‘Someone' explored and mapped South America before the arrival of Christopher Columbus


Paul Gallez, the famous historian born in Brussels in 1920 and now resident in Bahía Blanca (Argentina) has the honour of having discovered the river system of South America, from the Orinoco in Venezuela to the river Grande in Tierra del Fuego, on the map of the German Heinrich Hammer under his Latinised name Henricus Martellus Germanus. This map dates from 1489, several years before Christopher Columbus's first voyages, and is kept in the British Library in London (68 verso and 69 recto of the Add Ms 15760). There are many maps of the world dating from the sixteenth Century, known as Ptolemaic maps (copies of the original map drafted by Claudius Ptolemy of Alexandria) which depict South America as an immense peninsula joined in the north-west by the Mesoamerican isthmus to the Asian continent and to be more exact to China. In this way, the vast continent of South America forms the so-called ‘Dragon's Tail'. It is under this name that it is mentioned in Antonio Galvano's The Discoveries of the World from Their First Original unto the Year of Our Lord 1555  published in Lisbon in 1563. In order to situate ourselves historically, in time and place, it should be pointed out that before becoming regent of Portugal, the infante Dom Pedro, Duke of Coimbra, son of John I (John the Great, Master of Avis who reigned from 1385 to 1433) and brother of Henry the Navigator, made a long journey through England, France, Germany, the Holy Land, Hungary and Italy where he visited Rome and Venice. According to Galvano's Discoveries, in 1428, Dom Pedro came back from Italy with a map of the world which showed ‘the whole surface of the Earth' and where ‘the Strait of Magellan was in the Dragon's Tail' (Paul Gallez, La Cola del Dragón – América del Sur en los mapas antiguos, medievales y rencentistas, Bahía Blanca, Instituto Patagónico, 1990).

     Following a detailed study of this 1489 map, some astonishing conclusions can be reached: the Atlantic coasts of America were well known to European sailors of the end of the fifteenth century. As we shall also see later on, they were also familiar with the western coasts of South America and knew of the existence of the Pacific Ocean.

     We know little of the life of the German mapmaker Martellus, apart from the fact that he was born in 1440, lived in Italy and spent some time working in the Vatican. The presence of America on the 1489 pre-Columbian map greatly modifies many previous studies. The German mapmaker drew a further map of the world in 1490, which was purchased by Yale University Library in 1950. The 1489 map was acquired by the British Museum in 1821 and strangely enough is technically far superior to the Yale map, especially as the former shows the river system of South America.

     Roberto Almagiá, the eminent Jewish-Italian cartologist (a person who studies the history of maps), (see I Mapamundi di Enrico Martello e alcuni concetti geografici di Cristoforo Colombo La Bibliofilia, 42 (l940) Firenze pp. 288-311, and Monumenta cartographica vaticana, Vol. I, Vatican Library, 1994) drew our attention to the strange and astonishing shape of what he called the ‘fourth Asiatic peninsula' and added it to those already known, the Arabic, Indian and Malayan peninsulas. Almagiá mentions the letter that Columbus wrote to Ferdinand and Isabella the Catholic on July 7 1503, in which he informed them of his proximity to the mouth of the Ganges. In actual fact, this was not the case, but it is a clear indication that Columbus was fully conversant with the Martellus map.

     In 1942, the Argentine historian Enrique de Gandía suggested in his work Primitivos navegantes vascos that the fourth Asian peninsula might be South America. In 1966, the Boletín de la Real Sociedad Geográfica de Madrid (Jan-Dec) published an in-depth study of Martellus's map, although its conclusions did not go so far as to question the part played by Christopher Columbus nor the fact that the ‘fourth Asian peninsula' had been identified as South America.

    Dick Edgar Ibarra Grasso, the eminent Argentine anthropologist and historian, made considerable progress in the interpretation of Martellus's map, when in papers published in 1980 and 1986 he associated Ptolemy's ‘Sinus Magnus' with the ‘Southern Ocean' (during his fourth voyage Christopher Columbus was anxiously searching for a route or a channel leading to the Southern Ocean [Nito Verdera, Cristóbal Colón, originario de Ibiza y criptojudío, Ibiza, Consell Insular 1999, pages 59-66]) or the Greek ‘Megas Kolpos' , the Latin ‘Mare Magnum' referred to by Spanish sailors from 1513 onwards; that is the sea known to us in modern times as the Pacific Ocean (Gustavo Vargas Martínez, América en un mapa de 1489, Mexico 1996).

