To collect a fake is a mistake. To call a real work a fake is a sin. To prove that a fake is a genuine masterpiece is glory.

Tall, lanky, bespectacled Kurt Sundstrom, a young man of gravity and wit, the associate curator of the Currier Museum of Art in Manchester, N.H., and  expert on Renaissance and Baroque art, hit what he hoped was glory in 2001.

He had heard that the museum owned a painted stucco relief of the Madonna and Child by the 15th-century Florentine sculptor Antonio Rossellino, purchased in 1941 for $2,500 from a New York gallery that had itself gotten the work from the estate of Metropolitan Museum of Art president Robert DeForest. It had been exhibited until the 1950s and then was placed in storage. No one knew why. It could have been that its condition was too delicate to show or, worse, that it was a fake, made by the notorious Giovanni Bastianini, who, in the 19th century churned out a hoard of Rossellinos.

When Sundstrom encountered the relief and its wooden frame on a shelf in the basement, he instantly had what he calls the "Oh my God!" moment. He was stunned by how beautiful the relief was and, despite its rough condition, sensed a sort of electric pulse coming out of it. The Virgin is gorgeous in a solid, down-to-earth, very Italian way. She’s a placid young woman with a broad brow, sleepy eyes, a pleasingly pointed nose and a contented smile. She tightly and lovingly holds the robust Christ who displays his genitals -- the Renaissance way of signifying his being the son of man. Behind the pair is a painted rose trellis. The gilded wooden frame was beginning to separate from the heavy stucco (made of burnt marble, plaster, ash and wood chips.) Sundstrom was impressed that the worm holes didn’t go straight in and stop -- the sign of a sure fake -- but went straight in a little bit and then turned abruptly. Real worms burrow straight through the yucky-tasting paint or gilding and then chomp away delightedly in the wood just underneath.

To back up his blind impression that this was a rare and fine Rossellino, Sundstrom got the conservators Lance Mayer and Gay Myers, who worked for the Currier, to scrutinize the relief and give some advice on what needed to be done to bring it back to exhibitable shape. Although Mayer found the piece "blotchy and hard to understand," it looked to him like what a 15th-century work ought to look like. The conservation work would cost $12,000. The piece was driven to Mayer’s studio in New London by SUV.

Meanwhile, curator-detective Sundstrom followed another lead -- tracing the frame to see if the maker might be identified. He sent digital photos to the best expert in the field, Laurence Kanter, the curator of early Renaissance art at Yale’s museum and formerly curator of the Robert Lehman collection at the Metropolitan, a collection full of wonderful Renaissance and Baroque frames.

Wham!

Kanter condemned the "Rossellino" as a blatant fake, probably by the prolific Bastianini. He pointed to nearly a dozen fakeroos just like the "glory" Kurt Sundstrom had fiddled up in his eager brain. Sundstrom bowed. He had so little experience compared to Kanter, the master. Sundstrom called conservator Mayer -- stop the work!

Mayer did -- almost. Still intrigued as to the possible genuineness of the sculpture, he took paint samples -- so tiny that you couldn’t tell anything had been taken. Eight samples of the greens, the blues, and one light yellow from a rose. He sent them to Henry DePhillips, a chemistry professor at Trinity College in Hartford, who specialized in ancient paint. Upon analyzing the yellow bit, DePhillips discovered that it was lead tin yellow.

Glory!

Why? Because lead tin yellow came into use during the second quarter of the 15th century and by 1710 had never been used again. Bastianini presumably didn’t even know of the existence of lead tin yellow.

Kurt Sundstrom persuaded Laurence Kanter to come from Yale to the Currier to see the "Rossellino" in the flesh -- he chauffeured Kanter himself. It took only a minute or two for the expert to proclaim that the Virgin and Child was the real Renaissance McCoy. And Kanter admitted gracefully that he had been fooled by the photos and by all the proven Rossellino forgeries.

The recently restored masterwork is now proudly on view at the Currier. (To read the detailed history of the re-discovery of the treasure here’s a link to a series of excellent articles written by the reporter Mike Pride for the Concord Monitor.)

Lead tin yellow had an odd role in my career. It was in the mid-1970s when a British writer was about to slam as a phony Georges de la Tour’s amusing masterpiece in the Met, The Fortune Tellers, showing a young dandy being fleeced by gypsies who pretend to read his palm. The Brit claimed the costumes were not 17th century. He had found the word "merde" painted in the lace collar of the fetching gypsy girl conning the dandy. Not a 17th-century word. My conservator, Hubert von Sonnenberg, discovered that the "merde" had been painted in by the restorer the dealer of the painting had hired -- the restorer’s own quirky paintings were full of scatalogical words. He had left a calling card.

And, more importantly, von Sonnenberg confided to me in his studio that he had found a substance in the picture which virtually no curator knew about, lead tin yellow, which was never used after 1700.

Well, almost never. Years later von Sonnenberg told me that a bunch of newly found de la Tours and Caravaggios all had lead tin yellow. Still, he swore they were fakes. He believed that an assistant, whom he fired shortly after the Georges de la Tour incident, had listened in on his conversation with me and started up a studio faking 17th-century masters.

So, if you’re about to buy a 17th-century masterwork, don’t rely simply on lead tin yellow as a passport to glory.

THOMAS HOVING is author of Master Pieces: The Curator’s Game (2005).