The Piri Reis Map

Steven Dutch, Natural and Applied Sciences, University of Wisconsin - Green Bay
First-time Visitors: Please visit Site Map and Disclaimer. Use "Back" to return here.

The Map

The Piri Reis Map, shown below, is the oldest surviving map to show the Americas. It is not European, surprisingly, but Turkish. It bears a date of 919 in the Moslem calendar, corresponding to 1513 in the Western Calendar. It is in the Topkapi Palace in Istanbul, a fabulous museum and the locale for a truly awful movie in the late 1960's. (I've been there - the real place bears no resemblance to the place in the movie.) The map was lost for a long time and only rediscovered in the 20th century.

Apart from its great historic interest, the map has been alleged to contain details no European could have known in the 1500's, and therefore proves the existence of ancient technological civilizations, visits by extraterrestrials, or both.

The map is a portolan chart, a common form at this time. Instead of latitude and longitude grids, compass roses were placed at key points with azimuths radiating from them. That said, the east-west lines through the small rose off South America in the center of the map are a very good approximation to the Equator, both there and with respect to Africa. The small one at the very top of the map is a very good estimate of 45 north where the east-west azimuth hits the coast of France. The two big compass roses in mid-Atlantic are harder to place. They might locate the tropic lines (23-1/2 north and south) or they could represent 22-1/2 latitude (one-fourth of the way from equator to pole). Considering they are a bit closer to 45 degrees than the equator, the tropic lines are the best bet.

Erich von Daniken in Chariots of the Gods? claimed that the map closely resembled an azimuthal projection centered on Cairo.

At left is a real azimuthal projection centered on Cairo. This projection does tilt the Greater Antilles vertical and bring them up even with northern Europe. But it fails to bring South America below the bulge of Africa. And the equator, which is quite precise on the Piri Reis map, is curved.

The straight parallels of latitude show that the map cannot be azimuthal. It has to be a cylindrical projection, probably cylindrical equidistant if anything. A cylindrical equidistant projection has equally spaced parallels of latitude. It was rarely used in pre-computer times (there are better projections that are just as easy to construct) but has become a lot more common recently because it is the easiest projection to plot on a computer. (You just plot latitude and longitude directly without any mathematical alterations.)

At left is a direct comparison between the Piri Reis Map and the supposedly identical azimuthal equidistant projection. The scale is chosen to find the best fit with the western bulge of Africa.  Nothing matches. Spain on the azimuthal equidistant map is well to the right of western Africa, not directly above.

So clearly the claim that the Piri Reis Map matches a map centered on the Middle East is total garbage. At left is an azimuthal equidistant map centered on 0, 0. The fit of Africa and Spain is far better and the fit with Brazil is surprisingly good. Features on the South American coast down to southern Brazil can be identified with certainty. Beyond that, though, the map is fantasy. It doesn't match either South America or Antarctica very well.

So, apart from claiming vague similarities between the Piri Reis map and Antarctica, what positive, specific evidence do you have that the map shows Antarctica?

The Marginal Notes

The marginal writings on the map are very revealing. Translations are in The Oldest Map of America, by Professor Dr. Afet Inan. Ankara, 1954, pp. 28-34 and available at a number of Web sites. Until 1928 the Turkish language was written with Arabic letters, but the language on the map is Turkish, not Arabic.

Most important is that references to maps of Asia, plus some fragmentary lines south of Africa, indicate that this was originally a world map which was torn in half along the eastern edge. Wouldn't it be marvelous to see the other half?

Most of the bizarre claims made for the Piri Reis Map utterly ignore the marginal notes, which pretty conclusively show the map is entirely 16th century terrestrial in origin.

Cartography of the Piri Reis Map

Below is a tracing of the coastlines on the map. Western Europe and Africa are easily recognizable, the Azores, Canary Islands and Cape Verde Islands are fairly accurate both as to location and the number and arrangements of individual islands. Eastern South America is also easily recognizable, but there are a lot of things not so easily recognized. The map, by the way, is very clear on the existence of mountains in the interior of South America (in brown on the tracing).


