ACKNOWLEDGEMENTS

This project was conducted during a two-year Andrew W. Mellon Foundation Fellowship, from 1996 through 1998, in the Objects Conservation Laboratory at the Walters Art Gallery. The author is grateful to the Mellon Foundation for making this research possible. The results of this study were first presented at the Washington Conservation Guild (WCG) meeting on February 5, 1998. I would like to thank Dr. Pamela Vandiver of Museum Support Center of Smithsonian Institution, who heard the presentation and offered her insights. The paper was later presented at the 31st International Symposium on Archaeometry in Budapest, Hungary, on April 30, 1998. I am grateful to Dr. Mike Tite of the Research Laboratory for Archaeology and the History of Art in Oxford for his interest and thoughtful comments.

The author is indebted to Dr. Philip M. Piccoli of the Department of Geology, University of Maryland, for carrying out the SEM analyses, which were a major contribution to this study. The permission to use energy dispersive x-ray fluorescence at the Department of Conservation and Scientific Research at the Freer Gallery of Art, Smithsonian Institution, is gratefully acknowledged. Special thanks are owed to Dr. Steve Harvey of the Institute of Egyptian Art and Archaeology at the University of Memphis for his enthusiasm and curatorial assistance in the initial stage of this project, to Dr. Robert Mason of the Royal Ontario Museum West Asian Department, to Dr. Ian Freestone of the British Museum Department of Scientific Research, to Ms. Patricia Griffin of the Cleveland Museum of Art, and to Dr. Stephen Tong of Corning Incorporated and Dr. Robert Brill of Corning Museum of Glass, New York, for sharing their expertise and for providing specific information and references. Many thanks to scholars who provided access to their collections and valuable information: Dr. Helen Whitehouse, Ashmolean Museum, Oxford; Dr. R. A. Lunsingh Scheurleer, Allard Pierson Museum, Amsterdam; and Ms. Carol Andrews and Dr. A. Jeffrey Spencer, British Museum. The encouragement, support, and guidance of the Objects Conservation staff at the Walters Art Gallery are greatly appreciated.

APPENDIX

APPENDIX

Contributed by Dr. Philip M. Piccoli, Assistant Research Scientist, Department of Geology, University of Maryland, College Park, Md. 20742

Samples were analyzed using a JEOL 840A Electron Probe Microanalyzer (scanning electron microscope with analytical capabilities) at the Center for Microanalysis, University of Maryland. Samples were coated with approximately 300 angstroms of carbon using thermal evaporation techniques prior to analysis. For both EDS and WDS, an accelerating voltage of 15 kv was utilized. Samples were observed using secondary electron (SEI) and back-scattered electron imaging (BEI). BEI can be used to evaluate the relative mean atomic number of grains within a given sample (the higher the mean atomic number, the lighter the grain when viewed using BEI). Regions on some specimens thought to be representative of the sample as a whole were photographed using SEI and/or BEI techniques.

1 1. CONDITIONS EMPLOYED FOR EDS ANALYSES

All EDS data were collected using a standard Be-window EDS detector (Noran) connected to a Tracor Northern 5500 II analyzer. The beam size (analytical area) for EDS analyses was as large as the sample would allow. Data were collected over the energy range of 0–10 kv. A sample current of approximately 1.5 nA (corresponding to an EDS detector dead time of 20–40%) and a count time of 60 seconds were utilized to collect raw intensities. Raw data (count rates) were corrected using an SQ algorithm using top-hat filters to correct for fluctuations in sample current and accelerating voltage. Raw intensities were converted to concentrations using the PRZ correction procedure. The SQ algorithm forces analytical totals to 100%. All data were collected at a working distance of 30 mm. Reconnaissance analyses of the body revealed that sodium (Na), silicon (Si), lead (Pb), chlorine (Cl), potassium (K), calcium (Ca), iron (Fe), copper (Cu), aluminum (Al), magnesium (Mg), and antimony (Sb) were the only elements present in significant quantity to be detected readily by EDS; in addition manganese (Mn) and titanium (Ti) were also explicitly sought in the analyses. Theoretical, stored peaks for all 13 elements were fitted to the acquired spectra to ensure all peaks were accounted for; all chi-square values are of the order of 2 or below, a finding that suggests all major peaks were accounted for. The technique is limited because: (1) elements lighter than Na (such as oxygen [O] and carbon [C]) cannot be detected, (2) water (and hydroxyl) present as an essential structural constituent cannot be detected, and (3) no indication of crystalline or amorphous material can be obtained.

