New tests on a painting billed as the original version of the Mona Lisa, Leonardo da Vinci's 15th century portrait, have produced fresh proof that it is the work of the Italian master, a Swiss-based art foundation said on Wednesday.

The tests, one by a specialist in "sacred geometry" and the other by the Swiss Federal Institute of Technology in Zurich, were carried out in the wake of the Geneva unveiling of the painting, the "Isleworth Mona Lisa", last September.

"When we add these new findings to the wealth of scientific and physical studies we already had, I believe anyone will find the evidence of a Leonardo attribution overwhelming," said David Feldman vice-president of the foundation said.

The "Mona Lisa" in the Paris Louvre for over three centuries has long been regarded as the only one painted by Leonardo - although there have been copies - and claims for the Swiss-held one were dismissed by some experts last year.

But it also won support in the art world, encouraging the Zurich-based Mona Lisa Foundation - an international group which says it has no financial interest in the work - to pursue efforts to demonstrate its authenticity.

Feldman, an Irish-born international art and stamp dealer, said he was contacted after the public unveiling of the portrait - which shows a much younger woman than in the Louvre - by Italian geometrist Alfonso Rubino.

LEONARDO'S GEOMETRY

"He has made extended studies of the geometry of Leonardo's Vitruvian Man" - a sketch of a youth with arms and legs extended - "and offered to look at our painting to see if it conformed," Feldman told Reuters.

The conclusion by the Padua-based Rubino was that the "Isleworth" portrait - named for a London suburb where it was kept by British art connoisseur Hugh Blaker 80-90 years ago - matched Leonardo's geometry and must be his.

The Zurich institute, the Foundation said, carried out a carbon-dating test on the canvas of its painting and found that it was almost certainly manufactured between 1410 and 1455 - refuting claims that it was a late 16th century copy.

Earlier brush-stroke studies presented last September by U.S. physicist and art lover John Asmus concluded that both the "original" version and the Louvre crowd-puller were painted by the same artist.

The authenticity of the foundation's painting, discovered by Blaker in an English country house in 1913, has been fiercely challenged by British Leonardo authority Martin Kemp, who argued last year that "so much is wrong with it."

Feldman and foundation colleagues retort that Kemp has never followed up on invitations to come to see it.

Documents show that a painting of his wife Lisa was commissioned around the turn of the 16th century by Florentine nobleman Francesco del Giacondo. In French, the Louvre version is known as "La Giaconde" and "La Giaconda" in Italian.

Supporters of the "younger" version say it was almost certainly delivered unfinished to del Giacondo before Leonardo left Italy in 1506 and took up residence in France, where he died in 1519 in a small Loire chateau.

From the Giacondo house, it probably eventually found its way to England after being bought by a travelling English aristocrat, this account runs, while the Paris version was probably painted by Leonardo around 1516 in France.