Skip to main content

'The Virgin and Child with an Angel', after Francia: A History of Error

Ashok Roy and Giorgia Mancini
Technical Bulletin Volume 31, 2010


The Virgin and Child with an Angel (NG3927) came into the National Gallery Collection in 1924 with the Mond Bequest. It was believed to be by Francesco Francia (1450–1517/18), an attribution based on the design of the picture, an inscription and a date of 1490.

In 1954 a second version of the painting came to light and the new owner asserted that this work (now in the Carnegie Museum of Art in Pittsburgh) was an original by Francia, and that the National Gallery painting was a copy or a forgery.

The Gallery’s Trustees instigated technical inquiries and concluded that the London painting was most probably a forgery. This conclusion was called into question in the 1990s, and, on the basis of newer technical investigation, NG3927 was published in 1998 as an authentic early work by Francia.

In preparation for an exhibition Close Examination: Fakes, Mistakes and Discoveries (30 June–12 September 2010), NG3927 was again examined and further new technical evidence reasserted the 1950s conclusion, although for different reasons, that the picture was indeed a fake produced in Italy in the 19th century, perhaps based directly on the authentic Pittsburgh version. This article follows the changing fortunes of 'The Virgin and Child with an Angel' and explains the history of error in a long-running technical investigation.


Francia, fifteenth-century panel painting, forgery, fake, transfer, lead-tin-antimony yellow, technical study of fakes

Download article

'The Virgin and Child with an Angel', after Francia: A History of Error', Ashok Roy and Giorgia Mancini (PDF 29.35MB)

View high resolution infrared images

To cite this article we suggest using

Roy, A., Mancini, G., 2010. '''The Virgin and Child with an Angel", after Francia: A History of Error'. National Gallery Technical Bulletin Vol 31, pp 64–77.


Problems opening files? Get Adobe Reader [External link]