     Paul Gallez's great discovery, the river system of South America, took place in 1973 in Bahía Blanca: ‘I got up at three o'clock in the morning and went to look at the copy of Martellus's map which I had photocopied in London. One by one I started to identify the rivers: the Magdalena, the Amazon, the Orinoco, the Paraná, the Paraguay – surprising in its precision, all in astonishing precision. Then the Colorado, the Negro, the vast American peninsula, the Valdés peninsula, the river Chubut …all appeared. They were all as they should be, neither one too few nor one too many. Even the river Grande was shown in Tierra del Fuego. One even more amazing detail was the inclusion of the river Chubut, which was practically unknown until 1830. I realised that this was an incredible discovery and so I decided to publish it in a magazine of prestige and renown, so that the whole idea would not be dismissed as mere fantasy. I wrote the article and, during a visit to Bonn, I handed it over to the famous Erdkunde magazine' (interview given by Paul Gallez to Rubén Benítez, and published in La Nueva Provincia newspaper in Bahía Blanca, Argentina on May 25th 2003).

Gallez's article, fruit of a night of insomnia, was extremely well received in Germany and was then published in French under the title of Les grands fleuves d'Amérique du Sud sur le ptolémée londonien d'Henri Hammer (1489) (Erdkunde XXIX/4, Bonn 1975). 

Corredera de barqueta i rellotge d'arena / Corredera de barquilla y reloj de arena / Log and sand glass (Cortesía: Fundació Jaume I, Nadal, 1991)

The river system of South America on the Martellus map (1489) 

Paul Gallez states that a simple comparison with a modern-day map will allow an analytical study of the great South American rivers on the London Ptolemaic map (see map 1). Let us see if this is indeed the case:

Martellus: In the north of the Dragon's Tail there is an important river which crosses practically the whole continent from west to east and then flows into the Eastern Ocean. Its source is in a mountain range running parallel to the coast of the Sinus Magnus (Pacific), not in the range nearest to this sea, but in a parallel range rising slightly to the east.

Modern map: In the northern part of South America, the Orinoco River crosses practically the whole continent from west to east and then flows into the Atlantic Ocean. Its main tributary, the river Meta, rises in a sierra running parallel to the Pacific coast, though not in the Cordillera Occidental, nearest to this ocean, but in the Cordillera Oriental lying slightly further east.


 Martellus: To the south of the lower and middle stretches of the Orinoco, there is a mountain mass extending practically to the ocean shore.

Modern map: To the south of the lower and middle stretches of the Orinoco, the Guiana Highlands almost reaches the Atlantic. 


Martellus: To the south of the Guiana Highlands there is a river running east across the peninsula, the longest in the Dragon's Tail and the only one in the peninsula forming large lakes and swamps.

Modern map: To the south of the Guiana Highlands, the Amazon River flows eastwards. It is the longest river in South America and is the only one in the subcontinent, owing to its width, to give the appearance of a chain of lakes and swamps. 


Martellus: There is another river, separated from the Amazon by a mountain range, which flows into the Atlantic and whose sources are in a sierra lying in the middle of the Dragon's Tail.

Modern map: The Tocantins River is separated from the Amazon by the Alto Pará range. It flows into the Atlantic and has its sources in the central Brazilian highlands. 


Martellus: Another river rises in the same mountain range and flows into the ocean to the south of the easternmost cape of the Dragon's Tail.

Modern map: The San Francisco River rises in the Plateau and flows into the Atlantic to the south of the cape of Sao Roque, the easternmost cape in South America. 


Martellus: There is a long stretch of coast lying to the south of the San Francisco River where there are no rivers. There is a prominent sierra running parallel to the ocean.

Modern map: There is a long stretch of coast lying to the south of the San Francisco River where there are no rivers of any importance. The Serra do Mar runs parallel to the ocean for two thousand kilometres.  