The coastline of France and Iberia is well-drawn. There are four major rivers shown in Iberia, from north to south the Atlantic rivers are the Tagus and Guadalquivir, and the east-flowing rivers are the Ebro (north) and an unknown river in the south (there are several minor rivers it could be).

The rivers are very inaccurately located. The Tagus enters the Atlantic at Lisbon as shown, but does not have a hook in its upper reaches. The Duoro, to the north, does, but it's not shown. It looks very much as if the draftsman confused the two rivers.

By the way, the Spanish syllable guad- that begins so many place names comes from Arabic wadi, valley. Wadi-al-yahara, valley of the flowing water, became Guadalajara, for example.


The western bulge of Africa is pretty well drawn and the offshore islands are as well (though too large relative to everything else).

There are a couple of small rivers in Morocco that could correspond to the northernmost river. The river emptying at the center of the bulge is the Senegal and the next one south is the Gambia, followed to the south by the Guinea. The two rivers do not join but do approach closely. The south-flowing river is probably the Sassandra in the Ivory Coast.

The welter of lakes and rivers inland do not exist as shown but may reflect some garbled knowledge of the Niger headwaters and its inland delta.

Some people have claimed the map shows the Sahara as it was during the Pleistocene, when it had huge inland lakes. There are several reasons to doubt this:

North America

North America is frankly a mess on this map. The only voyages to North America by 1513 were voyages to Newfoundland beginning with John Cabot in 1498, and some Spanish sightings of the southeast coast of the U.S. It was only in 1513 that Balboa reached the Pacific and Ponce de Leon discovered people who can't punch ballots correctly in Miami Beach.

The marginal notes refer to some of the islands and coasts north of South America as "Antilia," clearly referring to the Antilles. The lack of good detail is puzzling since there must have been much better maps of the Caribbean by this time. If it's a real place at all - "Antilia" was a legendary island of the times. The big triangular island in the far northwest could be Newfoundland. It's close to the right latitude and even pretty much the right shape. Given that the most detailed knowledge of North America was in the north at this time, the big island off the coast is much more likely to be Nova Scotia than one of the Antilles. Supporting this is the fact that a nearby note refers to St. Brendan, an Irish monk who according to tradition sailed far into the North Atlantic in the sixth century. He might conceivably have reached Newfoundland or Nova Scotia but is pretty unlikely to have reached the Antilles.

The mess of North America is important. It's ridiculous to claim, as many people do, that there are ancient or extraterrestrial secrets lurking in this map when something as big as North America is so crudely drawn.

Robert Bywater and Jean-Pierre Lacroix published a very interesting hypothesis in Journal of Spatial Science vol 49 (1); 13-23 (2004) They suggest that the islands off North America might actually be Asia. The dream that the Americas might somehow be joined to Asia died hard, and remember, this map predates Magellan by a decade so nobody really knew how wide the Pacific was. As late as 1634, Jean Nicolet sailed into Green Bay expecting to meet the Chinese. It's worth considering.

Secrets in the map?

It's the other stuff that fascinates people. Among other claims:

Here's a map that does show the earth from space as seen from a point that roughly matches the Piri Reis Map (20N, 30W). We can see that any similarity between this map and the Piri Reis Map, apart from what terrestrial navigators knew in the early 1500's, is imaginary.

This projection is called an orthographic projection. Draftsmen of the 1500's would have been perfectly capable of drawing such a map given the geographic coordinates. You do not need to go into space to do it. For one thing, by this time there were globes to use as models.

At left is the same map with the Piri Reis map superimposed on it. The conclusions don't change: Europe and Africa, pretty good. South America, fair. In fact the crudeness of the cartography of the Caribbean coast is more obvious here. Similarity to North America: vague at best. Similarity to Antarctica: imaginary.

