivocal severing of the peninsula followed in the maps of two English geographers. The first of these is the map of Master Briggs. In this the island stretches from 23° to 44°, showing Cape Blanco, with Cape Mendocino and “Po. Sr. Francisco Draco” south of it, the latter in about 38°. The map bears the following legend: “California, sometymes supposed to be part of ye Westerne continent; but since by a Spanish charte taken by ye Hollanders it is found to be a goodly Ilande, the length of the west shoare beeing about 500 leagues from Cape Mendocino to the south cape thereof called Cape St. Lucas, as appeareth both by that Spanish Chart, and by the relation of Francis Gaule [Gali], whereas in the ordinarie charts it is sett downe to be 1700 leagues.” The other was that given in John Speed’s Prospect, which contains one of the maps of Abraham Goos of Amsterdam, “described and enlarged by I. S. Ano. 1626.” This carries up the outer coast of the island beyond the “Po[rto] Sir Francisco Dr[ake]” and Cape Mendocino. The coast of the main opposite the northern end of the island ceases to be defined, and is continued northerly with a dotted line, while the western shore of Hudson’s Bay is also left undetermined. De Laet, however, in 1630 still kept to the peninsula, placing “Nova Albion” above it. In 1636 W. Saltonstall’s English translation of Hondius’s Mercator presents an island, with the now somewhat common break in the main coast opposite its northern end. This gap is closed up, however, in another map in the same volume.

The map in Pierre D’Avity’s Le Monde makes California a peninsula, with the river St. Lawrence rising close to it, and flowing very near also to Hudson’s Bay in its easterly passage.

The circumstantial story of Bartolemé de Fonte, whose exploits are placed in 1640, at one time commanded a certain degree of confidence, and made strange work with the cartographical ideas of the upper part of the Pacific coast. It is now believed that the story was coined by James Petiver, one of the contributors to the Monthly Miscellany, or Memoirs for the Curious, published in London in April and June, 1708, in which first appeared what purported to be a translation of a letter of a certain Admiral De Fonte. In this a Spanish navigator—whose name was possibly suggested by a veritable De Fonta who was exploring Tierra del Fuego in 1649—was made to depart from Callao, April 3, 1640, and proceed up the coast to 53°, above which he navigated a net-work of interior waters, and encountered a ship from Boston which had entered these regions from the Atlantic side. To this archipelago, as it seemed, he gave the name of St. Lazarus; and to a river, leading from a lake with an island in it, he applied that of Velasco; and these names, curiously, appear in the fanciful maps which were made by Delisle and Buache in elucidation of the voyage in which they expressed not a little faith, though the Spanish antiquaries early declared that their archives contained no record of the voyage.

The Dutch, under De Vries, in 1643 had pushed up from Japan, and discovered, as they thought, an island, “Jesso,” separated from land on the west by a water which they called the “Detroit de Vries,” and on the American side by a channel which had an uncertain extension to the north, and might after all be the long-sought Straits of Anian. The idea of an interjacent land in the north Pacific between America and Asia is also said to have grown out of the report of a Portuguese navigator, Don João da Gama, who claimed to have seen such a land in sailing from China to New Spain. It long maintained a fleeting existence on the maps.

Two maps of Petrus Koerius, dated 1646, in Speed’s Prospect (1668), indicate what variable moods geographers could assume in the same year. In one we have an island and a determinate coast line running north to the straits; in the other we have a peninsula with two different trends of the coast north of it in half-shading. We owe to an expatriated Englishman a more precise nomenclature for the western coast than we had had previous to the appearance of his maps in 1646; and the original manuscript drawings preserved at Munich are said by Dr. Hale to be richer still in names. This is the Arcano del mare of Robert Dudley. He was born in Surrey in 1573, and whether the natural or legitimate son of the Earl of Leicester depends on the proof of the secret marriage of that nobleman with Lady Sheffield. An adventurous spirit kept him away from the enjoyment of Kenilworth, which he inherited, and he was drawn nearer to the associations of the sea by marrying a sister of Cavendish. He was among the many Englishmen who tried their daring on the Spanish main. He married a second wife, a daughter of Sir Thomas Leigh, whom he abandoned, partly to be rid of a stepmother; and out of chagrin at his failure to secure the dukedom of Northumberland, which had been in abeyance since the execution of his grandfather, Lady Jane Grey’s adherent, he sold Kenilworth to young Prince Henry, and left England in company with a daughter of Sir Robert Southwell. He now gave himself up to practical seamanship and the study of hydrography. The grand-duke of Tuscany gave him employment, and he drained a morass to enable Leghorn to become a beautiful city.

