Page:Monograph on Leonardo da Vinci's Mona Lisa (1915).pdf/22

 have been but natural to him to paint two versions, so as to secure the very best results he possibly could whilst she was sitting for him. But a very few months previously he had made two drawings of Isabella d'Este, when she gave him the commission for her portrait in oils, which he never painted. The fact that both the Isleworth and the Louvre Mona Lisas are painted on canvas substantiates this, for all of his other works extant to-day are on panel. The Louvre Mona Lisa is described in the new edition of Vasari as ' Canvas on Panel.'

In any case, we have it on the unquestioned and unquestionable authority of Fra Nuvolaria that he saw two of the master's pupils at work on two portraits; which means that they were—as was usual in those days with artists—filling in the details, he himself having painted the heads and principal portions of the portraits.

One of these portraits must have been the Mona Lisa, since, according to Vasari, Milanesi, and Ravaisson-Mollien, it was commenced in 1500. The other portrait we know Vasari was wrong in describing as that of Ginevra Benci; and in view of the fact that, at this very time, the Marchesa Isabella d'Este could not induce Leonardo to paint her portrait on any conditions whatsoever, is it not a logical inference to assume that this second portrait was a second version of the Mona Lisa, more especially as there is not a hint or a suspicion that Leonardo painted any one else's portrait at this period?

Again, what reason, what evidence, or what authority have we for even surmising that one of the portraits seen by Fra Nuvolaria was not the Mona Lisa? None, except Müntz's ipse dixit—'What they were is unknown.' But Müntz's opinions and conclusions are by no means infallible.

As an instance, I give the following specimen of some of the fiction that has been written about the Mona Lisa. Vasari, who had