The fit is actually not as good as the fit with the azimuthal equidistant map shown above.

Below is the Piri Reis Map with modern maps superimposed. We can see that Europe and Africa are pretty good but with lots of inaccuracy in detail. Promontories and bays are exaggerated, a natural enough tendency in a day when navigating by landmark was a matter of life and death. The Azores, Canary Islands and Cape Verde Islands are accurately located but again, exaggerated in size. Also note a hint of cartographic breakdown where the coast of Africa meets the right edge of the map.

Brazil is pretty recognizable, but South America is too big compared to Africa and Europe, the Atlantic is way too narrow, and South America is compressed east-to-west. Also, what are the big islands offshore? North America is essentially imaginary.

Now one thing we can count on navigators of the 1500's being able to measure accurately was latitude. On the east side we can clearly see the tip of France, so the top of the map represents about 50 degrees north latitude. So right away we can forget about this map showing Greenland, subglacial or not. The coast of subglacial Greenland, by the way, won't look very different from the present coast, for the simple reason that most of the Greenland coast is rock, not ice. There's nothing on the map that even vaguely resembles Greenland. 

The Piri Reis Map does not use any systematic projection, although as noted above it's close to a cylindrical equidistant. It tries to get features accurate to shape and relative location, and it tries to plot accurate latitudes, but there is no reasonable transformation of the present earth that will yield the Piri Reis Map. (You can, of course, come up with a mathematical transformation that will transform any map into any other map, but any transformation of the real world into the Piri Reis Map would be so convoluted and ad hoc that it would prove nothing.)

South America

The scale of South America above was chosen to give a good fit in latitude from the north coast to the tip of Brazil, presumably the best-mapped part at the time the map was drawn. We can see that the match between the modern map and the Piri Reis Map is pretty good for some distance south of that, both in scale and in geographic detail.

That long stretch of coast on the bottom of the map has been claimed to be Antarctica, a place not known to humans (according to orthodox history) until the 19th century. So let's compare a modern map of South America (left, below) with the Piri Reis Map (right).

Start with the obvious. The tip of Brazil is easy to place (A-a). To the west (b) we have a large river flowing into a broad recess. This can only be the Amazon (B). The big island to the northeast on the Piri Reis Map may be Marajo Island, the big island at the mouth of the Amazon. Whatever, the fact that there is no island in mid-Atlantic as shown doesn't bode well for the idea that this map drew on ancient advanced knowledge.

To the south, the sharp recess in the coast of Brazil (C-c) is easy to see on both maps. At d we have a large river with a big delta flowing out of a convex coastline, and a big island offshore (e). It's a nearly perfect match for the Orinoco (D) and the island is Trinidad (E). One of the two rivers at g is almost certainly the Magdalena (G) but it's not clear what the other one is. Possibly the Magdalena is the river to the east and the Darien is the river to the west. The coastal bend north of Panama is fairly clear (F-f) but everything north of that bears almost no resemblance to any modern maps.

Moving south, it's tempting to identify the big river at h with the Rio de la Plata (P), except the Rio de la Plata is too far south and empties into a large bay, not on a bulge in the coast. The Piri Reis Map actually matches the real coastal bulge at H far better, except there's no river there. But there is a city called Rio de Janeiro, or "River of January" because the discoverer mistook the complex bays there for the mouth of a large river. In fact, the real coastline there looks rather like the Piri Reis coastline, if you squint a bit. It certainly looks more like it than anything on the map looks like Greenland! If we buy this, the smooth concave indentation to the south (I-i) falls into place.

The southern compass rose on the map would place the tropic of Capricorn on the small coastal bump halfway between c and h, and that would favor the big river being the Rio de la Plata. So we have to conclude that either the latitudes or the coastline (or both) are inaccurate south of c. The coastal fit seems too good to discard, and the marginal notes in this area explain how Piri Reis synthesized his map from a number of sources, so it's not hard to see how latitude might have suffered a bit in the process. Remember, he didn't have the raw latitude observations to go on.

