Gazelle hide, scraped and dried and painted, about the size of a small tablecloth.
The coastlines run in gold and black and green. There are ships riding the open ocean. There is a man on an ox in the interior of a continent that almost nobody in the mapmaker's city had heard of.
Follow the western shore downward. Past the mouth of the Amazon. Past the bulge of Brazil. Further south than any European fleet had yet sailed. The coast keeps going.
Then it turns.
It swings east and runs along the bottom edge of the sheet, bay after headland, for what would be hundreds of miles, beneath the place where Antarctica ought to be.
The chart is signed and dated 1513. It was drawn in Gallipoli by an Ottoman sea captain who never crossed the Atlantic in his life.
Nobody would look twice at that bottom edge. Except that in 1960 a United States Air Force squadron was asked what it looked like, and wrote back.
The fragment in the Topkapi
Picture the Topkapi Palace in 1929, being emptied out.
The Ottoman sultans were gone. Their household in Istanbul was being turned into a museum, which meant that centuries of accumulated paper had to be pulled out of cupboards and sorted by people who had never seen most of it. Among those working through the material was Gustav Adolf Deissmann, a German theologian brought in to catalogue the non-Islamic holdings.
What he found was a piece of parchment with most of it missing.
The surviving sheet is roughly the left-hand third of what had been a world map. It holds the western coast of Africa, part of Iberia, the Atlantic, the Caribbean, and the long eastern shoulder of South America. Everything east of that is simply gone.
The signature belonged to Piri Reis, an Ottoman admiral and chartmaker, nephew of the corsair captain Kemal Reis, and later the author of a famous sailing manual, the Kitab-i Bahriye.
What made the map news in 1929 was not the south. It was the notes.
Piri Reis wrote on his own chart. In the margins and in the empty ocean he explained where the map came from, and he was unusually candid about it. He said he had compiled it from about twenty older charts and world maps. He named a map drawn by Columbus among his sources. He referred to charts he associated with antiquity, from the age of Alexander and from the Ptolemaic tradition.
He was not claiming to have surveyed anything. He was claiming to have copied, carefully, from things already old.
The coast along the bottom
Now go back to that southern shore.
The claim is simple to state and enormous if it holds. That the eastward-running coast at the foot of the Piri Reis map is Queen Maud Land, the Atlantic-facing arc of Antarctica. Not a guess at a southern continent. The actual coastline, with its bays and inlets in something like the right relative positions.
Antarctica was not sighted until the nineteenth century. Its true coast, beneath the ice, was not known in any detail until seismic sounding after the Second World War.
And that is the second half of the claim. What the map is said to show is not the ice front that ships see. It is the rock coast underneath, currently buried under something on the order of a mile of ice.
If that were so, somebody charted that shore while it was open water and dry land. Nobody has an accepted date for when that last was.
A sea captain, a professor, and a letter from Westover
The idea did not begin in a bookshop. It began with a retired American ship captain and engineer named Arlington H. Mallery, who had already argued in his 1951 book Lost America that pre-Columbian voyages to the New World were written into old charts.
Mallery worked over the Piri Reis map and said the southern portion matched Antarctica. In 1956 he took the argument onto a radio panel recorded at Georgetown University, alongside the Jesuit seismologist Daniel Linehan of the Weston Observatory, who had done sounding work in Antarctica, and a representative of the US Navy Hydrographic Office. The transcript circulated for years afterward.
Then the map found the man who would make it permanent.
Charles Hapgood taught at Keene State College in New Hampshire. He had already published Earth's Shifting Crust in 1958, a book on catastrophic movement of the crust that carried a foreword by Albert Einstein. He set students to work on the old charts, and in 1966 published Maps of the Ancient Sea Kings.
Hapgood's case was not only about Antarctica. He argued that the Piri Reis map and several other early charts preserved relative longitudes across the Atlantic that were better than anything Renaissance navigation should have produced. He concluded they descended from a lost source tradition, mapped by someone with a real surveying capability, long before 1513.
Inside that argument sits the document everyone quotes.
Hapgood wrote to the 8th Reconnaissance Technical Squadron at Westover Air Force Base in Massachusetts, asking cartographers there to evaluate the southern coastline. The reply, dated 6 July 1960 and signed by Lieutenant Colonel Harold Z. Ohlmeyer, is reproduced in his book. It says the lower part of the map agrees remarkably with the seismic profile of the ice cap made by the Norwegian-British-Swedish Antarctic Expedition of 1949. It says the geographical detail indicates the coastline had been mapped before it was covered by ice. It notes that the ice there is now about a mile thick.
An Air Force officer, on official letterhead, writing that sentence. That is the engine of the whole story.
Two years after Hapgood's book, Erich von Daniken put the map into Chariots of the Gods? and told his readers it resembled a photograph taken from high above Cairo. From there it went everywhere.
The case against
The conventional reading is that the coast turns east because the parchment ran out.
On this account, Piri Reis was working to the edge of his hide and bent the South American shore sideways to keep it on the sheet, a compression that early chartmakers used more than once. The historian Gregory McIntosh, whose book-length study of the map appeared in 2000, argues along these lines and finds no need for any lost survey.
There is also the matter of what Renaissance mapmakers expected to find. A great southern landmass, Terra Australis, was drawn on European maps for centuries before anyone saw it, on the theory that the northern continents needed a counterweight. Charts of the period sprout speculative southern coasts routinely.
Critics of Hapgood also point at his method. To align his coastlines he applied rotations, scale changes and grid corrections. His readers regard that as reconstruction of a damaged copy. His critics regard it as the freedom to make almost any coast match almost any other.
And then the geology. The standard account holds that continental ice has covered East Antarctica for millions of years, with the sheet forming on the order of tens of millions of years ago. On that timescale there is no human window at all. No people, and nothing to draw with.
None of that is a small objection. Taken together it is the reason the map is not taught as a mystery.
The part that will not go away
Here is the awkward center of it.
Nobody disputes that the Piri Reis map is a copy. Piri Reis said so himself, in his own hand, on the parchment. He listed roughly twenty sources and claimed some of them reached back into antiquity.
Every one of those sources is lost.
So the argument is not really about a chart drawn in Gallipoli in 1513. It is about what was on a stack of older maps that no living person has examined and no one can now produce. The mainstream account of that southern coast is an inference about a cartographer's habits. It is a reasonable inference. It is not an observation.
The Ohlmeyer letter is also still there. It has been quoted, reprinted and picked apart for more than sixty years, and it says what it says. Whether a photo-reconnaissance squadron was the right body to judge a Renaissance chart is a fair question. It was not asked as a favor to a fringe theory, and nobody has produced a retraction.
And the map keeps its strangest quality no matter which way the Antarctic question falls. It is a 1513 document that openly declares itself an assembly of older knowledge, in a period when most maps declared the opposite.
What the parchment will not say
A coast runs east along the bottom of a piece of gazelle skin.
Either it is South America folded to fit the page, and the resemblance to a buried shoreline is the kind of accident that happens when two irregular lines are laid over each other.
Or Piri Reis copied it faithfully from something much older, and the thing he copied knew a shore that has not been open to the sky in a very long time.
The hide has been read for nearly a century and has not settled it.