SAS Bulletin
Volume 19 Number 1/2 January /June 1996

Book Reviews

Lost Scents: Investigations of Corinthian "Plastic" Vases by Gas Chromatography-Mass Spectrometry. William R. Biers, Klaus O. Gerhardt, & Rebecca A. Braniff. MASCA Research Papers in Science and Archaeology 11. University of Pennsylvania Museum of Archaeology and Anthropology, Museum Applied Science Center for Archaeology, Philadelphia, 1994. xi + 72 pp., 16 figs. $30.00 (cloth).

Reviewed by Charles C. Kolb, Division of Preservation and Access, National Endowment for the Humanities, Washington, DC

Ceramics are one of the primary loci for the survival of organic residues of interest to archaeologists. The direct evidence of container use may be ascertained by the analysis of vessel contents, use wear, and external deposition. The identification of the original contents and functions of ancient ceramic cooking and storage vessels, among other types, leads to more finite socioeconomic interpretations. Most research has been devoted to the analysis of food residues rather than the delineation of perfumes and scented oils, the topic of the volume being reviewed.

During the past two decades, scientific analysis to determine the original contents of ceramic containers has become a significant and illuminating research area in archaeology, and frequently involves chemists or biochemists working cooperatively with archaeologists and art historians. Michael Deal (1990) has commented in the Bulletin about the importance of this significant interdisciplinary research and this MASCA volume documents this cooperation. Two of the authors, Biers and Braniff, are from the Department of Art History at the University of Missouri at Columbia, while Gerhardt is affiliated with the Experimental Station Chemical Laboratories-Mass Spectrometry Facility and the Department of Biochemistry at the university.

This slim volume documents the results of original research conducted at the University of Missouri on a typologically unique but important vessel category. In 21 pages of text accompanied by 21 pages listing chemical compounds, the authors propose new nondestructive extraction procedures for a more precise analysis of vessel contents. Although this research was not conducted at MASCA, it is appropriate that this cutting edge methodology is published in this important series, which includes other major works on vessel content analysis (Biers and McGovern 1990). (MASCA's facility and publications were recently profiled in an issue of the Bulletin [Miller 1993].)

"Plastic" vases are small containers molded in the shape of animals (such as hares and rams), mythological creatures (gorgons), and human busts or body parts (feet), which were distributed throughout the Mediterranean region (Italy and Sicily to Asia Minor and North Africa) during the seventh and sixth centuries B.C.E. The vessels derive their name from the Greek word plastos meaning molded or formed, although components may be wheelmade, handformed, or created in clay molds. Their small size, narrow necks, and tiny mouths are similar to the aryballos and the alabastron. Typically the containers were decorated in black gloss, but specific groups also had polychrome or incised detail.

The authors review briefly the chief centers of manufacture: Corinth, East Greece (Rhodes), and Western (southern Italy-Sicily). They follow Sparkes' (1991) recent study in defining ceramic shapes. In the past, scholars have assumed on the basis of archaeological contexts and vessel configurations that these containers held ancient precious perfumes or scented oils that needed to be dispensed frugally. The earliest literary works which cite perfume recipes and manufacturing processes date to the fourth and first centuries, and are summarized (Theophrastus Historia Plantarum, de Ordoribus IV.5.2; Dioscordies De Materia Medica I.56; and Pliny the Elder Naturalis Historia XIII.2.5). The authors also discuss previous interpretive scholarship regarding vessel functions, including religious, funerary, chthonic, folkloric, and picturesque.

Interestingly, plastic vases are not depicted in Greek art nor are they referred to by any ancient literary sources. In the past, interpretations depended nearly exclusively upon a knowledge of archaeological contexts and decorative parallels in vase painting. Although East Greek plastic vases have been studied using optical emission spectroscopy (Jones 1986), no vessels from Corinthian or other manufactories have been examined by physicochemical techniques. However, older techniques looked specifically for one product (such as olive oil) and ignored the existence of other substances.

Therefore, the authors sought to identify vessel contents more precisely in order to discern specific vessel function and clarify iconography. The methodology and initial results of a pilot project (Biers et al. 1988; Gerhardt et al. 1990) were based on a sample of only four vases. The goal of the current research is to develop and test a nondestructive method of analysis and to collect and examine as many examples of Corinthian plastic vases as possible. The present report, which expands and refines these earlier techniques and results, employs 24 vessels (20 Corinthian, 3 East Greek, and 1 Cretan), and identifies a range of substances. Vessels were obtained from more than nine collections, with the J. Paul Getty Museum and the Museum of Art and Archaeology at the University of Missouri each providing seven.

For fats and oils the primary methods of analysis include Fourier transformation spectrometry, infrared absorption, mass spectrometry, gas chromatography, and proton magnetic resonance spectrometry, among others, which identify specific fatty acids, cholesterol, triglycerides, and other components of organic materials. However, gas chromatography is especially well suited to identify organic materials such as oils and resins. Characterizing a wide range of natural products, capillary gas chromatography (GC) which is very sensitive and highly specific to the analysis of lipid mixtures affords the separation of a complex compounds into its components, which are then discerned by interfaced mass spectrometry (MS). Linked GC-MS holds considerable promise for use in the studies of vessel function, diet, and distribution of unique foodstuffs and other products such as dyes and perfumes.