Thereafter, the Piri Reis Map drifts into the Twilight Zone. It shows South America swinging far to the east. Given that the map so far has done fairly well in latitude, we can be sure the coastline is not Antarctica. Also, if the map draws on ancient knowledge to show things no 16th century explorer would have known, why is the coastline continuous? So why isn't there open water between South America and "Antarctica?" You can't seize on an accidental resemblance to a couple of bumps on the coast of Antarctica and blithely ignore the failure to show the Drake Passage!

Most damning of all to the Antarctica interpretation is that the marginal notes refer to the coast in this region being discovered by Portuguese ships blown off course. One note refers to the land being "very hot," which probably rules out Antarctica. The Piri Reis Map itself explicitly says the information in this area came from European sources. Atlanteans and extraterrestrials need not apply. We have isolated sightings of coast made by ships far off course and unsure of their location. Small wonder the map is wildly inaccurate.

Considering that we have had a good match so far by assuming the Piri Reis Map shows relative latitude accurately (although not nearly as well as north of the equator; the scale of South America is too large), and that coastal features like points and bays are accurately rendered, then south of the smoothly curving coast at I-i there must be a cusp on the coast (j-J). The next prominent point k could be the point beyond the Rio de la Plata (K). The latitude is about right compared to the rest of South America.

Above is an alternative interpretation of the mystery area. It requires us to assume the latitudes are badly off, something not hard to envision in maps of that era. However, it matches the curves in the coast. Point k might even correspond to the tip of Tierra del Fuego.


Above is a map of South America and Antarctica with the Piri Reis coastline in magenta. Southern South America and Antarctica are in the orthographic projection - in other words they do look like they would as seen from space. We can see the Piri Reis Map bears no resemblance at all to Antarctica. The 600-mile wide Drake Passage is not shown, nor are the large islands in the Weddell Sea. The latitude is thousands of miles off.

So in response to people who ask how to explain why the Piri Reis Map shows the coastline of Antarctica accurately, the answer is - it doesn't. It especially doesn't show the subglacial coastline of Antarctica, which corresponds to the existing coastline of Antarctica around most of the continent anyway.

Rule #1 For Interpreting Ancient Maps (If You Want A Best Seller)

Anything that matches (or can be made to seem like a match to) existing cartography is proof that the cartographer had access to secret knowledge. Anything that doesn't match, doesn't count.

Rule #1 For Interpreting Ancient Maps (If You Seriously Want to Learn Anything)

The map can be no better than its portrayal of the areas that were well explored in the time and place the map was drawn. If it has significant errors in known geography, claims that the map shows unknown geography are simply worthless.

Some Real Mysteries About the Map

The map seems to show more detail than Europeans were likely to have in 1513. Pizarro hadn't been to Peru, yet, so how did Piri Reis know about the Andes? Did somebody hear tales of mountains far inland? Also, the detail on the South American coast seems a bit rich for 1513. Was the map begun then and completed later? Was the map copied later and the date miscopied? But if the map was derived from ancient sources that contained details otherwise unknown to Europeans, why are so many parts of it so crude?

There's also a marginal note opposite South America that says "It is related by the Portuguese infidel that in this spot night and day are at their shortest of two hours, at their longest of twenty two hours. But the day is very warm and in the night there is much dew." That would indicate a far southern latitude, but note that the report explicitly comes from the Portuguese, not from arcane ancient sources. It's possible that some Portuguese expedition was blown very far south, not to Antarctica where the days are rarely "very warm," but perhaps to 50 south or so. 

Let's Hear it for Piri Reis

For 1513, this map shows an astonishing amount of detail. The notes on the map explain that the map was synthesized from about 20 maps, many of which were captured from Spanish and Portuguese ships in the Mediterranean. It was also supplemented by accounts given by captured Spanish and Portuguese sailors.