Under authority of Ferdinand II., he assumed the title of Duke of Northumberland, which was recognized throughout the empire. He died in 1639. The Arcano has thirty-three American maps; but the Munich manuscript shows thirteen more. One of the Pacific coast, which records Drake’s explorations, is annexed; but with Dudley’s text there is another showing the coast from Cape Mendocino south, which puts under thirty degrees north a “golfo profondo” of undefined inland limits, with “I di Cedros” off its mouth. The bay with the anchor and soundings just north of thirty degrees, called in the fac-simile “Pto di Nouova Albion,” corresponding, it would seem, to San Francisco, is still seen in this other chart, with a more explicit inscription,—“Po: dell nuovo Albion scoperto dal Drago Cno Inglese.”

In 1649, in Texeira’s chart, there is laid down for the first time a sketch of the coast near the Straits of Anian, which is marked as seen by João da Gama, and extends easterly from Jesso, in the latitude of 50°. Gama’s land lived for some time in the charts.

We have another of Speed’s maps, five years later (1651), which appears in the 1676 edition of his Prospect, in which that geographer is somewhat confused. He makes California an island, with a break in the coast line of the main opposite its northern extremity, and its northwest point he calls “C. Mendocino,” while “Pt. Sir Francisco Draco” is placed south of it; but rather confusedly another Cape Mendocino projects from the main coast considerably further to the north. A map of Visscher in 1652 reverts, however, to the anterior notions of Mercator; but when in 1655 Wright, an Englishman, adopted Mercator’s projection, and first made it really serviceable for navigation, in his Certain Errors in Navigation, he gave an insular shape to California.

The French geographer Nicolas Sanson introduced a new notion in 1656. California was made an island with “Pto de Francisco Draco” on the west side, somewhat south of the northern cape of it. On the main the coast in the same latitude is made to form a projection to the north called “Agubela de Cato,” without any extension of the shore farther northward. The map in Petavius’s (Petau’s) History of the World (London, 1659) carries the coast up, but leaves a gap opposite the northern end of the insular California. The atlas of Van Loon (1661) converts the gap into the Straits of Anian, and puts a “terra incognita” north of it. Danckerts of Amsterdam in the same year (1661), and Du Val in various maps of about this time make it an island. The map of 1663, which appeared in Heylin’s Cosmographie, gives the insular California, and a dotted line for the main coast northward, with three alternative directions. A map of the Sanson type is given in Blome’s Description of the World, 1670. Ogilby’s map in 1671 makes it an island, following Montanus’s Nieuwe Weereld.

Hennepin had in his 1683 map made California a peninsula, and in that of 1697 he still preserved the gulf-like character of the waters east of it; but the same plate in the 1698 edition is altered to make an island, as it still is in the edition of 1704. The French geographer Jaillot, in 1694, also conformed to the insular theory, as did Corolus Allard in his well-known Dutch atlas. Campanius, copying Hennepin, speaks of California as the largest island “which the Spaniards possess in America. From California the land extends itself [he says] to that part of Asia which is called Terra de Jesso, or Terra Esonis. The passage is only through the Straits of Anian, which hitherto has remained unknown, and therefore is not to be found in any map or chart,”—all of which shows something of Campanius’s unacquaintance with what had been surmised, at least, in cartography. All this while Blaeu in his maps was illustrating the dissolving geographical opinions of his time. In 1659 he had drawn California as an island; in 1662 as a peninsula; and once more, in 1670, as an island. Coronelli in 1680, and Franquelin in his great manuscript map of 1684 had both represented it as an island.

