New data of 87Sr/86Sr ratio in classical marble: an initial database for marble provenance determination

⁸⁷Sr/⁸⁶Sr ratios are presented for white marbles collected from some of the most famous classical quarry areas of the Mediterranean: Carrara, Paros, Naxos, Pentelikon, Dokimeion, Hymettus, Thasos, and Proconnesus. These ratios range from 0.7071 to 0.7092. The ranges of the different quarry areas are notably superimposed; this prevents a common adoption of this parameter in distinguishing the quarry locations for the classification of the provenance of marbles used during classical times for architectural and sculptural artefacts. However some quarry areas show peculiar distributions of their isotopic values, which could help to assign the provenance to an unknown marble artefact.Adding these new data to the scarce number of strontium isotopic compositions measured on white marbles and available in the relevant literature, permits the creation of an initial database of 11 classical quarry locations. This database confirms that the strontium isotope ratio could be used in marble provenance determination, together with petrographic and geochemical methods, as an ancillary technique.     

INVESTIGATION OF ELECTRON PARAMAGNETIC RESONANCE PEAKS IN SOME POWDERED GREEK WHITE MARBLES*

Within the context of white marble provenance research, 42 powdered white Greek marbles have been characterized using electron paramagnetic resonance (EPR) spectroscopy. Phase analysis was first carried out to separate calcitic and dolomitic marbles, then EPR spectra of Mnuc were studied. Investigation of other marble EPR spectra, especially around g — 2.00, led us to characterize five different signals. The intensities of Mn²⁺ lines permit the separatum of eight quarries into two groups, allowing a first discrimination, in the g = 2,00 region, only one signal has been found to be useful for provenance research objectives. The results are cheering and should lead to many other similar experiments.     

SCIENTIFIC CHARACTERIZATION OF THE POHORJE MARBLES,SLOVENIA

As a former part of the great Roman Empire, Slovenia has many archaeological sites featuring buildings and objects entirely or partly constructed from marble whose provenance is doubtful. In Slovenia, the most probable source of such marble is the Pohorje Mountains. For the purpose of supporting further provenance studies, these marbles are fully scientifically characterized. The techniques used are petrographic and geochemical analysis, stable isotope ratio analysis and EPR spectroscopy. The results show that the Pohorje marbles are highly heterogeneous in both their isotopic and geochemical parameters as well as grain sizes. The parameters of the different Pohorje marble outcrops are compared between themselves and with the parameters of known ancient quarries in the Mediterranean and Austria. The use of a multi‐technique approach with combined parameters allows the best possible discrimination.     

PROVENANCE DETERMINATION OF NEOLITHIC TO CLASSICAL MEDITERRANEAN MARBLES BY STABLE ISOTOPES*

Two stable isotopic data bases for ?¹³C and ?¹⁸O values of marble have been accumulated: (1) 590 analyses from 42 Classical quarries of Greece, western Turkey, Tunisia, and Carrara, Italy, and (2) potential Early Bronze Age (EBA) marble sources in the central Cyclades, comprising 192 analyses from 16 sites on the islands of Paros, Naxos, Ios, and Keros. The data bases include the most important marble occurrences in the eastern Mediterranean and so can be used to source artefacts of any age. Neolithic artefacts found at Franchthi Cave, where no marble occurs locally, have been attributed to the Peloponnese and the Cyclades, direct evidence for early trade. In the Cycladic EBA, Naxos and Keros were the principal marble sources for the abundant artefacts found at Keros. The commercial marble trade in Archaic Greece began in the central Cyclades where the tradition apparently went back some 5000 years, to the Neolithic. Classical Greece and especially Rome expanded the number and locations of commercial marble sources throughout the Mediterranean.     

MARBLE PROVENANCE INVESTIGATION OF ROMAN SARCOPHAGI FROM THESSALONIKI*

The provenance question of the marbles used for making the sarcophagi of local production of Thessaloniki in the Roman period had been posed by archaeologists in the past. The hypothesis was that the material came mainly from the quarries of Thassos Island. This paper presents the scientific analysis of 23 characteristic samples from selected sarcophagi at the Museum of Thessaloniki, using three different techniques: electron paramagnetic resonance (EPR) spectroscopy, stable isotope analysis and maximum grain size measurements, plus in-situ examination. New combinations of parameters were used in order to resolve fully the provenance of all samples. The results showed that the vast majority of the objects were indeed made in Thassian marble but from three different quarries, namely Vathy/Saliara (dolomitic marble), Aliki (calcitic marble) and Acropolis/Phanari (calcitic marble). The calcitic marbles from Aliki and to a lesser degree from Acropolis/Phanari are met in all the basic types of sarcophagi (framed type, plain type, garland sarcophagi). The use of the better quality and snow-white dolomitic marble from Vathy/Saliara quarries is relatively restricted and is often found in works that are above the average regarding the quality of their workmanship. In addition, two sarcophagi were made in Pentlelic marble and one in Proconnesian. The scientific analysis results provide a good confirmation of the archaeological estimates and, furthermore, the conclusions constitute henceforth a secure basis for the study of the entire sculptured works of Thessaloniki.     

