5 DISCUSSION

On the basis of tin oxide enrichment in the patina, the presence of mottled areas containing enhanced levels of iron, and the pseudomorphosis of structural details, a hypothesis can be formulated, based on the research by Geilmann. This hypothesis would state that one possible environment for the corrosion observed on these bronzes would have been burial in a porous soil, with both oxygen and carbon dioxide readily available but with no chloride ion corrosion. The corrosion products and pustules should show no evidence of chloride corrosion, nor should the pustules be associated with cuprous chloride. This hypothesis is confirmed by the analytical and microstructural studies carried out here. That the corrosion environment was not deficient in carbon dioxide can be seen from the extensive malachite deposits on the surface, the conversion of the lead globules to lead carbonates, and the presence on the Nike and Venus of azurite crystals. In the case of the Venus, these crystals can be seen growing on individual malachite fibers as tiny deep blue compact crystals. The outer surfaces of the corrosion pustule on the Togati are also malachite rather than basic copper chlorides.

The associated fragments of red brick or tile embedded in the corrosion products of all four bronzes not only link them together but indicate that the environmental conditions were favorable to the permeation of groundwaters and to oxygenated soil conditions and also allowed enough space adjacent to some of the object surfaces for the fibrous malachite crystals to grow. Unlike warty corrosion associated with chloride ions, the kind of warts seen on the surface of these bronzes are chemically stable formations under proper storage or exhibition conditions and do not require especially strict humidity control in display or storage. They had, for example, been kept in a special case with regulation of the humidity to below 40% RH, but since there is no chloride instability, such measures are unnecessary. Similarly, during conservation treatment there is no need to remove the warty corrosion, unless there are good aesthetic arguments for doing so.

These four bronzes show an interesting variety of corrosion phenomena, ranging from fibrous malachite to tin oxide patina, and their surfaces preserve details of the “original surface” very well in a matte tin oxide patina. The bronzes were mechanically cleaned during conservation, revealing this fine detail, but they were also treated with a respect for the preservation of corrosion and associated material overlying the fine patina. It is easy, with hindsight, to criticize earlier methods for the conservation of archaeological bronzes, such as electrolytic cleaning, glass-bead air-abrasion techniques, use of chemical stripping reagents or removal of outer layers of the patina, sometimes to reveal a cuprite surface. An analogy is frequently made to medicine, where earlier treatment techniques have been discarded. The only difficulty is in gauging the health of the patient. We know that we overtreated the patient in the past, but there is still a danger that by cleaning and removal of corrosion, we are losing precious clues to association, burial, and environment. Practices in this respect are not uniform throughout the conservation profession. In any event, a variety of treatment options are available to the conservator, depending on the object; the advantages and disadvantages of cleaning have to be evaluated carefully. This paper could not have been written if the bronzes had been cleaned of all extraneous corrosion products.

ACKNOWLEDGEMENTS

Thanks are due to Bruce Christman, head, Conservation Department, Cleveland Museum of Art, for allowing Lisbet Thoresen, associate conservator, Department of Antiquities Conservation, J. Paul Getty Museum, and me to study and sample the Cleveland Nike; to Eric Doehne for carrying out the many microprobe analyses; to WestCoast Analytical Services for performing the ICP-MS analyses; to Jerry Podany, head, Department of Antiquities Conservation, J. Paul Getty Museum, for commenting on the manuscript; and to Karen Manchester, Antiquities Curatorial Department, J. Paul Getty Museum. Special thanks are due to Eva Sander, conservator, who assisted in preparing an English translation of the paper by Geilmann in 1986.