In 1698 the English geographer Edward Wells, in his New Sett of Maps, showed a little commendable doubt in marking the inlet just north of the island as “the supposed Straits of Anian,”—a caution which Delisle in 1700, with a hesitancy worthy of the careful hydrographer that he was destined to become, still further exemplified. While restoring California to its peninsular character, he indicated the possibility of its being otherwise by the unfinished limitations of the surrounding waters. Dampier in 1699, in chronicling the incidents of the voyage with which he was connected, made it an island.

In 1701 one would have supposed the question of the insularity of California would have been helped at least by the explorations overland of Father Kino the Jesuit which were begun in 1698. His map, based rather upon shrewd conjecture than upon geographical discovery, and showing the peninsular form of the land, was published in the Lettres Edifiantes, vol. v., in 1705. In 1705 the map in Harris’s Collection of Voyages preserves the insular character of California. In 1715 Delisle expressed himself as undecided between the two theories respecting California, but in 1717 he gave the weight of his great name to an imagined but indefinite great gulf north of the California peninsula, which held for a while a place in the geography of his time as the “Mer de l’ouest.” Homann, of Nuremberg, in 1719 marked the entrance of it, while he kept to the insular character of the land to the south; as did Seutter in his Atlas Geographicus published at Augsburg in 1720. Daniel Coxe in his Carolana had a sufficient stock of credulity—if he was not a “liar,” as Bancroft calls him—in working up some wondrous stories of interior lakes emptying into the South Sea. In 1727 the English cartographer Moll converted the same inlet into the inevitable Straits of Anian. The maps in such popular books as Shelvocke’s Voyages (1726) and Anson’s Voyages (1748), as did sundry maps issued by Vander Aa of Amsterdam, still told the mass of readers of the island of California; as had Bruzen la Martinière in his Introduction à l’histoire (1735), and Salmon (using Moll’s map of 1736) in his History of America.

Meanwhile, without knowing it because of the fogs, Behring, in 1728, had pushed through the straits now known by his name into the Arctic Seas, and had returned along the Asiatic shore in continued ignorance of his accomplishment. It was not till 1732 that another Russian expedition was driven over to the Alaskan shore; and in 1738 and 1741 Behring proved the close proximity of the two continents, and made demonstration of their severance.

At this time also the English were making renewed efforts from the side of Hudson’s Bay to reach the Pacific; and Arthur Dobbs, in his Countries adjoining to Hudson’s Bay (1744), gives a variety of reasons for supposing a passage in that direction, showing possible solutions of the problem in an accompanying map.

The Spaniards, who were before long to be spurred on to other efforts by the reports of Russian expeditions, were reviving now, through the 1728 edition of Herrera, more confidence in the peninsular character of California; though Mota Padilla in his Nueva Galicia, in 1742, still thought it an island.

The French map-maker Bellin, in his cartographical illustrations for Charlevoix in 1743, also fell into the new belief; as did Consag the Jesuit, in a map which he made in 1746.

The leading English geographer Bowen in 1747 was advocating the same view, and defining the more northerly parts as “undiscovered.” In 1748 Henry Ellis published his Voyage to Hudson’s Bay,—made in 1746-1747, and mentions a story that a high or low tide made California an island or a peninsula, and was inclined to believe in a practicable northwest passage. In 1750 Robert de Vaugondy, while preserving the peninsula, made a westerly entrance to the north of it, which he marks as the discovery of Martin d’Aguilar. The lingering suspicion of the northerly connection of the California Gulf with the ocean had now nearly vanished; and the peninsula which had been an island under Cortés, then for near a century connected with the main, and then again for more than a century in many minds an island again, was at last defined in its proper geographical relations.

