FURTHER STUDIES IN THE COMPOSITIONAL VARIABILITY OF COLOURLESS ROMANO‐BRITISH VESSEL GLASS*

Previous studies of colourless Romano‐British vessel glasses have suggested that, regardless of vessel type, they show considerable compositional homogeneity. Intriguing differences in variability (as opposed to mean composition) have, however, also emerged. This paper reports on a compositional study of 243 vessels, that is larger and more carefully controlled than in previous studies of this kind. Unexpected compositional differences have been found both between and within the four vessel types studied. We discuss the implications of these results in the context of different models that have been proposed for glass‐making and glass‐working in the Roman world.     

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.     

DIRECT EVIDENCE OF PRIMARY GLASS PRODUCTION IN LATE BRONZE AGE AMARNA,EGYPT

This paper presents direct archaeological evidence for the primary production of glass at the LBA site of Tell el‐Amarna, in the form of numerous finds of semi‐finished glass. The diagnostic microstructural and chemical composition of these finds is presented alongside macroscopically similar finds, of shapeless lumps of finished white glass. The semi‐finished glass was found to contain numerous inclusions of residual quartz and newly formed lime‐rich crystal phases, but no added colorant. However, several samples of semi‐finished glass have antimony oxide levels comparable to those of finished white glass, suggesting that the opacifier was added to the initial glass‐making batch. The results are discussed in comparison with the evidence for glass‐making from Qantir – Pi‐Ramesse, identifying some close technical similarities, which indicate that the same technological processes were used at both sites for the production of glass.     

Rationales in Old World Base Glass Compositions

It is long known that most Egyptian and Roman base glass compositions show a remarkably small scatter in their chemical composition. By plotting appropriately reduced base glass compositions in ternary phase diagrams it is demonstrated that the compositional fields defined by the compositional scatter are closely related to eutectic regions within the relevant phase diagrams. This is interpreted as to be due to an eutectic melting regime, i.e. partial melting in the presence of a crystalline buffer or residuum, and not primarily a result of strict recipe and raw material control. Furthermore, it is demonstrated that possibly two independent melting temperature indicators are correlated, suggesting a factual relationship between melting temperature and melt composition. This evidence is taken to develop a “partial batch melting model” for these early glasses, as opposed to the “total batch melting model” of Mediaeval and early modern glasses. Some archaeological implications of this model are briefly discussed.     

STATISTICAL ANALYSIS OF CHEMICAL COMPOSITIONAL DATA AND THE COMPARISON OF ANALYSES

Recent statistical work on approaches to analysing compositional data - where variables sum to a constant for each row of a data matrix - may encounter difficulties when applied to data of the kind typically arising in scientific archaeology. The reason is that results obtained may be unsatisfactory from a substantive viewpoint for identifiable technical reasons. This paper explores and illustrates some possible resolutions of the problem. A feature of the approach used is to analyse subsets of the variables on separate scales. A synthesis of the results obtained from separate analyses is essential and the use of multiple correspondence analysis for this purpose is illustrated.     

APPLICATION OF LEAD ISOTOPE ANALYSIS TO A WIDE RANGE OF LATE BRONZE AGE EGYPTIAN MATERIALS*

Lead isotope analysis was applied to Egyptian materials from the Late Bronze Age in order to investigate the relationship between these different materials, many of which have lead as a significant component. The galena kohls analysed can be provenanced to Gebel Zeit, a large mining site known to have been active during the period. However, the source of lead metal is different and seems to be outside Egypt, along with the source of copper. Lead‐based pigments such as lead antimonate that were used in glass and glazes seem mostly to come from Egypt, although they may well contain a component of ‘Mesopotamian’ lead. In the Predynastic period, galena from many sources is being exploited for use as kohl. However, by the Middle Kingdom (2055–1650 bc ), extraction has concentrated on one source, Gebel Zeit, perhaps reflecting increasingly centralized control and/or the use of large‐scale exploitation. It therefore appears that a complex pattern of trade in lead‐based materials was evident, with lead metal and galena being separate commodities from separate sources and treated as such.     

PANTELLERIAN WARE: A COMPREHENSIVE ARCHAEOMETRIC REVIEW*

Pantellerian ware is a Late Roman cooking ware whose production centre was established on the island of Pantelleria by the pioneering research of Fulford and Peacock almost 20 years ago ( Peacock 1982 ; Fulford and Peacock 1984 ). Archaeological and archaeometric studies carried out by the authors of the present contribution during the past four years have aimed to fully characterize this ceramic class. Recurrent ceramic forms, their distribution over time and space, their petrographic characteristics and their chemical identity, as well as possible raw materials and their technological properties, were considered. The present paper is a comprehensive review of this archaeometric work and aims to establish a ‘reference group’. Using a representative number of samples of Pantellerian ware that were recently discovered in the island through archaeological field surveys or surface and submarine excavations, it was possible to characterize in detail the compositional variability of this ware in terms of chemistry and petrography. Furthermore, the physical properties of this ceramic type have been defined in order to better understand its performance characteristics, mainly in response to induced thermal stress. In the meantime, the experimental mixing and tempering of locally sampled raw materials have shed light on the ancient manufacturing process and have led to an approximation of the original paste.     

