Follower of Leonardo da Vinci, probably before 1750, "Portrait of a Woman", called "La Belle Ferronnière". (55 x 43.5 cm) 21 5/8 x 17 1/8 inches. Est. $300/500,000. Sold for more than $1.5 million. Photo: Sotheby's

NEW YORK, NY (REUTERS).- A portrait once believed to have been painted by Leonardo da Vinci sold for more than $1.5 million in New York on Thursday, around treble the top price estimated ahead of the auction.

Auction house Sotheby's said another version of the portrait, "La Belle Ferronniere," is in the Louvre in Paris and it is that painting that is now generally agreed to have been by da Vinci himself or one of his pupils, while the painting sold in New York was a later copy.

The painting is thought to be of Lucrezia Crivelli who was a mistress of Ludovico Sforza, the Duke of Milan, and was painted sometime before 1750 by a follower of the Renaissance master.

But for more than 80 years the identity of the artist was in dispute and the controversial work was the subject of a slander trial in the 1920s, as well as two books.

"The story of "La Belle Ferronniere" is as much about the aesthetic and scientific foundations of modern art history as it is about the authenticity of the painting itself," Sotheby's said in a catalog note.

Sotheby's said on its website that the painting, artist unknown, sold on Thursday for $1.538 million, compared to pre-auction estimates of $300,000-$500,000. It did not give details of the buyer.

"La Belle Ferronniere" was brought to the United States in 1920 by newlyweds Harry Kahn, an American serviceman during World War One, and his French wife, Andree, who received the painting in 1920 as a wedding gift from her grandmother.

It was thought to be have been painted by da Vinci and authenticated by a French art expert.

But when Kahn later tried to sell it to the Kansas City Art Institute in the United States for $250,000, Joseph Duveen, a leading art dealer in London, told a reporter who questioned him about the sale that he was sure it was a fake.

This sparked a decade long legal battle that brought into question Duveen but also the very foundations of connoisseurship, according to Sotheby's.

The deal fell through and Kahn's wife sued Duveen for slander, claiming he was trying to control the art market.

The case riveted the art world. Although other art experts supported Duveen's assessment, because of a lack of scientific and archival evidence the jury failed to reach a verdict. Duveen settled out of court before a retrial.

Experts believe the portrait must date before 1750 because it contains lead-tin yellow, a color that was used in paintings up until the late 17th century.

Unlike the Louvre version, which was painted on a poplar panel, typical during da Vinci's time, the Hahn painting was done on canvas, which was commonly used much later. (Writing by Patricia Reaney and Belinda Goldsmith, Editing by Miral Fahmy)

PROVENANCE: Said to be in the French Royal Collections, from which it was supposedly removed and sold during the Revolution by August Cheval de Saint-Hubert and purchased by General Louis Tourton, 1796;¹
Comte de Betz, before 1847;
His sale, 1847, where acquired by Antoine Vincent ("Tableaux à M. Vincent vendu après la Collection du Comte de Betz provenant du General Tourton, Leonard da Vinci et Claude Lorrain.");
By descent in the family to his grand-daughter, Louise de Montaut;
By whom given to Andrée Lardoux and her husband Harry Hahn, Kansas City, by 1920;
With Leon Loucks, 1975;
Returned to the heirs of Andrée and Harry Hahn, 2009.

EXHIBITED: Kansas City, Nelson-Atkins Museum, 8 December 1946 - 5 January 1947;
Los Angeles, Hartwell Galleries, April 1947;
San Francisco, Dallas and Houston, National Dry Goods Association, 1947.