2 2. CONDITIONS EMPLOYED FOR WDS ANALYSES

British Glass Industry Research Association (BGIRA) standard 3 (for Na, Al, Pb, Sn, and Fe), Corning Museum of Glass standard A (for Si, K, Sb, Ca, Mn, and Cu), Meionite Scapolite (for Cl), in addition to cobalt (Co) and arsenic (As) metals were used as standards. Unknowns and standards were analyzed using a sample current of 40nA, a 10 x 10 μm rastered beam, an accelerating voltage of 15 kv, and count times of up to 30 seconds (reduced when counting statistics were better than 0.5%). Peak and background measurements were made on all standards and unknowns. K(alpha)-lines were analyzed for most elements; however, M-lines (Pb) and L-lines (Sn, Sb, and barium [Ba]) were also used. A rigorous WDS protocol was used in order to avoid complex overlaps of Pb and Sn with alkalis and to allow for lower detection limits. Using this protocol, the minimum detection limit (1 sigma) for elements determined by WDS were calculated and are reported in table 3. Data were collected using the software package pcTASK (Geller MicroAnalytical); raw intensities were converted to concentrations using the PRZ correction procedure. In most cases, analytical uncertainty presented in table 3 is considerably less than sample variability (compositional), revealing that samples are heterogeneous. In order to minimize any possible effect of migration of alkalis in the samples, alkalis (Na and K) were analyzed first, a large beam was used, and the beam was rastered.

REFERENCES

Baines, J., and J.Malek. 1980. Atlas of ancient Egypt. Oxford: Phaidon.

Barber, D. J., and I. C.Freestone. 1990. An investigation of the origin of the color of the Lycurgus cup by analytical transmission electron microscopy. Archaeometry32:33–45.

Freestone, I. C.1982. Applications and potential of electron probe micro-analysis in technological and provenance investigations of ancient ceramics. Archaeometry24:99–116.

Freestone, I. C.1998. Personal communication. Department of Scientific Research, British Museum, London.

Freestone, I. C., N. D.Meeks, and A. P.Middleton. 1985. Retention of phosphate in buried ceramics: An electron microbeam approach. Archaeometry27:161–77.

Friedman, F. D, ed. 1998. Gifts of the Nile: Ancient Egyptian faience. New York: Thames and Hudson.

Gabriel, B.1985. SEM: A user's manual for materials science. Metals Park, Ohio: American Society for Metals.

Johnston, W.1999. William and Henry Walters, the reticent collectors. Baltimore: Johns Hopkins University Press.

Kaczmarczyk, A., and R. E. M.Hedges. 1983. Ancient Egyptian faience: An analytical survey of Egyptian faience from predynastic to Roman times. Warminster, England: Aris & Phillips.

Kingery, W. D., and P. B.Vandiver. 1986. Ceramic masterpieces. New York: Free Press.

Mason, R. B., and M. S.Tite. 1997. The beginnings of tin-opacification of pottery glazes. Archaeometry39:41–58.

Mysliwiec, K.1996. In the Ptolemaic workshops of Athribis. Egyptian Archaeology9:34–36.

Nenna, M.-D., and M.Seif el-Din. 1993. La vaisselle en fa�ence du Mus�e Gr�co-Romain D'Alexandrie. Bulletin de Correspondance Hell�nique, II �tudes chroniques et rapports (�cole Fran�aise D'ath�nes)117:565–603.

Nicholson, P.1993. Egyptian faience and glass. Princes Risborough, England: Shire Publication Ltd.

Nicholson, P.1998. Materials and technology. In Gifts of the Nile: Ancient Egyptian faience, ed. F. D.Friedman. New York: Thames and Hudson. 50–64.

Nordyke, J. S.1984. Lead in the world of ceramics. Columbus, Ohio: American Ceramic Society.

Petrie, W. M. F.1911. The pottery kilns at Memphis. In E. B.Knobel, W. W.Midgeley, J. G.Milne, M. A.Murray, and W. M. F.Petrie, Historical studies II. London: School of Archaeology in Egypt. 34–37.