PROVENANCE STUDIES OF ROMAN AFRICAN RED SLIP WARE USING NEUTRON ACTIVATION ANALYSIS

African Red Slip ware (ARS) from Carthage and San Sisto Vecchio (Rome) was analysed by neutron activation analysis. The Carthage tableware was linked with Oudna and the type 1 lamps were probably from a northern ARS/lamp factory other than the sources currently investigated. Two samples were central Tunisian, probably manufactured at El Ala. The San Sisto samples divided into three groups and three outliers of undetermined provenance. Group 1 was central Tunisian, probably from El Ala, and was composed entirely of terra sigillata chiara C forms. Group 2 was of unknown origin within northern Tunisia. Group 3 was the largest subset and most of its constituent samples were associated with El Mahrine. Additionally, four samples were linked with Oudna.     

COOKING WARES IN ANCIENT SYRIA (FIRST TO 10TH CENTURIES ad): RECONSTRUCTING THE PRODUCTION CONTEXTS FROM THE CONSUMPTION SITES

This paper deals with Brittle Ware, a cooking ware product that appears to have been highly standardized and widely distributed in Syria during the Roman and the Byzantine periods. The study intends to determine the distribution of Brittle Ware through time and space, by combining typology and a thorough examination of the fabrics in the field, using binocular microscopy (n = 2807 diagnostic fragments). Based on the fabric groupings, petrological and chemical analyses were then undertaken. Resting on an archaeological background and several analytical methods, this paper aims at going beyond a limited provenance study, by characterizing the Brittle Ware production system and thus providing some insights into the economy of ancient Syria.     

ISOTOPIC DISCRIMINANTS BETWEEN LATE BRONZE AGE GLASSES FROM EGYPT AND THE NEAR EAST

This paper presents oxygen, strontium and neodymium isotopic analysis from a series of Late Bronze Age glasses from Egypt and Mesopotamia. It was found that oxygen and neodymium isotopes alone cannot readily distinguish between glasses from the various sites. However, combined Sr and Nd isotope analysis separate the data into three groups: an Egyptian group with relatively low Sr and Nd ratios; a Late Bronze Age (LBA) Nuzi group with high Sr and low Nd ratios; and an intermediate Sr and high Nd ratio grouping of glasses from Tell Brak. These findings suggest that most of the glass from Nuzi and Tell Brak had different raw materials and hence the glass was probably produced at different manufacturing sites. However, one glass ingot found at Tell Brak (TB1) appears to have Nuzi‐type Sr–Nd characteristics. This is the first positive identification of multiple production sites in LBA Mesopotamia and an exceptional example of a glass that may have been exchanged from one LBA site to another.     

A STUDY OF CHINESE PORCELAIN RAW MATERIALS FOR DING,XING, GONGXIAN AND DEHUA WARES*

The raw material used largely determines the type and quality of porcelain produced. Twenty-three raw materials used for making Ding, Xing, Gongxian and Dehua porcelain bodies were studied using principal component analysis. Results show that for Dehua wares, only one raw material was used. For Ding wares, although there were many possible raw materials, the use of one alone is unlikely. Lingshan clay was the main raw material but it has to be mixed with other raw materials: Pinjiawa clay, quartz, feldspar and dolomite. The Xing pieces can be divided into three groups and Gongxian wares into two groups, which are discussed in detail.     

Wavelength‐dispersive X‐ray fluorescence analysis of ancient glasses*

The application of X‐ray fluorescence to the study of the composition of ancient glasses has been limited by the large amount of material required for samples. In this paper the setting of a reduced sampling XRF method for the analysis of ancient glasses is described. The method involves the preparation of glass beads by melting the sample with lithium tetraborate flux. By means of reference and synthetic samples, the method allows regression curves to be set, covering broad ranges from ppm to high concentrations. By means of a single program, up to 31 elements of interest in the study of ancient glass are analysed. The reproducibility, sensitivity and reliability of the method are discussed. The results demonstrate that 0.2 g of glass is sufficient to obtain accurate and sensitive analyses for most of the elements of interest.     

NEUTRON ACTIVATION ANALYSIS OF ROMAN AFRICAN RED SLIP WARE KILNS

Roman African Red Slip ware (ARS) sherds from four ARS kiln sites in modern-day Tunisia (El Mahrine, El Ala, Oudna and Pheradi Maius) have been analysed using instrumental neutron activation analysis. Multivariate statistical methods (including cluster analysis and Manhattan City Block distances), with a dilution correction applied to the data, were used to interpret the concentration data. Reference groups for each of the four kiln sites were established. Both El Mahrine and Oudna were described by two dilution-related groups. These kiln groups were all shown to be compositionally distinct from each other and suitable for use in provenance studies.     