CARBON AND OXYGEN ISOTOPIC RATIOS: A DATA BASE FOR CLASSICAL GREEK AND ROMAN MARBLE

An isotopic data base of δ¹³C and δ¹⁸O analyses of 528 samples collected from 39 classical Greek and Roman quarries of Turkey, Greece, Italy and Tunisia has been accumulated. The discriminative possibilities of the data base vary from quarry to quarry. Most quarries have a distinctive pattern for either or both variables, but some have more than one grouping. The data also overlap for many quarries preventing a unique provenance assignment. If the alternatives are known so that the total pool of possible sources is reduced, the data base can then discriminate and assign a provenance for most classical marble artifacts.     

AN EPR AND ISOTOPIC STUDY OF THE MARBLES OF THE TRAJAN's ARCH AT ANCONA: AN EXAMPLE OF ALLEGED HYMETTIAN PROVENANCE*

The provenance of 20 marble samples drawn from the Trajan's arch at Ancona, which is supposed to be made of Greek, Hymettian marble, has been established on the basis of the independent use of EPR and isotopic data. The results of the two methods are in good agreement and unequivocally indicate a Proconnesian provenance. Sixteen samples are assigned to this site, whereas two are classified as untypical Proconnesian samples and the remaining two are assigned an unlikely Parian provenance. No indication exists for a possible Hymettian origin of the marbles. A multi‐method approach, which employs all of the experimental variables simultaneously, demonstrates that the 18 samples identified as Proconnesian belong to several different quarrying areas within the site. This result is taken as an indication that the Proconnesian quarries were run following a highly organized, semi‐industrial production model. The origin of the Hymettian/Proconnesian misunderstanding and the possible misclassification of other Roman monuments in Italy are briefly commented on.     

The marbles of two early Christian churches at Latrun (Cyrenaica,Libya)

The provenance of twenty-seven architectural marbles from the two 6th century AD Christian churches of Latrun (Cyrenaica, Libya) has been established by discriminant function analysis of twelve isotopic, EPR and petrographic/morphological variables. Sixteen samples (59%) originate from Proconnesos, while 8 samples (30%) are assigned to the calcitic quarries of Thasos and 1 to the Docimium quarries near Afyon. The last two samples are of uncertain provenance and might be Proconnesian as well. These results demonstrate that the suggestion made long ago by Ward-Perkins that the presence on non-Proconnesian, possibly Thasian, marbles at Latrun was well founded. Particularly interesting is the strong correlation found between provenance and type of the architectural element sampled. Thasian marbles at Latrun were used exclusively for the manufacture of chancel screens. Seven out of nine screens sampled have this provenance, suggesting that the use of different materials is related to issues of work organisation and specialisation in different marble production centres.     

Roman lava quarries in the Eifel region (Germany): geochemical data for millstone provenance studies

Roman millstones of assumed Eifel origin were produced and exported in huge quantities to many parts of Roman Europe and can be used as tracers for trade patterns in Roman times. This study presents for the first time a raw-material centred geochemical definition of the 13 well-known Roman basaltic lava quarries in the Quaternary East and West Eifel volcanic fields. The aim is to find a way to characterize the individual exploitation sites precisely by means of their geochemical composition and to recommend a standard procedure for further millstone provenance analysis. Based on major and trace element composition analyses by XRF, a differentiation of each quarry is achieved by using a combination of geochemical discrimination and cluster and discriminant analysis. Therefore, for provenance analysis of a unknown millstone sample the following procedure is recommended: (1) geochemical discrimination to check if it actually belongs to the volcanic Eifel quarries; (2) cluster analysis to check to which quarry or cluster of quarries the millstone can be affiliated; and (3) discriminant analysis to assign the sample to its quarry of origin with a certain confidence.     

The Asiatic marbles of the Hadrian's Villa at Tivoli

Multi-method provenance studies, including petrographic, isotopic, electron paramagnetic resonance and trace chemical analyses, have been carried out on 20 white, 9 black and 1 red artifacts purposely selected to investigate the use and distribution of sculptural marbles at the Hadrian's Villa. A large fraction of the marbles tested (21 samples, 70%) are shown to be from Asia Minor, mostly originating from the recently discovered site of Göktepe near Aphrodisias (15 or 71%). All the 9 black samples investigated and 6 out of 11 white Asiatic marbles are from Göktepe, the remaining being Docimium marble from Iscehisar (4 samples) and Aphrodisias marble from the city quarries (1 sample). The single red sculpture tested proved to be Carian red marble from the Iasos quarries, whereas non-Asiatic marbles include 3 Carrara and 6 Pentelicon samples. The selection of marbles tested is preliminary and incomplete, but, despite this, the results are noteworthy, especially since the marble of other sculptures from the Villa has already been identified as Göktepe. Together with other published results, the marble distribution at the Hadrian's Villa seems to suggest that the use of sculptural marbles in the Roman world underwent considerable changes around the turn of the 1st and the 2nd century AD. The evidence supporting this hypothesis is briefly summarized in the conclusions.     

