aesthetic aspect or technical information contributed by the original wood
surface has already been eliminated. Therefore, the decision to remove a
small amount of material that had never been visible from the exterior, in
a process that could greatly facilitate the aesthetic improvement of the
painted surface as well as the future stability of the panel, can be justified.
If, to take a hypothetical example, a panel were to lose both its back sur-
face as well as its painted surface, would the core material retain any value
as a work of art or even as a historical document?
It should be stressed that the removal of original material—even
that which was never visible on the surface—remains a radical decision
and should not be undertaken as a matter of course. The fitting of these
wedges is a dangerous operation and, unless it is very precisely executed,
offers little advantage over simpler methods of regluing. When executed
precisely, however, it produces a repair of exceptional stability and durabil-
ity while allowing uniform, uninterrupted expansion and contraction
across the panel. It also permits extremely accurate surface-level and cur-
vature adjustments with minimal aesthetic compromise.
An interesting solution for Botticelli’s Man with a Medallionin the
Uffizi Gallery was recently presented by Ezio Buzzegoli and Marco Marchi
ofthe Soprintendenza per i Beni Artistici e Storici di Firenze e Pistoia
(Buzzegoli, Marchi, and Scudieri 1993). In this example, the panel retains
its original surface, but a split traveling upward from the bottom was caus-
ing the pastigliamedallion held by the sitter in the picture to fracture.
With a scalpel, conservators removed the “skin” of wood around the split
in one continuous piece, fitted the split with wedges, and reattached the
skin so as not to disturb the overall aesthetic.Attachment of the two panels
After all the splits in the Metropolitan panel werefitted with wedges,
thereby improving the surface leveling and overall curvature, the two pan-
els were then aligned (Fig. 26). The top of the Metropolitan panel already
had a routed track from the attachment of the old addition. It was decided
to rout a similar track into the mahogany of the Washington panel with-
out cutting into the original poplar (Fig. 27). Short poplar blocks were
then made to bridge the two panels. The blocks were fitted and glued a
few at a time, beginning at the center, so the paint surface could be care-
fully leveled as each piece was glued in place (Fig. 28). Before each piece
was glued, the end-grain joint between the panels was filled from the back
with gesso to produce a tighter fit. Each fixed piece provided a point of
leverage from which to level the next piece—and so on, until the track was
completed. Rabbit-skin glue thickened with calcium carbonate was used
again as an adhesive.
The sides of the Washington panel were wider than those of the
Metropolitan panel because of the additions that were made around its
perimeter (Figs. 25, 29). No additions had been made to the Metropolitan
panel because, although the same amount of wood was missing around
its perimeter, as long as the panel existed as a separate entity, there was
no pressing need to reconstruct it. Besides, it would eventually only be
cropped by the frame. Other evidence (the raised lip or barb at the edges
of the painted surface) made it clear what was missing, and this was
considered sufficient. The decision to add the missing wood in the354 Bisacca
Figure 26
The Washington panel (left) and the
Metropolitan panel, aligned in position for
permanent rejoining.
Figure 27
Planing ofthe track between the two panels.
Note the gesso filling added into the slightly
irregular joint at the center.