LITERATURE AND REFERENCES: A. Venturi, "Leonardiana," in L'Arte, vol. XXVII, 1924, p. 49-50, reproduced in black and white;
R. Cortissoz, "La Belle Ferronnière: Coverage of the Hahn-Duveen Trial," in The Herald Tribune, February - April 1929;
"Duveen Settles," in The Art Digest, vol. V, 1 May 1930, p. 13;
H. Hahn, The Rape of La Belle, Kansas City 1946, with an introduction by Thomas Hart Benton;
A. Decker, "The Multimillion-Dollar Belle," in Art News, summer 1985, pp. 86-97, reproduced p. 86, details pp. 92, 94, 95;
M. Secrest, Duveen: A Life in Art, New York 2004, pp. 224-243;
J. Brewer, "The Lure of Leonardo," in The Lure of the Object, Stephen Melville, ed., New Haven 2005, pp. 3-14;
J. Brewer, "Art and Science: A Da Vinci Detective Story, " in Engineering and Science, no. 1/1, 2005, pp. 32-41;
J. Brewer, The American Leonardo: A Tale of Obsession, Art and Money, Oxford and New York 2009

NOTE: This striking and (in)famous portrait is another version of a composition in the Louvre, thought to depict Lucrezia Crivelli, a mistress of Ludovico Sforza, the Duke of Milan.² Depicting a lady in three-quarters profile, the sitter in the present painting is seated behind a stone balustrade that is only just suggested by the strip of gray paint at the bottom. The woman gazes out of the canvas, while a slight smile is just suggested by the upturned corners of her mouth. Her hair is parted down the center and pulled back, and she wears as jeweled headband across her forehead. The sitter's pale, milky skin is offset by the rich red gown and green ribbons she wears, and the square, embroidered neckline of her gown is complimented by her delicate necklace.

Although the attributions of both the present painting and the work in the Louvre have in the past been the subject of debate between experts and scholars, it is today typically agreed that the Louvre version is either by Leonardo himself or one of his pupils, while the present version is thought to be a later copy. The history of these judgments, however, is not nearly as straightforward as the above would seem to imply. On the contrary, the history of the present painting is intertwined with the fields of connoisseurship and art history as they emerged and asserted themselves in the early part of the 20^th century. The story of La Belle Ferronnière is as much about the aesthetic and scientific foundations of modern art history as it is about the authenticity of the painting itself.

This Belle Ferronnière was brought to the United States in 1920 by newlyweds Harry and Andrée Hahn. Harry, an American serviceman during World War I, had met and married a young French woman named Andrée Lardoux. With the war over, Harry returned home to Kansas City with his new bride and this picture, a wedding gift from Andrée's godmother, Louise de Montaut. Shortly after their arrival in the U.S., the Hahns decided to sell their picture, contacting several dealers on the East Coast as well as local dealers in Kansas City. Later, they claimed to have nearly reached a deal with the Kansas City Museum for $250,000, when a reporter at the New York World, who had heard that a woman in Kansas City was trying to sell a Leonardo, telephoned the famed art dealer Sir Joseph Duveen in London to ask his opinion on the painting. It was 1AM on the morning of 17 June 1920 when Duveen answered his phone. The art dealer had never seen the Hahn's picture, but nevertheless he told the reporter that it must be a copy and that the real work was in the Louvre. Although he couldn't know it at the time, this seemingly routine telephone call would unleash a decade long legal battle that brought not only Duveen himself, but also the very foundations of connoisseurship, under trial.

In November 1921, the Hahns served papers on Duveen at his gallery in New York City. He was being sued for slander, for as the Hahns claimed, his comments to the press had not only been false, but also had been purposely calculated to drive them and their picture from the market. Duveen rallied a host of art experts to his side, including Bernard Berenson, Wilhelm Valentiner and Roger Fry, who all unanimously agreed that the Hahns' picture was a copy. He even orchestrated - at his own expense - a confrontation between the Louvre's and Hahns' pictures in Paris. There too, the unanimity of opinion between the various art experts was almost unheard of in the history of connoisseurship: the Louvre's picture was the original, while the Hahn's was a later copy. Duveen could sense his victory in the courtroom.

But as the trial began, it became clear that the jury was not on his side. The average American had little patience for a group of "experts" who could not provide one bit of hard, factual evidence that the work was a copy, and who, when asked to explain how they had reached their conclusions about the pictures, were reduced to vague statements about the superiority of their "eyes." The Hahns' lawyer saw his opportunity: while Duveen's counsel was attempting to show how unified this disparate group of experts was in their opinions on the Hahns' picture, the Hahns' lawyer began to poke holes in this unified front by questioning the experts on their opinions of the Louvre picture. Many had written articles in the past claiming that it too was a copy. If, he asked the jury, the experts had changed their opinions of the Louvre painting, what was to say that their current opinions of Hahns' picture might not also change? In 1929, with no scientific or archival evidence to back up the testimony of Duveen's experts, the jury could not reach a decision, and the case ended in a hung jury.³ Duveen ultimately settled out of court, paying the Hahns $60,000 in damages.

