PRODUCTION TECHNOLOGY OF ROMAN LEAD‐GLAZED POTTERY AND ITS CONTINUANCE INTO LATE ANTIQUITY

A broad selection of Roman lead‐glazed pottery dating from the first century ad through the fifth century ad was studied to establish locations of workshops and to address their technology of production. The ceramic bodies were analysed by ICP–AES. In addition, lead isotope analysis was undertaken on a selection of glazes. These findings suggested that there were several regions responsible for the production of lead‐glazed ceramics in the western Roman world, including central Gaul, Italy and, probably, Serbia and Romania. Using the body compositions as a starting point, the glazing techniques employed by each of the potential workshops were examined using electron probe microanalysis. It was determined that there were two primary methods of glazing. The first method used lead oxide by itself applied to non‐calcareous clay bodies, and the second method used a lead oxide‐plus‐quartz mixture applied to calcareous clay bodies. Based on these data for clay composition and glazing method, transfer of technology from the Hellenistic east to the western Roman world was proposed. Likewise, the inheritance of lead‐glazing technology into late antiquity was established by making comparisons to lead‐glazed ceramics dating to the seventh to ninth centuries from Italy, the Byzantine world and Tang Dynasty China.     

INVESTIGATING MATERIAL DECAY OF HISTORIC BUILDINGS USING VISUAL ANALYTICS WITH MULTI‐TEMPORAL INFRARED THERMOGRAPHIC DATA

This paper shows how visual analytics methodology can be used to facilitate interpretation of multi‐temporal thermographic imagery for the purpose of restoration of cultural heritage. We explore thermographic data in a visual environment from the unifying spatio‐temporal perspective in an attempt to identify spatial and spatio‐temporal patterns that could provide information about the structure and the level of decay of the material, and the presence of other physical phenomena in the wall. The approach is tested on a thermographic dataset captured on the façade of a Romanesque building from the 13th century—the Cathedral in Matera (Italy).     

NON‐DESTRUCTIVE INVESTIGATION OF BRONZE ARTEFACTS FROM THE MARCHES NATIONAL MUSEUM OF ARCHAEOLOGY USING NEUTRON DIFFRACTION*

Non‐destructive neutron diffraction techniques were applied to determine composition and microstructure data, and hence to derive information on manufacturing techniques of Picenum bronze artefacts. Furthermore, texture analyses were carried out on standards and suitable ancient artefacts to investigate their potential role in archaeometallurgical studies. This is a first step towards an overall characterization of the bronze collection of the Marches National Museum of Archaeology, Italy, and the analytical demonstration of relations with and differences compared to Etruscan metallurgy. The work is also an example application of the powerful non‐destructive archaeometrical approaches provided by time‐of‐flight neutron diffraction.     

STATISTICAL ASPECTS OF POTTERY QUANTIFICATION FOR THE DATING OF SOME ARCHAEOLOGICAL CONTEXTS*

This paper describes some statistical analyses of a particular archaeological material (pottery) originating at some sites in the city of Tours. An important part of the archaeological study of pottery is the comparison of ceramic assemblages to establish the absolute dates of contexts. In this paper, a statistical model is built to assess this comparison. The statistical procedure uses classical tools (correspondence analysis, linear regression and resampling methods) in an iterative scheme. Archaeologists may find in the paper a useful set of known statistical methods, while statisticians can learn a method of ‘arranging’ well‐known techniques. No method is new, but their combination is characteristic of this application.     

COMPARING GLASS COMPOSITIONAL ANALYSES*

In a recently published study of Romano‐British colourless glass compositions, using inductively coupled plasma spectroscopy, 28 glasses from Colchester sampled in a previous study were resampled. This was done deliberately, with a view to examining the repeatability of results from sampling on different occasions. We report on our results here, developing in the process some simple statistical methodology that could be applied in similar situations. The potential for combining analyses undertaken at different times is discussed and illustrated.     

AMPHORAE FROM LYONS: PETROGRAPHIC AND CHEMICAL ARGUMENTS*

This paper briefly presents a method for the interpretation of quantitative petrographic data and chemical composition data from amphorae found in Lyons with the aim of determining whether or not the amphorae are of local production. We will try to define the exact place of each argument in our reasoning about the attribution of the origin of these ceramics, while insisting on the zones of uncertainty for each method. It is important to emphasize here that in assigning the origin of the Lyons amphorae all the arguments must be used conjointly. The presence of rhyolite in the temper is not enough to determine the Lyons origin of an amphora, and the geochemical arguments are often insufficient to make a decision. It is a reasoned application of petrographic, geochemical and archaeological arguments which enables us to pronounce an attribution.     

TRACE ELEMENT GEOCHEMISTRY OF NATIVE COPPER FROM THE WHITE PINE MINE,MICHIGAN (USA): IMPLICATIONS FOR SOURCING ARTEFACTS*

Eighty-five samples of native copper from the White Pine mine in upper Michigan were analysed by neutron activation to document geochemical variability within a single locality. Silver, sodium, antimony and scandium contents vary by two orders of magnitude, whereas arsenic values vary by three orders of magnitude. This variability reinforces the need for provenance studies of artefacts to examine sufficient samples to test for heterogeneity within the source area. The White Pine suite contains some of the most impure native copper that has been analysed to date. Arsenic and silver values of these impure samples overlap with those of European coppers, but consideration of several elements can still distinguish native copper from European copper.     

THE SCIENTIFIC IDENTIFICATION OF ARCHAEOLOGICAL JET-LIKE ARTEFACTS

Jet has been widely used in the past for jewellery but was just one of a range of black lithic materials employed. When worked into small objects these materials can be visually indistinguishable, yet discrimination is important because of the restricted occurrence of jet. X-ray fluorescence (XRF) is a rapid non-destructive discriminatory technique, but not all samples are clearly classified, and inhomogeneity may pose problems. A reassessment of the use of XRF is reported. Other techniques considered include electron spin resonance, infrared spectroscopy, Raman spectroscopy, thermogravimetric analysis, scanning electron microscopy and X-radiography. These were tested on geological materials to define an effective methodology for archaeological material. The recommended approach is a combination of XRF and X-radiography, although preliminary Fourier transform infrared results showed promise. The X-ray techniques were tested on Romano-British assemblages from Catterick and Stanwick.     

MULTIVARIATE ANALYSIS OF COMPOSITIONAL DATA: APPLIED COMPARISONS FAVOUR STANDARD PRINCIPAL COMPONENTS ANALYSIS OVER AITCHISON'S LOGLINEAR CONTRAST METHOD

There has been debate about whether standard principal components analysis is appropriate for the multivariate analysis of compositional data (e.g. oxide composition of glass), Loglinear transformation has been recommended by Aitchison as a prerequisite. This paper argues that previous comparisons of methodological merits have tended to circularity of argument by making assumptions about the form of a good multivariate result. To break the circularity of argument the authors have introduced randomized variables into five data sets. A good result must recognize these randomized variables as noise and place them near the centroid of the principal components scattergram of variable loadings. Standard principal components analysis is found to perform better than loglinear transformation in its ability to recognize the randomized variables. It is concluded that loglinear transformation tends to introduce spurious structure into a table of compositional data. This paper is followed by a comment by M. J. Baxter.