MODIFIED MAHALANOBIS FILTERS FOR GROUPING POTTERY BY CHEMICAL COMPOSITION*

A statistical procedure for grouping pottery in provenance studies by chemical data is presented, which now is routinely in use in our laboratory. It is based on the Mahalanobis filter method and X² -statistics, and can be used for both establishing groups and assigning single sherds to already known groups, thus replacing principal components analysis or cluster analysis and avoiding their problems in grouping pottery. The new method is able to consider correlations, uncertainties of measurement and constant shifts of the data in case of dilution. In particular, considering dilution effects results in both a better assignment to and separation of reference groups and is also equivalent to the compositional data approach, if log-transformed data are used. Other distortions of data (e.g., mixing of clays) can also be considered.     

THE UTILIZATION OF TRACE CHEMICAL COMPOSITION TO CORRELATE BRITISH POST-MEDIEVAL POTTERY WITH EUROPEAN KILN SITE MATERIALS

This study uses X-ray fluorescent analysis to compare the trace chemical composition of seventeen pottery sherds excavated from sites in Britain with the trace analyses of twenty four sherds from continental sources. Results indicate that certain elements, notably zirconium, titanium, rubidium, and strontium show little variation within a particular sherd and are consistent among samples from the same source. Certain other elements appear to show wide variation which is possibly attributable to firing conditions, burial or is connected with the glaze. The comparisons of analyses indicate that nine of the seventeen British sherds can be closely correlated with particular continental sources. Tentative correlation is suggested for five sherds. The remaining three sherds clearly do not come from any of the sources investigated.     

CHEMICAL COMPOSITION OF A BOG BODY ADIPOCERE

Chemical analyses were carried out on adipocere obtained from a bog body recovered from a peat bog at Meenybradden, County Donegal, Ireland. Chromatographic (thin-layer chromatography and gas chromatography) and mass spectrometric analyses, combined with microanalytical chemical transformations, have yielded detailed compositional information. An absence of intact triacylglycerols, diacylglycerols and monoacylglycerols indicates that hydrolysis is complete. Consequently the adipocere is composed mainly of fatty carboxylic acids. The high proportion of palmitic and stearic acids, together with depleted oleic acid content, indicates that extensive reduction and, possibly, β-oxidation have occurred during burial in the peat bog. Hydration of the double-bond in oleic acid has also occurred, as is shown by the formation of 10-hydroxystearic acid. The monoenoic fatty carboxylic acids that are present, composed mainly of C₁₈ and C₁₆ compounds, comprise a mixture of positional isomers. The results are compared and contrasted with those obtained from previous studies of the lipid composition of other adipoceres and cadavers of archaeological interest.     

MOESSBAUER SPECTROSCOPIC,CHEMICAL AND MINERALOGICAL CHARACTERIZATION OF IBERIAN POTTERY

Three types of Iberian pottery from the same excavation and archaeological level have been studied by Moessbauer spectroscopy, atomic absorption spectroscopy, X-ray diffraction, and transmitted-light polarizing microscopy. The results are compatible with a common origin, a firing temperature below 850 °C, and a different fabrication technique for each pottery type.     

COMPOSITION AND ORIGIN OF IAPYGIAN POTTERY FROM ROCA VECCHIA,ITALY

We have studied 35 samples of fragments of ceramic containers dated between the eleventh and the eighth centuries BC, found at the archaeological site of Roca Vecchia, some 20 km from Lecce along the Adriatic (Salentine) coastline. Analysis of the elemental composition of the finds has been performed by means of atomic emission spectroscopy in plasma and the results obtained have been treated with diverse multivariate statistical techniques. Except for some minor differences, the statistical analysis confirms the grouping structures indicated by typological and stylistic evaluations of the samples, in accordance with the validity of the above methods.     

LOCAL POTTERY PROVENIENCE STUDIES: A ROLE FOR CLAY ANALYSIS

This article deals with a problem of pottery provenience in which Neutron Activation Analysis was employed to establish chemical ‘fingerprints’. Because of the special nature of the problem, it was necessary to collect clays systematically to reveal the compositions of pottery that would have been made in the vicinity. By so doing, it was possible to ascertain the local provenience of a range of common household wares. Large numbers of vessels were analysed, making possible a meaningful application of statistics in the evaluation of the data.     

CORRECTIONS FOR DRILL-BIT CONTAMINATION IN SAMPLING ANCIENT POTTERY FOR NEUTRON ACTIVATION ANALYSIS

The routine measurement of tungsten concentrations in pottery sampled with tungsten-carbide drill bits has enabled the measured concentrations of cobalt, tantalum, and lutetium to be corrected for drill-bit contamination. The corrected data, for cobalt at least, are sufficiently accurate that they regain their value as chemical indicators of pottery provenance.     

A NEW METHOD FOR THE DETERMINATION OF THE PROVENANCE OF WHITE MARBLES BY CHEMICAL ANALYSIS OF INCLUSION FLUIDS: THE MARBLES OF THE MAUSOLEUM OF BELEVI/TURKEY*

A new method is presented in this paper to pinpoint the origin of white marbles using the chemical analysis of fluids occurring as micro-inclusions in calcitic or dolomitic marbles. Beside the conventional methods usually applied for this purpose, the analyses of significant anions and cations in fluids extracted from white marbles were used to characterize different types of marbles. Using this new approach it is possible to distinguish different groups of marbles which cannot be separated isotopically. A case study was performed on the mausoleum of Belevi, close to Ephesus in Turkey, and the quarries in the area of the mausoleum. The marbles used for the architecture of the mausoleum cannot be distinguished isotopically but were split into two groups when using fluid inclusion analysis. The corresponding quarry for the high-quality marble (cella walls, columns etc.) turned out to be the Kentli Çiftliği quarry, which also supplied the marbles for the classical Artemision. According to the chemical and fluid data it can be shown, however, that neither the adjacent famous antique quarry of Belevi nor the underground mine of Kusini, some 5 km away from the mausoleum, are the sources for the marbles used for the construction or decoration of the mausoleum.     

ON THE SOURCE OF CLAY USED FOR COLOGNE ROMAN POTTERY

Chemical analyses have been made of Roman wares found in Cologne and of bearded man jugs found both in Cologne and Frechen. The question of the influence of the approximately 1600 year difference in time between the manufacture of the two major groups of wares is addressed. It is concluded that the Roman wares were very probably made of Frechen clay.