Reeder, E.1988. Hellenistic art in the Walters Art Gallery. Princeton, N.J.: Princeton University Press. 105–6.

Rhodes, D.1973. Clay and glazes for the potter. Radnor, Pa.: Chilton Book Company.

Scheurleer, R.1986. Thirteen drinking cups. In Enthousiasmos: Essays on Greek and related pottery, ed. H. A. G.Brijder, A. A.Drukker, and C. W.Neeft. Amsterdam: Allard Pierson Museum. 147–56.

Scheurleer, R.1998. Personal communication. Allard Pierson Museum, Amsterdam.

Shortland, A. J., and M. S.Tite. 1998. The interdependence of glass and vitreous faience production at Amarna. Ceramics and Civilization8:251–65.

Spencer, A. J., and L.Schofield. 1997. Faience in the ancient Mediterranean world. In Pottery in the making: World ceramic traditions, ed. I.Freestone and D.Gaimster. London: British Museum Press. 104–9.

Tite, M. S., and M.Bimson. 1986. Faience: An investigation of the microstructures associated with the different methods of glazing. Archaeometry28:69–78.

Tite, M. S., I.Freestone, R.Mason, J.Molera, M.Vendrell-Saz, and N.Wood. 1998. Review article: Lead glazes in antiquity—methods of production and reasons for use. Archaeometry40:241–60.

Vandiver, P. B.1982. Technological changes in Egyptian faience. In Archaeological ceramics, ed. J. S.Olin and A. D.Franklin. Washington, D.C.: Smithsonian Institution Press. 167–79.

Vandiver, P. B.1983. Appendix A: The manufacture of faience. In A.Kaczmarczyk and R. E. M.Hedges, Ancient Egyptian faience: An analytical survey of Egyptian faience from predynastic to Roman times. Warminster, England: Aris & Phillips. A1–A144.

Verita, M., R.Basso, M. T.Wypyski, and R. J.Koestler. 1994. X-ray microanalysis of ancient glassy materials: A comparative study of wavelength dispersive and energy dispersive techniques. Archaeometry36:241–51.

Wallis, H.1897. Egyptian ceramic art: The MacGregor collection. London: Taylor and Francis. 81–85.

AUTHOR INFORMATION

YUNHUI MAO received her B.A. in chemistry and studio art from Bowdoin College in 1991 and her M.S. in art conservation science, majoring in objects conservation, from the University of Delaware in 1996. She has completed a graduate internship at the Freer Gallery of Art, Smithsonian Institution, and a two-year postgraduate Mellon Fellowship at the Walters Art Gallery, Baltimore. Currently, she is a visiting objects conservator at the Walters Art Gallery. Address: Walters Art Gallery, 600 N. Charles St., Baltimore, Md. 21201

Received for review April 14, 1999. Revised manuscript received November 2, 1999. Accepted for publication February 15, 2000.

NOTES

1.. MacGregor was a well-known collector of Egyptian antiquities during the 19th century. His collection was sold at Sotheby's in London in 1922. Henry Walters purchased his pieces from a Paris- and New York-based art dealer, Dikran Kelekian, in 1923, a year after the London sale.

2.. Xeroradiography was carried out at the Division of Conservation and Technical Research at the Walters Art Gallery, using the Xerox 126 system for xeroradiography. Xeroradiographs were taken at 150kv, 2.4 mA, and 10 seconds.

3.. Energy dispersive x-ray fluorescence analysis was undertaken using Omega-5, a modified Spectrace 6000, at the Department of Conservation and Scientific Research at the Freer Gallery of Art, Smithsonian Institution. Data were collected at 40 kv, 0.99 mA, and 100 seconds.

4.. Each sample area was first cleaned with ethanol to remove surface dirt. The sample was removed using a jeweler's saw and then was impregnated in a polyester resin. The sample was polished using sandpapers first, followed by various grades of Micro-mesh, up to 6,000. Micro-mesh is made of silicon carbide, thus eliminating the possibility of introducing aluminum as a contaminant to the faience body. The polishing residue trapped in the voids was removed by immersing the sample in an ultrasonic tank containing distilled water for approximately 2–5 seconds, depending on the needs of each sample.