Martellus: The river rising in this range runs first west and then southwest in the inland area of the Dragon's Tail. It then meets another river whose source is in a sierra situated further to the northwest. Together they form a mighty river that runs south and then southeast before flowing into the eastern ocean through a wide estuary.

Modern map: The Paraná river, whose main tributaries rise in the Serra do Mar, flows first west and then southwest, through inland Brazil. It then meets the Paraguay, which flows from the Mato Grosso Plateau. Together they form the Lower Paraná, which runs south and then southeast and flows into the Atlantic through the River Plate Estuary. 


Martellus: Further south, two rivers, which rise in the same sierra, run parallel to each other east-southeast towards the eastern ocean.

Modern map: Further south, the Colorado and Negro rivers, which rise in the same Mendoza-Neuquén sierra, run east-southeast to the Atlantic parallel to each other. 


Martellus: To the south of these rivers, a large peninsula juts out far into the sea. It is the only peninsula on the whole of the eastern coast of the Dragon's Tail.

Modern map: To the south of the Negro river, the Valdés peninsula juts out more than a hundred kilometres into the Atlantic. It is the only peninsula of any importance on the whole of the eastern coast of South America. 


Martellus: To the south of the peninsula there is a river rising in the western mountain range and flowing eastwards.

Modern map: To the south of the Valdés peninsula, the Chubut River rises in the Andes and runs east. 


Martellus: There is only one river in Tierra del Fuego that flows into the Atlantic.

Modern map: In Tierra del Fuego, only the Grande River flows into the Atlantic. 


     The identification of all the great South American rivers on Henricus Martellus's 1489 pre- Columbian map based simply on their geographical position, their main characteristics, and the direction in which they flow, provides indisputable proof that the Dragon's Tail can be identified as South America on this map and therefore on the maps of all the cartographers who have copied it, as well as those who, as in the case of Waldseemüller, were unaware of what they were actually copying. In particular, the Paraná-Paraguay system, with its Y shape opening to the north-northeast, its lower stretches which turn from the south to the southeast, and its broad estuary where it flows into the ocean, forms a figure which is absolutely unique. Martellus's map depicts it with complete accuracy, both in form and course, with the addition of the Serra do Mar, a mountain formation also quite unique in the world in its positioning parallel to the coast. And as if that were not enough, the dimensions of the system relative to the other major rivers are also absolutely correct together with its position in relation to these rivers and the coast. Indeed, the Paraná-Paraguay river system in itself is sufficient to reveal the South American identity of the Dragon's Tail.

     To believe that the correct depiction of all the major South American rivers is fruit of a ‘mere coincidence' in the fantasy of a mapmaker drawing an imaginary peninsula is, at the very least, to show ignorance of the odds. Dr. Gallez says: ‘the success of this research has encouraged us to attempt to identify other geographical elements in this part of Martellus's map. To this end, he has resorted to a little known cartographic technique, ‘the distortion grid method'. 

Astrolabi / Astrolabio / Astrolab (Cortesía: Fundació Jaume I, Nadal, 1991)


      Austrian and Swiss geographers have used the distortion grid with considerable success in order to determine the degree of error in positions on maps dating back to the Middle Ages and the Renaissance.

     The majority of such maps bear no indication of latitude or longitude, except in a few cases, the equator and the tropics. The distortion method consists of working out the position of the meridians and parallels based on the location of mountains, rivers and coasts, which are generally easy to identify.

     So the cartographic analysis and the building up of a network of meridians and parallels must depend on our ability to identify, with certainty, a sufficient number of places. We have just identified, as Paul Gallez reminds us, all the major rivers in the Atlantic basin as well as several mountain ranges. On the other hand, no attempt has been made to say anything about the Pacific basin owing to its limited river system, both on Martellus's map and as it is in reality. However, this problem has been solved. The eastern river system in the Dragon's Tail, which has been successfully identified through its geographical location, orientation and length of rivers, served as a base for building up the grid of meridians and parallels. In turn, this grid of meridians and parallels will enable us to identify rivers and lakes in the Pacific basin, as well as several capes on the Atlantic coast previously unidentifiable. 