The coast line long remained, however, shadowy in the higher latitudes. Buriel, in his editorial notes to Venegas’s California, in 1757, confessed that nothing was known. The French geographers, the younger Delisle and Buache, published at this time various solutions of the problem of straits and interior seas, associated with the claims of Maldonado, De Fuca, and De Fonte; and others were found to adopt, while others rejected, some of their very fanciful reconciling of conflicting and visionary evidences, in which the “Mer de l’ouest” holds a conspicuous position. The English map-maker Jefferys at the same epoch (1753) was far less complex in his supposition, and confined himself to a single “river which connects with Lake Winnepeg.” A map of 1760, “par les Srs Sanson, rectifiée par Sr Robert,” also indicates a like westerly entrance; and Jefferys again in 1762, while he grows a little more determinate in coast lines, more explicitly fixes the passage as one that Juan de Fuca had entered in 1592. The Atlas Moderne, which was published at Paris, also in 1762, in more than one map, the work of Janvier, still clung to the varieties presented by Delisle ten years before, and which Delisle himself the next year (1763) again brought forward. In 1768 Jefferys made a map to illustrate the De Fonte narrative; but after 1775 he made several studies of the coast, and among other services reproduced the map which the Russian Academy had published, and which was a somewhat cautious draught of bits of the coast line here and there, indicating different landfalls, with a dotted connection between them. One of Jefferys’s own maps (1775) carries the coast north with indications of entrances, but without attempting to connect them with any interior water-sheds. Going north from New Albion we then find on his map the passage of D’Aguilar in 1603; then that of De Fuca, “where in 1592 he pretends he went through to the North Sea;” then the “Fousang” coast, visited by the Spaniards in 1774; then Delisle’s landfall in 1741; Behring’s the same year; while the coast stops at Mount St. Elias. In his 1776 map Jefferys gives another scheme. “Alaschka” is now an island athwart the water, dividing America from Asia, with Behring’s Straits at its western end; while the American main is made up of what was seen by Spangenberg in 1728, with a general northeasterly trend higher up, laid down according to the Japanese reports. The Spaniards were also at this time pushing up among the islands beyond the Oregon coast. In 1774 Don Juan Perez went to Nootka Sound, as is supposed, and called it San Lorenzo. In 1775 another Spanish expedition discovered the Columbia River. Janvier in 1782 published a map still perpetuating the great sea of the west, which Buache and others had delineated thirty years before. The English in 1776 transferred their endeavors from Hudson’s Bay to the Pacific coast, and Captain James Cook was despatched to strike the coast in the latitude of Drake’s New Albion, and proceed north in search of a passage eastward. Carver the traveller had already, in 1766-1768, got certain notions of the coast from Indian stories, as he heard them in the interior, and embodied them with current beliefs in a map of his own, which made a part of his Travels through the interior parts of North America, published in 1778. In this he fixed the name of Oregon for the supposed great river of the west, which remained in the end attached to the region which it was believed to water. In 1786 the Frenchman La Pérouse was on the coast. In 1789 the English and Spanish meeting on the coast, the English commander was seized. This action led to a diplomatic fence, the result of which was the surrender of Nootka to the English.

Meanwhile a Boston ship, the “Columbia,” commanded by Captain Kendrick, in company with the “Washington” (Captain Gray), was on a voyage, which was the first American attempt to sail around the globe. They entered and named the Columbia River; and meeting Vancouver, the intelligence was communicated to him. When the English commander occupied Nootka, the last vestige of uncertainty regarding the salient features of the coast may be said to have disappeared under his surveys. Before they were published, George Foster issued in 1791 his map of the northwest coast, in which the Straits of Juan de Fuca were placed below 40°, by which Captain Gray is supposed to have entered, on his way to an open sea, coming out again in 55°, through what we now know as the Dixon entrance, to the north of Queen Charlotte’s Island; the American navigator having threaded, as was supposed, a great northern archipelago. Vancouver’s own map finally cleared the remaining confusion, and the migratory Straits of Juan de Fuca were at last fixed as the channel south of Vancouver’s Island which led to Puget Sound.