Not a map from some ancient Atlantean civilization, not a map created by extraterrestrials, but a first class piece of naval intelligence. Considering that it was created by a sailor whose country never participated in the age of exploration, and that it's drawn wholly from second-hand sources, it's an astonishing piece of work. It seems to contain up-to-the-minute details derived from enemy maps, many of which would have been tightly-guarded secrets.

There's a class of crank that hates the idea that other people might have real accomplishments, because they never accomplish anything themselves. So Shakespeare didn't write his plays, other people did; Robert Peary didn't reach the North Pole as he claimed, and so on. And Piri Reis wasn't a gifted admiral and good intelligence analyst, but had to get help from ancient lost documents. Get a life, folks.

Same Old, Same Old

Here's a recent e-mail I got. My comments are in red.

So let me get this straight, regarding the Piri Reis map,

You feel that people are totally wrong and probably liars, when they claim:

"In 1953 ... the U.S. Navy Hydrographic Bureau ... Arlington H. Mallery, an authority on ancient maps ... Mallery discovered the projection method used. ... the map was totally accurate. ... The Hydrographic Office ... were ... able to correct ... errors in the present days maps" <>

Yes, they are totally wrong. Certainly the people who created the web page he cites did absolutely no original work of their own but are merely parroting older pseudoscience works.

Reality check here. In 1953, we had just fought a major war in the Atlantic, where errors in maps could lead to ships being sunk and battles being lost. We're to believe a 16th century map was more accurate than charts used to fight naval warfare in World War II when ships and lives depended on cartographic accuracy? And what specific errors were found and corrected? Where's the documentation that any of this ever actually happened?

Searching for Arlington H Mallery on line is revealing. He was a prolific author of cult archeological theories, and at least a couple of times he got into professional journals, only to get smacked down soundly for his errors.

Regarding your own page

"Most of the bizarre claims made for the Piri Reis Map utterly ignore the marginal notes, which pretty conclusively show the map is entirely 16th century terrestrial in origin."

Surely, assuming the translations of his notes are correct:

"The-hand of this poor man has drawn it and now it is constructed. From about twenty charts and Mappae Mundi-these are charts drawn in the days of Alexander, Lord of the Two Horns, which show the inhabited quarter of the world; the Arabs name these charts Jaferiye-from eight Jaferiyes of that kind and one Arabic map of Hind, and from the maps just drawn by four Portuguese which show the countries of Hind, Sind and China geometrically drawn, and also from a map drawn by Colombo in the western region I have extracted it. By reducing all these maps to one scale this final form was arrived at. So that the present map is as correct and reliable for the Seven Seas as the map of these our countries is considered correct and reliable by seamen."

Which seems to conclusively prove that he may have drawn this in the 16 century, but a lot of it is based on much older information.

Duh. If you draw a map from existing sources, they are certainly older. Note the logical leap from "older" to "much older" or "ancient."

"One note refers to the land being "very hot," which probably rules out Antarctica."

Well, if the ice was gone when they were there, it would be hot, wouldn't it? He also writes "there are white-haired monsters ... , and also six-horned oxen." - only quote what fits, eh? ;)

Gee, come on up to Wisconsin in May. The ice is gone (usually) but tell me if it's "very hot." The quotes about white-haired monsters and six-horned oxen further illustrate the inaccuracy of the source.

"The Piri Reis Map itself explicitly says the information in this area came from European sources. Atlanteans and extraterrestrials need not apply. "

Actually, it says what Piri Reis thought - not who actually originally authored the maps (not that I'm trying to prove extraterrestrials, you just seem unnecessarily arrogant and condescending here.)

Oh, here we go again. Saying something based on technical information is "arrogant," arguing about it with no background whatsoever, and expecting to be taken seriously, isn't. More of the Self-Appointed Expert syndrome. Aww, pity-poo.

Return to Pseudoscience Index
Return to Professor Dutch's Home Page

Created 8 July 1998, Last Update 02 June, 2010

Not an official UW Green Bay site