     To be able to build the distortion grid, Gallez considered that the drawing of meridians and parallels on the London Ptolemaic map should be carried out using as a base only the geographical features shown on the map and identified with certainty, ignoring the lines of longitude marked on the Yale map, which represent an accumulation of errors made by Marinus of Tyre, Ptolemy and possibly even Henricus Martellus himself and his ‘mysterious informers'. In short, the distortion grid has been drawn based on geographical features we have successfully identified, both for lines of latitude and longitude. 


     The elements at our disposal are, in order of importance, the rivers, the coasts and the mountains. Paul Gallez took the London Martellus map and started out by drawing in the parallels at 10º intervals, using a modern map of South America as a guide. The 10º N parallel of latitude runs parallel to the Meta and Orinoco rivers and falls outside the picture. The equator enters the continent through the Amazon estuary, continues to the north of this river and cuts across the southern foothills of the Guiana Highlands. It crosses a swampy region which might be the river Negro and then cuts through the Andes to the south of the source of the Magdalena river, before reaching the Pacific Ocean. In the grid we can see that the equator forms a straight line between the Atlantic and the river Negro swamps where it then dips down before resuming its normal direction. This is only a very slight distortion.

     The 10º S parallel enters the continent to the south of the Cabo Sâo Roque, near the mouth of the San Francisco River. It passes north, then south and then again north of this river, cuts across the Tocantins, runs parallel to the Amazon and reaches the Pacific.

      The 20º S parallel cuts through the coastal mountain chain, then the Upper-Paraná and the Upper-Paraguay rivers. It crosses the Peruvian highlands and meets the Pacific to the south of Lake Titicaca.

      The 30º S parallel crosses the Brazilian coast to the south of the Serra do Mar, then the Paraná between Corrientes and Santa Fé. From there, through the lack of any points of reference on Martellus's map, we run it parallel to the 20º S parallel.

     The 40º S parallel crosses the coast between the mouths of the Colorado and Negro rivers. It then crosses the river Negro and continues across to the Pacific.

     The 50º S parallel meets the continent to the north of the Strait of Magellan, represented here by the Patagonian isthmus. The lack of details on the map precludes greater precision.  

The drawing of the meridians is carried out in the same way as that of the parallels.

     The meridian 40º W of Greenwich meets the continent between the mouth of the river Tocantins and Cabo Sâo Roque. It leaves slightly to the north of Cabo Frio, after crossing the San Francisco River.

     The 50º W meridian meets the continent at the mouth of the river Amazon, below the equator, runs close to the sources of the Tocantins, and to the west of the source of the Sâo Francisco. It then cuts across the Upper Paraná and the southern part of the Serra do Mar, reaching the coast 30º S.

      The 60º W meridian cuts into the continent very close to the Orinoco delta. It crosses the Manaus swamplands, runs to the west of the Mato Grosso Plateau, reaches the Paraná, which it crosses at about 31º S and once again at 33º S, and then leaves the continent at 39º S.

     The 70º W meridian crosses the Meta and the upper stretches of the Amazon, leaving the Colombia range to the west. It cuts across Lake Titicaca and runs near to the Pacific coast at 20º S, staying in the mountain regions where it crosses the upper stretches of the Colorado and the Negro tributaries such as the Upper Chubut. It then crosses the Patagonian Isthmus (the Strait of Magellan) and reaches the ocean to the south of Tierra del Fuego.

     The 80º W meridian crosses the coast at the equator and then again to the south of the Promotorium Satyrrorum, which Dick Edgar Ibarra Grasso (La Representación de América en mapas romanos de tiempos de Cristo, Buenos Aires 1970) has identified as Punta Aguja. 


     In Dr Gallez's opinion, the cartographer has made only two mistakes of importance, the location of the Brazilian Highlands too far to the west, and the mountains spawning the Patagonian rivers too far to the east. In both cases the true geographical facts were unknown for many centuries after the drawing of Martellus's Ptolemaic maps, the Brazilian Highlands until the seventeenth century, and the upper stretches of the Patagonian rivers until the mid nineteenth century.

    As a final result of this analysis,' says Gallez, ‘we can see that Henricus Martellus's pre-Columbian Ptolemaic maps are drawn up with a precision and accuracy worthy of admiration, which cannot be explained in terms of the history of cartography as we know it today.'