This settlement did little to change the market's opinion of the authenticity of the Hahns' picture. Duveen continued to dominate the international art market, and the picture was locked away in a bank vault in New York during the Depression and World War II. In 1947 Harry Hahn published a book, The Rape of La Belle, which blasted the art establishment for its treatment of the picture, and which claimed to conclusively prove that it was by Leonardo. The book coincided with renewed efforts to sell the picture, both in the United States and abroad. Some of these attempts included figures from the art world, such as the renowned conservator Helmut Ruhemann, who opined that the Hahn painting was, in fact, a Leonardo, and the controversial promoter of pictology, Maurits van Dantzig. However, it all came to nothing and the work disappeared from public view.

In 1993, La Belle Ferronnière was examined by Leonardo expert Professor Martin Kemp. While Kemp did not think that the painting was a Leonardo, he did believe that it had age, dating it to the first half of the 17^th century. He also suggested an attribution to a northern European painter, perhaps Laurent de la Hyre.

Recent technical examination of the painting has revealed compelling new evidence about the genesis of the picture, much of which supports Kemp's theory. As noted, while the Louvre example is painted on poplar panel, the expected and typical support for a late quattrocento Florentine portrait, the Hahn picture is painted on canvas, a material which became more common in the following centuries.⁴ Indeed, the priming layer on the canvas is a double red pigment, common in French pictures from the late 17^th to the early 19^th centuries. Pigment analysis has also suggested a date ante quem for the picture. Most compelling is the use of lead-tin yellow, a color which was employed by artists from very early and up until the late 17^th century. After that time, however, it began to be replaced on the artist's palette by other yellows, such as Naples yellow and ochre, and the formula for lead-tin yellow itself was eventually lost and appears to have disappeared by the early nineteenth century. It was not until 1941 that scientists were able to isolate the composition of the compound, and in particular the presence of tin, which separated it from other yellow pigments.⁵ These features, then, would suggest that the Hahn Belle Ferronnière was most likely painted by a French artist (or an artist using French materials) in the years up to about 1700/50, after which time some of the pigments used would have become obsolete, unavailable or forgotten. It is important to remember, however, that technological advances are currently improving our knowledge of artists' materials, and it is possible that future investigations will yield new information on the use of this and other pigments. Such findings are enticing, for they suggest that La Belle Ferronnière may finally emerge not only from the shadow of Leonardo da Vinci, but also from close to a century of controversy and critical neglect, to be understood on its own terms.

1. This provenance is traditional to the picture; however, it is also highly speculative. The first firm documentation we have of the picture is its appearance at auction in 1847 as part of the collection of the Comte de Betz, as referenced in Brewer, The American Leonardo, p.132.
2. Ludovico was one of Leonardo's most important patrons, and the artist also painted another of his mistresses, Cecilia Gallerani, in his famed Lady with an Ermine, in the Czartoryski Museum, Cracow, Poland.
3. Duveen had experts in archival research and x-ray analysis examine La Belle Ferronnière but he decided not to use their testimony in the court case, thereby depriving the jury of some of the evidence that could have swayed them to his side.
4. Examination has confirmed that the Hahn La Belle Ferronnière was not originally painted on panel and then transferred to canvas at a later date, as the inscription on the back suggests.
5. Naples yellow began to become more common on artists' palettes from the late 17^th century on, and in particular with painters who had an Italian association. It seems most likely that various yellows were used, and that lead-tin yellow was used in Northern Europe longer than further south, perhaps as late as the 1770s. Indeed, its use in the ceramics industry in the Netherlands in the second half of the 18^th century is recorded, although it seems likely that innovations in this area would lag behind those in painting.