Rembrandt's "The Man With the Golden Helmet" is the most recent example of the search for an answer to one of art's eternal questions. How do you prove a work of art is by a particular artist?

There are no absolute answers in the field of art authentication, and truth is more a matter of general agreement. Art experts have long been skeptical about the questionable Rembrandt, which hangs in the Staatliche Museums in West Berlin.

A consensus has recently developed, however, about "The Man With the Golden Helmet." Among those who say the work is not a Rembrandt are international Rembrandt authority Jan Kelch and experts at the Rembrandt Research Project in Amsterdam and the Staatliche Museums, which recently published a guide to its collection suggesting that the work is a good contemporary imitation of the 17th-century master.

The artist is a difficult one for art connoisseurs generally. Rembrandt didn't keep good notes on what he created -- and where the paintings went. His more than 100 pupils often worked with him on canvases. Experts must examine characteristic qualities, such as thickness of paint, the way a figure is posed and the mood of a work.

Lots of other artists are massively copied or faked -- perhaps to provide future art historians something to argue over. One artist who can get people debating for hours is Giorgione (1477-1510); works possibly by him at the J.B. Speed Art Museum in Louisville and the National Gallery of Art here are current subjects of controversy. Not much is known about Giorgione. He was an apprentice in Giovanni Bellini's workshop along with Titian, and the work of all three painters is frequently confused.

Jesse Wright, director of the J.B. Speed museum, isn't convinced that his museum's Giorgione -- "The Three Stages of Man," which is nearly identical to another painting of the same name in the Pitti Palace in Florence -- is the real thing. "I am convinced it is a painting of the period," he said. "It could be a Giorgione, and it is conceivable it is not."

In Washington, curators at the National Gallery believe their Giorgione -- "The Allendale Nativity," named after a previous owner and donated to the museum by Samuel Kress -- is genuine, but others disagree.

Works of art frequently are discovered to be fake or "improperly attributed" to an artist. On occasion, a lightly regarded picture is found to have been painted by a major artist -- for example, Frederick Church's "Icebergs," which was found hanging in an obscure boarding school, later sold at auction for $2.5 million -- but usually it is discovered that a lesser known artist actually created a piece formerly attributed to some master.

Last year, the National Gallery reattributed more than 250 works in its collection based on research by the curatorial staff. This is part of an ongoing effort at the museum to conclusively document all of its works. While some pieces were elevated in terms of their attribution -- for instance, a sculpture that had been labeled "Florentine, 16th Century" was upgraded to "Benvenuto Cellini" and a painting thought to be by a minor Italian Renaissance artist, Dosso Dossi, was reattributed to Tintoretto -- most of the others were downgraded. An "attributed to Cimabue" became "by followers of Cimabue" and several paintings once thought to be by English landscape artist Thomas Gainsborough are now said to have been done by Gainsborough Dupont who, it should be said, lived at the same time as the better known artist with the similar name.

It is generally more difficult to prove that a work of art is authentic than to cast a shadow of a doubt upon it, due to the general lack of documentation (receipts, letters in which works are described, photographs) that informs scholars who did them and when. With more modern pieces, there is far less difficulty in finding the trail from the artist's studio through the art gallery where it was sold to the collector. Possibly the piece was sold several times, but bills of sale exist to show conclusively the "provenance" -- a work's history of ownership.

There is no great difficulty in proving that the wall and ceiling paintings in the Sistine Chapel in the Vatican are really by Michelangelo since they were never moved, but most other older works of art changed hands in largely undocumented fashion for centuries.

Artists haven't always signed their works nor were these pieces always written about by contemporaries such as Giorgio Vasari, whose "Lives of the Artists" has helped later art historians understand how Michelangelo and others worked, what pieces they created and who commissioned them.

Vasari himself was something of a chauvinist for his native Florence, promoting home-town artists with only grudging recognition of artists elsewhere, such as in Venice where Giorgione lived and worked. Written documentation as to authenticity can be quite scarce -- especially so for an artist's sketches and preliminary drawings, which have historically been thought of as less important than the "finished" paintings and sculptures -- and one has to look elsewhere for evidence.

Increasingly, curators have taken their inquiries to laboratories to learn more about the work. This often involves microscopic examination of paint pigments and the canvas or paper or wood that the picture was painted on, as well as use of infrared, ultraviolet or X-ray technology. Infrared light penetrates several layers of paint to reveal, for instance, a hidden signature, while ultraviolet light shows any restoration and overpaint. X-rays help identify any changes the artist made in the course of creating the work or detect whether one picture was painted on top of another. Laboratory analysis cannot tell the curator if a particular artist did the work but can indicate whether the piece was created in the time the artist was working.

Most of the major museums in the country have some sort of in-house laboratory facilities for this kind of analysis but in a pinch they call on outside professionals. The McCrone Research Institute in Chicago and the Institute of Paper Chemistry in Appleton, Wis., are the first places museum officials are likely to call.