Finally, the existence of Martellus's 1489 map allows us to assert that Christopher Columbus would have been in possession of excellent maps of South America before embarking on his voyages of discovery.



   Waldseemüller offers a configuration of recently discovered land areas on his 1507 map of the world and Carlos Sanz (Los antiguos mapas del mundo [15th-16th Centuries], Madrid, 1961, p 90) says that ‘the Western Hemisphere appears in the middle of an ocean, well separated from Asia, and Waldsseemüller's extremely original projection offers a most accurate and true picture, particularly so in the case of the small hemispheres decorating the upper part of the map'.

    Carlos Sanz mentions that in Europe in 1507 the existence of the Pacific Ocean was as yet unknown. However this is not true. The proof of this is that, during his fourth and final voyage, Christopher Columbus searched long and hard for a passage to the Southern Sea around Central America. For further information on the map mentioned above, you can visit the first part of this website.

   On Waldseemüller's map, on the cartouche or hemisphere drawn next to Amerigo Vespucci, we can distinctly make out the bend formed by the Pacific coast of South America between Peru and Chile, around the Roadstead of Arica (18º 29 S). This bend cannot possibly have been drawn purely by chance, so Waldseemüller must have been in possession of information not generally known in Europe; some maps which had to have been drawn by sailors who had explored the Pacific coast of South America prior to the arrival of the Spaniards. It should be remembered that Vasco Núñez be Balboa only set foot on the Pacific coast in 1513.

   It is indisputable that the bend of the coast drawn by Waldseemüller in 1507 is amazingly similar to the shape of the coast of Arica (Chile), as we can see on the modern-day Mercator projection. In addition, we have already seen how Henricus Martellus's map shows Lake Titicaca and the present-day Punta Coles and so there is also a bend on the coast, though not quite so pronounced, as he had used a different projection.

 The Walsperger map (1448)    

     This is a map of the world drawn in Konstanz in 1448 by Andrew Walsperger, a Benedictine friar from Salzburg. The south is at the top and the east is on the left. The great peninsula on the left is South America. The large castle depicted to the north of South America represents the ‘Earthly Paradise'. It is kept in the Vatican Library (Ms. Palat. Lat. 1362).  In this map we can observe the Red sea and Arabia, the Persian Gulf and India, the Gulf of Bengal and the Malayan Peninsula, under the name of the Golden Chersonese, the name given to it on the maps of Marinus of Tyre and Ptolemy.

     The great gulf which then appears is the Pacific Ocean, reduced to about a twentieth of its actual size, since it is copied from Ptolemy's map. To the east of the Pacific, as in all the sixteenth century maps and also Martellus, we then find the vast continent of southeast Asia, the Dragon's Tail, that is, South America. 

Brúixola / Brújula / Compass (Cortesía: Fundació Jaume I, Nadal, 1991)


     We cannot but be impressed by Paul Gallez's powers of observation, fruit of extensive, objective analysis, and we shall see yet another example of this when he demonstrates quite decisively that Christopher Columbus was in possession of a copy of Walsperger's map. It is of common knowledge that, during his third voyage in 1498, Columbus reached the Gulf of Paria between Trinidad and Venezuela, which he christened ‘Tierra de Gracia', and saw the western branch of the Orinoco. Afterwards, Columbus wrote to the King and Queen of Spain and told them that he had discovered ‘a river flowing out of Paradise'. Many historians, including some of the best, have simply come to the conclusion that ‘his mind is rambling and he thinks he has discovered the Earthly Paradise' (Antonio Rumeu de Armas; La epopeya colombina, p 111, José Manuel Gómez Tabanera Las raíces de América, Madrid, Istmo, 1968). I received an e-mail from Paul Gallez on April 17 2003 which read: ‘Rumeu de Armas and Pérez de Tudela have, like many other academics, reached an age at which they are unwilling to admit new ideas. Science must remain as they have known it; that is to say, that they are opposed to progress.'

     However, Paul Gallez claims that his interpretation of the map shows that the castle depicted on Walsperger's map covers the mouth of the Orinoco, and that on pointing out its closeness to Paradise, Columbus has merely read his maps correctly. 