Ideally, substances should be measured carefully both in the container and the context in which the vessel was found in order to minimize the effects of the postdepositional environment. An examination of substances from the interior and the exterior vessel walls, mouth, and base may also assist in determining container reuse and depositional contamination. None of the specimens examined had any visible, measurable remains of oils or other substances, although the specimen from the Toledo Museum of Art still contained soil which was analyzed separately and had chemical compounds similar to those found in the vessel's fabric.

The current study employs two separate extraction solvents (methylene chloride and methanol) to penetrate the ceramic pores and remove samples for analysis. Instrumentation includes a Kratos MS25 double-focusing mass spectrometer equipped with a Carlo-Erba 4160 Fractovap gas chromatograph and a DS-55 data system operating on a Data General Nova 4X computer. Specimen patterns were compared via library search routines to the approximately 112,300 compounds included in the major databases (NIH/EPA and NIST Computer Libraries, Wiley/NBS Registry of Mass Spectral Data, and Eight-Peak Index of Mass Spectra). A BenchTop PC/Probability-based Matching System was also used.

The research determined that nondestructive GC-MS extraction and analysis can retrieve compounds from the fabric of the ancient vessels. The procedures did not affect the fabric or decoration and no vessel was damaged in any way by the solvents used. Three vessels contained more than 100 identifiable compounds and others contained very few (potentially the result of thorough cleaning following excavation). The number of compounds per vessel ranges from 114 to 18. Modern contaminants (the pesticide DDT, silicone compounds, plasticizers from glues and plastic packaging, caffeine, and nicotine) were notable. Cholesterols, long-chain fatty acids, and over 40 terpenoids were discerned, but in no instance were the authors able to match specifically the contents of a vessel with any recipe extant in the ancient sources. The famous iris perfume could not be identified, but information about the origin and nature of juniper, pine, and cypress resins were delineated.

In sum, the authors suggest that these vessels contained an oil, perhaps olive, scented with a resin (represented by specific fatty acids and terpenoids), and that these contents were not among the very costliest perfumes. In addition, there was no correlation between vessel configuration and contents. While the technique is non-destructive, it is also costly and labor intensive, and subject to modern contamination. The chemical evidence is consistent with pungent (rather than floral) scented contents and the enfleurage method of extracting scents may have been used in Corinth in the sixth century B.C.E. The authors suggest that it is possible that these scents may have served as insect repellent or in embalming. Appendix II suggests protocols for treating intact specimens during excavation to avoid modern contamination.

This is an important study, both for the results reported and the initial interpretations derived, but also for the nondestructive analytical procedures developed and tested. No longer is it necessary to pulverize a potsherd before extraction; therefore the artifact is not destroyed or damaged. The extraction solvents, unlike strong acids or bases used in gas-liquid chromatography (GLC), do not alter the composition of the original fatty acids in the oils nor are aldehydes or ketones, which may be useful "markers," destroyed. In the future it will be possible to determine more precisely the places of origin of the contents of ceramic vessels and elucidate the production and distribution of ancient craft products and foodstuffs.


Biers, W.R., S. Searles & K.O. Gerhardt
1988 Non-destructive extraction studies of Corinthian plastic vases: methods and problems: a preliminary report. In J. Christiansen and T. Melander (eds.), Proceedings of the Third Symposium on Ancient Greek and Related Pottery, Copenhagen, 1987, 33 47. Ny Carlsberg Glyptotek and Thorvaldsens Museum, Copenhagen.

Biers, W.R., & P.E. McGovern (eds.)
1990 Organic Analysis of Ancient Vessels: Materials Analysis and Archaeological Investigations. MASCA Research Papers in Science and Archaeology 7. University of Pennsylvania, University Museum, Philadelphia.
Deal, M.
1990 Exploratory analysis of food residues from prehistoric pottery and other artifacts from eastern Canada. SAS Bulletin 13(1):6 12.

Gerhardt, K.O., S. Searles & W.R. Biers
1990 Corinthian figure vases: non-destructive extraction and gas chromatography-mass spectrometry. In W.R. Biers and P.E. McGovern (eds.), Organic Contents of Ancient Vessels: Materials Analysis and Archaeological Investigations, 41 50. MASCA Research Papers in Science and Archaeology 7. University Museum, University of Pennsylvania.

Jones, R.E.
1986 Greek and Cypriote Pottery: A Review of Scientific Studies. Athens: British School at Athens.

Miller, N.
1993 Laboratory profile: Museum Applied Science Center for Archaeology (MASCA). SAS Bulletin 16(1):2 6.

Sparkes, B.A.
1991 Greek Pottery: An Introduction. Manchester University Press, Manchester and New York.

Back to the Society for Archaeological Sciences