available for working. Second, they perfected the design of
shaft furnaces to allow continuous production of the metal
by making it possible to add new ore and tap off slag without
shutting down the furnace. And third, they developed the
technique of amalgamation, by mixing in a crucible ground,
unpurifi ed gold with mercury; the two combined to form an
amalgam (gold being the only metal attracted to mercury),
leaving behind the impurities. Th e amalgam was then forced
through leather to leave the gold behind or was heated to va-
porize the mercury (with its lower boiling point), which was
then condensed through cooling, the only known ancient ap-
plication of distilling.
Th e Romans practiced and sometimes improved on all
the Greek techniques of casting, forging, and annealing to-
gether with the more delicate metalworking techniques. Ad-
vances in enameling and the production of gold leaf were two
of their principal contributions in the latter category. Ac-
cording to Pliny in his Natural History, Roman metalworkers
could beat a single ounce of gold into 750 leaves of foil, each
4 inches square: a remarkable achievement, even if gold is the
most malleable of metals. Enameling they appear to have bor-
rowed from the Celts of western Europe, who as early as the
fourth century b.c.e. were applying a fi ne powder of ground
glass and metallic oxides mixed with water to the surface of
small bronze objects (brooches seem to have been the most
common) that were then heated until the enamel fused. At-
tractive, multicolored designs were achieved by placing dif-
ferent oxide mixtures into separate cells on the surface of
the metal object, formed in various shapes either by cutting
directly into the metal (called champlevé) or by applying a
mesh of wires over the surface (cloisonné).
For the hollow casting of small metal objects, the Ro-
mans used the same cire-perdue, “lost-wax,” process as the
Greeks. Metalworkers of both cultures were also capable of
producing life-sized bronze statues using the same technique
but by casting many parts individually and then fastening
them together. To ensure that the constituent elements would
fi t properly, the artist started from a full-sized model of the
entire piece in clay or plaster, from which he would take sep-
arate two-piece molds of the individual parts. Th ese molds
would then be coated on the inside with a layer of melted
wax of the same thickness as the fi nal bronze product, and
an inner core of clay would be inserted. Th e two halves of
each mold would be reassembled, the wax melted out, and
the liquid bronze added. Once all the parts had been cast,
they would be welded together by fi rst heating the points
of attachment almost to melting and then applying molten
metal around the joint, causing the two pieces to bond. Once
cooled, the welding on the surface was fi led down to the level
of the cast pieces, fi ne details chiseled out of the metal or
engraved into it, patches added to camoufl age the inevitable
casting fl aws, and the statue given its fi nal polishing. With
this technique, multiple copies could be made of the same
original model.
Although there are documentary claims that the Romans
were casting large bronze statuary as early as Romulus in
the eighth century b.c.e., it seems more likely that it was the
Etruscans who fi rst borrowed from the Greeks the cire-per-
due technique: one of the earliest surviving Italian examples
is the famous Capitoline wolf (without the familiar suckling
twins Romulus and Remus, who were added in the Renais-
sance), dated to around 600 b.c.e. But within a few centuries
the Romans were producing bronze statues of a size and qual-
ity that matched their predecessors’ work, an achievement in
the fi ne casting of metal that was not reproduced until the
Renaissance, when the technique of cire-perdue casting was
reinvented.THE AMERICAS
BY PENNY MORRILL
Th e history of metallurgy in the Americas is based primarily
on the archaeological record. Although most sites have been
looted over the centuries since the arrival of the Spaniards,
several tombs have been opened and excavated in modern
times to reveal magnifi cent works in gold. In the interpre-
tation of these mortuary caches, it is helpful to consider the
cautionary note sounded by the historians Paloma Carcedo
Muro and Izumi Shimada, that these works may not have
been produced exclusively for burial. Th ey have said that gold
artifacts were used in public contexts—courts and temples,
for example—to propagate symbolic messages. Th en, on theFrom classical Greece to the Roman Empire statu-
ary of cast bronze was as popular as sculpted stone,
though proportionately less survives today because
the material demands of postclassical societies made
bronze artistic creations vulnerable to recasting into
more practical tools. Many of the few large bronze
statues that grace Athens’ National Archaeological
Museum or the piazza of the Capitoline Museums in
Rome owe their survival to a combination of ancient
accident and modern luck—the Zeus/Poseidon and
Horse with Jockey now in Athens were recovered from
a shipwreck off the east coast of Greece in 1928—or
to misplaced reverence: in Rome the bronze Marcus
Aurelius on horseback escaped medieval recasting
because he was misidentifi ed as Constantine, the
fi rst Christian emperor in the Church’s view. This is
our sole surviving formal equestrian statue from the
ancient world, among what must have been the thou-
sands to which Pliny alludes when he tells us in his
Natural History that the style was almost certainly de-
veloped by the Greeks and was very popular among
the Romans.THE SURVIVAL OF BRONZE STATUARY
686 metallurgy: The Americas