      When he explains the events that took place during the third voyage of discovery, Fray Bartolomé de las Casas explains that the Admiral ‘came to the conclusion that the world was not round, contrary to the opinion of astrologers and philosophers, but that, while the hemisphere of Ptolemy and the rest was indeed round, this other one here, of which they knew little, was not completely so. He imagined it as half a pear with a stalk, or like a woman's breast on a round ball, and that the part with the nipple was higher and closer to the air and heaven (…); and it seemed to him that the Earthly Paradise might be found on this nipple'.

     Referring to the shape of the earth, Columbus states textually: ‘I had always read that the world, land and sea, was spherical, and the authority and experience of Ptolemy and all the rest who have written about this place supported and demonstrated this idea, together with eclipses of the moon and other illustrations they make from East to West, such as the elevation of the North Pole in the southern hemisphere; now I have seen so much deformity that I started to think about the world and found that it was not round as they write, but that it is shaped like a pear which is round except where the stalk is, which there is higher…”

     Las Casas remarks: ‘And so it appears that, for these reasons, the Admiral did not reason well that the earth was round…' (Historia de las Indias, Vol I, Madrid, Biblioteca de Autores Españoles, 1957 pp 375-376). 

In order to find out what degree of truth there was in Columbus's claims about the shape of the Earth, on November 16 1981 I travelled to Madrid to visit the Military Geographical Service and the School of Geodesics and Topography. I was there able to talk to Colonel Angel Paladín Cuadrado, an expert in geodesics, and I showed him what Columbus had written. To my great surprise, he told me that he was already familiar with the passage and that Columbus was perfectly correct. Indeed, in ‘Geodesics in the Space Age', p. 73, a speech given by José María Torroja Menéndez on taking his seat in the Real Academia de Ciencias Exactas, Físicas y Naturales on June 25 1969 we can read: ‘The first results in the study of the shape of the Earth were obtained in 1958 from observations carried out by Vanguard I satellite which gave a flattening value of alpha = 1/2988.3 instead of the previously accepted value of 1/297. These observations also allowed O'Keefe to show that the two hemispheres are not symmetrical, thus confirming what Columbus had guessed, that the Earth must be shaped like a pear, in the account of his third voyage (…). These conclusions were later confirmed by the Vanguard II and III satellites. The North Pole stands approximately fifty metres above and the South Pole fifty metres below the average sea level. Outside the polar circles, the average level is about twenty-five metres lower in the northern hemisphere and higher in the southern hemisphere. In addition, the equator turned out not to be round. The greatest diameter, that crossing Brazil, is about four hundred metres longer than that lying at right angles to it'. (Nito Verdera, La verdad de un nacimientoColón ibicenco, pp.107-111, Madrid, Kaydeda, 1988).

     So despite Las Casas, who claimed that ‘the Admiral did not reason well' it turns out that Christopher Columbus declared that the Earth was pear-shaped 470 years before this could be confirmed by artificial satellites. Columbus was apparently in possession of other amazing scientific information, as well as Walsperper and Martellus's maps. I can well understand why Las Casas might think that Columbus was mistaken or was drawing wrong conclusions as to the shape of the Earth, since the Dominican was lacking in the scientific knowledge that would entitle him to give an opinion of any weight. However, when the scholar, Juan Manzano Manzano (Colón y su secreto – El predescubrimiento, p. 250 and 251, Madrid, Ediciones Cultura Hispánica 1982) refers to Columbus's third voyage and the problem with the spherical nature of the Earth he says ‘Columbus (he uses the word Ligurian) accepted without question the common belief that the Earth was round until his third voyage, during which he discovered (poor man!) that the eastern hemisphere below the equator had a protuberance or a pinnacle similar in shape to a woman's breast, where he believed that the Earthly Paradise could be found'. However, it turns out that Columbus was far from being a ‘poor man'. The eminent Spanish historian would have done well to consult the Real Academia de Ciencia Exactas, Físicas y Naturales or the Madrid School of Geodesics on Columbus's claims about a pear-shaped Earth.

     However, we should not be too surprised: ‘Young rebels are open to change, while the old are not prepared to question ‘scientific truths' in which they have believed all their lives' (Paul Gallez, Predescubrimientos de América, p. 49, Bahía Blanca, Argentina, Instituto Patagónico, 2001).