The style of these panels, MM1 and its companion, MM2 (the right wing), which seem to be in very good condition, is close to that of Lorenzo Monaco, with elongated figures, keen facial features, elegant rhythms and attractive silhouettes. There seems also to be some influence in the sharpness of the forms, notably in the draperies, from Starnina, Lorenzo’s contemporary, who died c.1413. It should be stressed that these are works of high quality. A feature worthy of note is the double framing seen in both panels – an open ogival tracery set within and playing against the pointed arch. The present writer can find no comparable arrangement in paintings by any other artist in this period, including the two most inventive, Lorenzo Monaco and Starnina. The painter of these panels displays considerable decorative sense and visual wit in relating the figures to this double framing: note, in the left-hand wing, the way that St Christopher’s palm winds through the tracery at the left to expand in the area between that and the external fame, and the way that the point in the tracery at the right pushes in towards the Child’s head; or, in the right-hand wing, the tight confinement of St Dorothy’s head in the upper field of the ogive.

The panels seem always to have been considered to be the wings of a triptych, but an alternative put forward by Dr Laurence Kanter seems more plausible. He writes:

the eccentric and inventive play with the spatial device of the painted frame mouldings . . . would make much more sense as tabernacle doors than as wings of an altarpiece triptych, and would better explain the coats of arms and ledge on which the figures stand. The stacking of the figures atop one another this way would also make better sense. If this proves to be correct, you are likely not missing a painted central image: they could have flanked a niche with sculpture, or a sacrament tabernacle . . .¹

As Dr Kanter notes, it would be important to examine the backs of these panels for any traces of hinges.

These panels have been attributed to Alvaro Pirez Portoghese, although it is not known by whom. Alvaro is becoming a more widely studied painter but those paintings generally reproduced as his do not seem to match these panels particularly closely in style and they also seem less decisive in their forms and staging and less incisive in definition of forms and clarity of thought. Alvaro rarely shows the decorative elegance of these panels, although there is – as Christopher Daly has pointed out to the present writer – an Annunciation by him,² which shows an elegant penetration of the tooled border by Gabriel’s wing – but in that panel Sienese influence is paramount. The figures in the Mells panels are also more elongated than any found in his work. However, Daly points out that a homeless Virgin and Child in Glory with SS Julian and Francis (Fototeca Zeri, inv. 33491) does show close similarities of physical type to the saints in the Mells panels.

A second candidate, also suggested by Mr Daly, is Scolaio di Giovanni, probably to be identified with the so-called Borgo alla Collina Master.³ Scolaio and Alvaro worked together in the Palazzo Datini in Prato in 1411 and there are considerable similarities in their styles. As far as Scolaio is concerned, the present wings come closest in type and to the saints in the wings of the reconstructed Passagli-Spinelli triptych.⁴ In the right-hand wing of that triptych the play of St Christopher’s palm against the moulding of the picture-field is comparable to that in the left wing of the present pair. But, in the opinion of the present writer, the connection although reasonably close falls short of proof.

A third candidate, put forward by Dr Kanter, is the Pisan painter Battista di Gerio, (documented 1414–18) and he has referred the writer to discussions of Battista by Drs Maria Teresa Filieri and Carl Brandon Strehlke.⁵ The writer would accept – judging from the reproductions provided by these scholars – that Battista too might have executed the Mells panels. But he feels that, if Battista was responsible for them, they must be dated some years earlier than any of the paintings documented as his or attributed to him. Thus, while there are some similarities between the saints in the Mells panels and those flanking the Virgin in Battista’s signed and dated (1418) triptych in Pieve di Camaiore, and a common acknowledgement of Starnina’s types, the resemblances seem insufficiently close for certainty about an attribution to Battista.⁶

In summary, while the stylistic area and approximate date of these impressive panels are clear, the present writer does not think that the issue of their authorship can yet be resolved.