When it surfaced in 1957, it was too good to be true: a purported 15th-century world map depicting an island to the far west labeled Vinilandia Insula -- the fabled Vinland -- proof positive, it seemed, that Norse explorers had reached North America long before Columbus.

Thanks -- but no thanks -- the British Museum told the intermediary who offered to sell it to them. It's a phony.

Later that year, however, New Haven, Conn., book dealer Lawrence Witten bought the map and an accompanying medieval manuscript for his wife, paying $3,500. Soon after, he visited Yale University Library to view a seemingly unrelated manuscript fragment purchased by Thomas E. Marston, the library's curator of medieval and renaissance literature. Witten asked to borrow it.

That night, Marston got an excited call from Witten. Marston's manuscript, Witten's manuscript and the map were all written in the same hand, Witten said. Furthermore, worm holes in all three works matched up. They apparently had been bound together, with Marston's manuscript as the meat in the sandwich. The map had to be real.

Thus began the affair of the "Vinland Map," a 13-by-19-inch sheet of parchment depicting not only Vinland, but also remarkably detailed renderings of Iceland and, especially, of Greenland, which -- if the map is real -- is portrayed as an island for the first time in history.

Forty-five years after the map's "discovery," its authenticity remains a subject of fierce debate. In the last two months, the journal Analytical Chemistry has published two articles by front-line combatants in the dispute.

One, by retired Smithsonian research chemist Jacqueline Olin, argued that the presence of anatase, or titanium oxide, in the ink did not mean the ink was modern, as had been alleged in earlier research. She suggested the ink may well have been medieval, made from a simple leaching process from the titanium-rich mineral ilmenite.

The other, by Kenneth Towe, also a retired Smithsonian analyst, reminded readers that the map's anatase had a crystalline structure identical to commercial anatase, a ubiquitous synthetic compound used to enhance colors in paint. Olin's analysis, Towe charged, was "a 'rehash' that is too often biased, misleading or inaccurate."

In May, Danish businessman Jorgen Siemonsen, a well-known debunker of Viking frauds who is agnostic on the map, will sponsor a debate between believers and skeptics as part of a conference on the "Dynamics of Northern Societies."

And coming a month later will be a book-length study titled "Maps, Myths and Men, the Story of the Vinland Map," which will make the case that it is a 1930s forgery by a German Jesuit priest intent on making the Nazis look like fools.

At this juncture, a preponderance of evidence points toward forgery, but the argument is not over, and the stakes are high. If it is authentic, the map is priceless, the oldest known depiction of North America. Yale's Beinecke Rare Book and Manuscript Library, the map's current resting place, at one point reportedly insured it for $25 million. If it is not authentic, however, it is an amusing curiosity -- worth what Witten paid for it, perhaps, but not much more.

The Yale library refused a request to discuss the map except to say that it takes no position on its authenticity.

"It's going back and forth, and it will continue going back and forth," said William W. Fitzhugh, director of the Smithsonian Institution's Arctic Studies Center. "The people who made the studies are defending them, and I don't think it will ever be solved."

For 15 years after the Witten phone call, belief in the map's authenticity was ascendant. The British Museum's skeptics became believers. Philanthropist Paul Mellon bought the map and gave it to Yale. Then in 1960 a Norwegian husband and wife team discovered remains of a Norse encampment in Newfoundland, proving that Greenlanders had, in fact, reached North America.

There was dissent, however. Researchers were furious with Yale for keeping the map out of sight until Columbus Day 1965, choosing public relations pizzazz when orthodoxy called for prompt publication of scientific analyses. A team of British investigators questioned whether the ink was medieval. Paleographers questioned the handwriting.

And then there was the fact of the map itself. "The Norse never made maps," said Norwegian-born historian Kirsten Seaver, author of "Maps, Myths and Men." "When you are Norwegian and you see something like that, you say it is so fake, there's no use bothering with it. And how would a 15th-century mapmaker know Greenland was an island?"

"There was a medieval warm period" that may have allowed the Icelanders to colonize Greenland in the 10th century, but "it wasn't that much warmer" than today, Seaver said. "You'd be hard put to sail around Greenland today."

Still, the debate did not turn until 1973, when chemist Walter C. McCrone analyzed the map with polarized light microscopy and found that the yellow "aging stain" seeping from beneath the map's lines was made of synthetic anatase, a substance patented in 1917.

"That seemed to put it in the grave," Fitzhugh recalled.

But others disagreed. Olin made anatase from ilmenite, using a process that would have worked for medieval scribes. "There have been too few medieval inks analyzed" to make categorical statements about them, Olin said in an interview, responding to criticisms by McCrone, and later, Towe, that her anatase did not match the map's.

The case for authenticity was strengthened in 1987, when a University of California at Davis team led by Thomas Cahill reanalyzed the map using X-rays and concluded that titanium was present, but only in minute quantities, calling into question McCrone's analysis. Yale trumpeted these results in a 1995 revision of its original report on the map.

McCrone, who died in 2002, never wavered in dismissing it as a fraud. And that year a British team from University College, London, used laser analysis of the stain to draw conclusions identical to McCrone's. "I'm not taking any glamour away from McCrone," said chemist Robin J.H. Clark, co-author of the British study. "We used a completely different technique to obtain the same conclusion. I think the matter is over."

Not quite. At the same time, an Olin-led team published results from radiocarbon testing that dated the parchment to 1434 A.D. -- proof that the paper was old enough, even if the map wasn't.

And it isn't, Seaver said. Fascinated, she read everything written about Norse exploration since the 18th century and gradually homed in on the Rev. Josef Fischer, a German Jesuit cartographer and prolific Norse historian who died in 1944, as her chief suspect.

Seaver theorizes that Fischer, upset at Nazi persecution of the Catholic Church, drew the map in the late 1930s and deliberately larded it with religious references. "The Nazis loved to talk about a Nordic heritage, and the map was great for them," Seaver said, "except that it also told the story of how the Roman Church had been there from the start. It presented them with a wonderful dilemma."

The map, maybe the earliest including North America, is at Yale University, which is neutral on its authenticity.