ROMAN GLASS‐MAKING AT COPPERGATE,YORK? ANALYTICAL EVIDENCE FOR THE NATURE OF PRODUCTION*

Ceramic vessels and associated vitreous debris, excavated at Coppergate, York, UK, have been interpreted as the remains of Roman glass‐making from the raw materials. This paper reports the results of analysis of this assemblage by XRF, ICPS, XRD, SEM– EDAX and thin‐section petrography. These findings suggest that some ceramic vessels, used as crucibles, have been subjected to temperatures up to 1200°C, well above the firing temperatures of the local domestic assemblage from which they were selected. Analysis of quartz‐rich debris, intimately mixed with glass in some samples and in others interleaved with glassy phases, indicates partially reacted glass‐making raw materials, with α‐quartz, tridymite and cristoballite phases represented. This could represent evidence either of a failed attempt to frit the raw materials, or a batch that had not fully fused. Variability in the composition of glass at the site is viewed in the light of glass‐making technology, and possible interpretations concerning this episode are discussed in the light of the archaeological evidence.     

LOG‐RATIO COMPOSITIONAL DATA ANALYSIS IN ARCHAEOMETRY*

Compositional data arise commonly in archaeometry, in the study of artefact compositions where the variables measured either sum to 100%, or can be viewed as a subset of such a set of variables. There has been debate in Archaeometry about the appropriate way to analyse such data statistically, which amounts to argument about how the data should be transformed prior to statistical analysis. This paper reviews aspects of the debate and illustrates, using both simulated and real data, that what has been proposed as the ‘correct’ theoretical approach—log‐ratio analysis—does not always work well. The reasons for this are discussed.     

ROMAN WINDOW GLASS: A COMPARISON OF FINDINGS FROM THREE DIFFERENT ITALIAN SITES

Thirty‐three samples of window glass and five glass lumps coming from three Italian archaeological sites—the Suasa excavations (Ancona, settled from the third century bc to the fifth to sixth centuries ad ), the Roman town of Mevaniola (Forlì‐Cesena, settled from the Imperial Age up to the fourth century ad ) and Theodoric's Villa of Galeata (Forlì‐Cesena, settled from the sixth century ad onwards)—were analysed to track the changes in the chemical composition and manufacturing technology of window glass through the centuries. The aims of this study were: (1) to establish the origin of the raw materials; (2) to verify the chemical homogeneity among samples coming from different sites and/or produced using different techniques; and (3) to sort the samples into the compositional groups of ancient glass. The analysis of all the chemical variables allowed two groups to be distinguished: (a) finds from Mevaniola and Suasa; and (b) finds from Galeata. All the samples had a silica–soda–lime composition, but the analysis of minor elements—in particular, of Fe, Mn, and Ti—made it possible to split the samples into two groups, with the higher levels of these elements always found in the Galeata samples (HIMT glass). In conclusion, it can be asserted that the main differences between the samples are related to their chronology.     

ANALYSIS OF FIRST MILLENNIUM bc GLASS VESSELS AND BEADS FROM THE PICHVNARI NECROPOLIS,GEORGIA*

The Pichvnari necropolis on the Black Sea coast of Georgia lies in an area known in the late first millennium as ‘Colchis’, on part of the trade route leading to the Orient. The burials of the necropolis date to the late fifth century bc and frequently contain grave goods, including extremely well‐preserved polychrome glass beads and core‐formed vessels. This paper presents a study of these vessels both stylistically and archaeologically and using SEM–WDS and LA–ICPMS. It reveals that the vessels have compositional differences that may point to multiple manufacturing sites. One of the vessels appears stylistically unique and may exhibit one of the earliest uses of manganese as a decolorizer. Major and minor element data for the vessels suggest that they may belong to the same ‘Levantine’ group as many Roman glass objects, suggesting that a source of sand on the coast of the Levant could have been used in their production. The beads clearly show glass with both natron‐ and plant ash‐based flux with distinct rare earth compositions, showing multiple sites of production, some of which were probably either in the Middle East or the Indian subcontinent.     

REGIONAL VARIATIONS IN BLOOMERY SMELTING SLAG OF THE IRON AGE AND ROMANO‐BRITISH PERIODS*

This study highlights regional variation in the composition of iron‐smelting slag produced in England prior to the medieval period and attempts to link slag composition to the type of ore smelted. For many sites, the slag compositions were consistent with the use of limonite ore, but there is evidence that siderite ore was smelted at sites in Sussex in the late Iron Age/Romano‐British periods. A compositional comparison of smelting slags and slag inclusions in Iron Age currency bars, using data from Hedges and Salter (1979 ), illustrates the potential of smelting slag compositional data in provenance studies of early iron objects.