complex paintings,^26 even if separations at glue lines or fissures interrupted
the structural continuity of the panel. They also distributed throughout
the whole panel the forces originating at supports, hanging points, con-
nections, and so on, and helped to conserve the painting by reducing the
negative effects of swelling and shrinkage caused by unavoidable moisture
changes (Uzielli 1994).
On all but the last point, there is general consensus. However, the
question of whether crossbeams and backframes help conserve a painting
is still quite controversial today, both in its theoretical and practical aspects.
In fact, the real problem lies not in the lesser or greater complexity of
the backframe but, rather, with the type and stiffness of the connections
between the panel and crossbeams, as well as the stiffness of the cross-
beams themselves.
Until the early fifteenth century, connections between panels and
frames (including crossbeams and engaged frames, where appropriate)
were made mostly with nails. Later, various types of sliding crossbeams,
resting on the back face of the panel, were devised. In the late fifteenth
and early sixteenth centuries, dovetailed crossbeams (trapezoidal in cross
section), inserted in tapered or (rarely) parallel grooves mortised into the
thickness of the panel, were often used. On a few occasions, crossbeams
were glued.
The two criteria taken into consideration by artisans, along with
their individual views and experience, in order to obtain the “controlled
mobility” required from the crossbeams were, first, the distance between
the crossbeams—defining the size of the transversal strip of panel that was
“entrusted” to a crossbeam (which in fact was highly variable), and second,
the thickness ratio (i.e., ratio of the thickness of the panel to thickness of
the crossbeam), with an approximate range of 1:2–1:3.
It should be noted, however, that any general statement regard-
ing the design of the backframe may do a disservice to the creativity and
ability of the artisans. The few images given here serve only as examples
(Figs. 15–17).Nailed crossbeams
Nailing is one of the oldest and most frequently used means of connecting
pieces ofwood. At least with regard tothirteenth- and fourteenth-century
panels, nailing should not be considered primitive, rough, or technologi-
cally inadequate. On the contrary, a careful analysis shows just how wise
and skillful the artisans were who conceived the structures and nailed
them together.^27
Nails were made of soft, wrought iron. The shanks were square
or rectangular in cross section, tapered from their large, thin, round heads
to their acuminate points (Fig. 18). They were driven by hammer into par-
tially prebored holes and were clinched back into the wood in a U shape to
ensure optimum resistance against pullout.^28
The spacing of nails was regular and obviously well thought out.
No strict spacing rules applied; the artisan’s wisdom defined the direction
in which the nails were inserted (from the front toward the back or vice
versa, or in both directions) (Fig. 19).^29
Great care was usually taken in separating the nail’s end (head or
clinched point) from the ground layer (Fig. 20) to prevent repercussions on
the paint layers, such as surface irregularities or possible future emergence122 Uzielli
Figure 15
Coppo di Marcovaldo, Madonna in trono col
Bambino, reverse. Church of Carmine,
Florence.