Walter McCrone, founder of the McCrone Institute, was asked to examine the J.B. Speed's Giorgione, and he concluded that the paint and canvas of the picture date to the years in which the artist was working. "It could be by Giorgione -- I can't say that it isn't," he said.

He was more positive when he determined that a painting reputed to have been done by German artist Matthias Gru newald (1470-1528) in the Cleveland Museum of Art was a forgery because the paint contained a kind of white lead that wasn't in use until the 20th century. McCrone created a controversy a few years back when he claimed that the image in the Shroud of Turin, which has been said to be the face of Jesus, was formed not with blood, as had been claimed, but with paint.

Unless a "fake" work is created at the same time as the original and with the same basic materials, chemical analysis can generally spot the differences. Paints, for instance, have continually undergone change and improvement over time. Canvases, too, can be dated according to their fiber composition and the regularity of the weaving (machine-made canvases didn't come into existence until the mid-19th century).

Dutch forger Han van Meegeren fooled many collectors into purchasing paintings "by" 17th-century artist Jan Vermeer that he had actually painted himself. However, his application of formaldehyde in the paint -- which made it crack quickly and look old -- was eventually discovered.

The Institute of Paper Chemistry also analyzes pigment for works done on paper, but it chiefly examines the fiber content (for instance, wood, straw or rag) and the pulping process to determine the paper's approximate age. For example, the paper that most newspapers are printed on was not created until 1869, though "rag goes back to when Adam was a kid," Bill Kreuger, a consultant to the institute, said.

He pointed out that a number of different and distinct elements can give clues to when the paper was made, such as watermarks (the brand of the manufacturer) and dyes as well as the glues, rosins, starches and waxes that often are added to paper.

"You have to look at a lot of things when you check a work's authenticity," he said. "The style and technique are one aspect, but I'm not equipped to discuss an artist's work in that way. All I do is supply one part of the puzzle. In the field of chemical analysis, it's not so much what you know, it's do you know where to find it."

Chemical analysis provides a factual basis for discussing a work's authenticity. "Science -- that is, factual, repeatable evidence -- takes the matter out of the area of heated emotions," said Ross Merrill, a conservator at the National Gallery.

But it is still only a starting point. It is connoisseurship -- the sense an expert has of the less tangible qualities of a picture, such as the kind of brushstrokes used and the artist's preferred subject matter, as well as the overall composition and drawing -- that leads to questions about a work's authenticity in the first place. If the scientific analysis is not conclusive, as with the J.B. Speed's Giorgione, the matter may still rest in the area of heated emotions.

Up until this century, connoisseurs of art learned about the style of particular artists by traveling to the museums and other places where their works were shown. Bernard Berenson (1865-1959) was the first connoisseur to use photographs of works as a way of saving time in evaluating the authenticity of other pieces by the same artists. That shorthand way of helping to train one's eye is still considered extremely important, and the J. Paul Getty Museum in Malibu, Calif., is attempting to compile photographs of every recognized work of art ever created.

Still, experts continue to seek out factual evidence for making correct attributions. Moussa Domit, former assistant director at the Corcoran Gallery, pointed out that labels or inscriptions on the back of a stretcher or frame may give some clues to where a picture has been and when. He once found a painting attributed to Monet that had a label stuck on the stretcher belonging to a frame shop in Hartford, Conn. Through that and subsequent research, he determined that the picture could not have been by the artist, though it might have been done by an American Impressionist of the same era as Monet.

"One thing about Monet is that he signed most of his paintings," he said. "In addition, there are records of most of the Impressionists' exhibits as well as extensive correspondence between the artists in which they mentioned particular paintings. With all that, it is difficult to find a Monet popping up from nowhere."

Descriptions of a work in an artist's notebook also help validate it and, certainly, if the artist is alive, he or she can be asked directly. That's not always foolproof, however. Toward the end of his life, Norman Rockwell denied that he had painted certain pictures that he undoubtedly had done early in his career because he was ashamed of them.

Surrealist painter Giorgio de Chirico (1888-1978) signed documents saying that works painted in the 1950s and '60s had actually been done in the 1920s. The reason for this is that his earlier works regularly sold at higher prices than the later pictures, and he tried to cash in. After a while, no one took his word anymore and simply went to his dealer for verification.

Laboratory tests probably could not have distinguished which paintings had been done in the 1920s and which in the '50s; it was solely a matter for connoisseurship. In the end, it all comes down to instinct and taste. From the moment the artist dabs paint on his brush until the time an authority claims the picture is authentic a work of art is subject to questions. Collectors, historians and even the artist all look at it and wonder if it is worthy of its creator and how it stands on its own. Art is an existential pursuit, transcending the bounds of fact and logical reasons, and validating a work requires no less feeling and intuition.

"How can we prove the authenticity of a given work of art?" George Shackleford, curator of European painting at the Houston Museum of Art, asked. "For that matter, how can we prove our own existence?"