The marble of the David of Michelangelo: A multi-method analysis of provenance

The origin of the marble block utilised by Michelangelo for sculpting the David has been determined with the aid of a multi-method provenancing procedure including spectroscopic (EPR), isotopic and other miscellaneous variables. Data processing was carried out using discriminant function analysis. The provenancing procedure was split into three sequential steps aimed at obtaining increasing spatial resolution. High values of relevant probabilistic parameters indicate that the David's marble originates from Carrara and, specifically, from the Fantiscritti quarries (Miseglia), thus confirming the assignment proposed in the 19th century on the basis of simple autoptic examination. The role of different provenancing techniques in the three stages of the analysis is briefly discussed, as is the performance of instrumental methods as compared with the ability of a trained human eye.     

PETROGRAPHIC AND ISOTOPIC CHARACTERIZATION OF THE MARBLE OF THE ISLAND OF TINOS (GREECE)*

Most outcrops of good‐quality crystalline marble in Mediterranean countries were exploited in Greek and Roman times. Hence, a wide variety of marble is now found in archaeological excavations. The precise determination of the provenance of a marble object is of great archaeological importance, and is now mostly obtained by coupling the petrographic study of a thin section with the analysis of the C and O stable isotopes, or by stereomicroscopy and EPR. The existing databases are considerable but still insufficient, because our knowledge about and study of ancient Mediterranean quarries remain incomplete. The contribution of this research is to add new petrographic and isotopic data on small quarries on the Greek island of Tinos that were exploited in antiquity. This marble belongs to the same geological horizon; it is quite pure, sometimes dolomitic and characterized by a low‐T, high‐P metamorphism that produced a limited recrystallization (MGS varying from 0.64 to 2.50 mm) on marine limestone protoliths. The petrographic features are quite distinctive: the fabric is strongly lineated and often stressed; and the accessory minerals are ubiquitary quartz, ore minerals, graphite and muscovite. As far as the isotopic data are concerned, δ¹³C_PDB varies from 1.1 to 2.7, and δ¹⁸O_PDB from ?1.7 to ?11.4. Both the features and the data have been compared with those of similar marbles used in antiquity, showing that their combination mirrors the fingerprint of the Tinos marble.     

PROVENANCE OF MARBLES FROM NAXOS BASED ON MICROSTRUCTURAL AND GEOCHEMICAL CHARACTERIZATION

A detailed quantitative microstructural study coupled with cathodoluminescence and geochemical analyses on marbles from Naxos demonstrates that the analysis of microstructures is the most sensitive method to define the origin of marbles within, and between, different regions. Microstructure examination can only be used as an accurate provenance tool if a correction for the second‐phase content is considered. If second phases are not considered, a large spread of different microstructures occurs within sample sites, making a separation between neighbouring outcrops difficult or impossible. Moreover, this study shows that the origin of a marble is defined more precisely if the microstructural observations are coupled with cathodoluminescence data.     

A Multi‐Method Database of the Black and White Marbles of Göktepe (Aphrodisias), Including Isotopic,EPR, Trace and Petrographic Data

A database of the ancient marbles of Göktepe near Aphrodisias is reported, and includes 160 white, black and bichrome samples from 13 quarries and two waste deposits. The quarries are grouped into four districts. Districts 1 and 2, to the north, produced exclusively black or dark grey marbles (nero antico). Districts 3 and 4 produced white statuary marble. Nero antico and a black and white stone were also quarried in district 4, where the black marble surfaces again. The samples were characterized using petrography, isotopic analysis, EPR spectroscopy and trace analysis. The Göktepe white exhibits an unparalleled combination of fine grain, low EPR intensity and high strontium concentration that often allows easy identification. The isotopic values are less distinctive and similar to those for Carrara. In most cases, however, they are tightly grouped and also quite indicative. The black marbles exhibit a much weaker metamorphic overprint. With respect to the white marbles, they show analytical similarities but also notable differences. Therefore, establishing their provenance can be more difficult and may require the support of petrographic analysis. The results suggest that the Göktepe marbles were among the most prized and widespread statuary marbles used in Roman times, particularly from the early second century ad onwards.     

ELECTRON SPIN RESONANCE STUDY OF WHITE MARBLES FROM THARROS (SARDINIA): A REAPPRAISAL OF THE TECHNIQUE,POSSIBILITIES AND LIMITATIONS*

The characterization of various samples of white marble from the archaeological site of Tharros (Sardinia) is the starting point for the reexamination of the capabilities of electron spin resonance (ESR) spectroscopy as a tool for identifying the quarries of origin of ancient marble artefacts. Existing methods and data bases are employed for the assignment of our samples, but also reviewed to see if a more extensive use of the technique is feasible. It turns out that ESR spectroscopy, coupled with other available methods, may readily provide valuable information, often yielding the final answer to the problem of marble provenance. However, it is also clear that much work has still to be done to standardize the data collection and analysis procedures and to expand the available data base.     

AN UPDATED,MULTI‐METHOD DATABASE OF EPHESOS MARBLES,INCLUDING WHITE,GRECO SCRITTO AND BIGIO VARIETIES*

An updated database of Ephesos marbles including 244 samples from 16 quarrying districts is reported. The unknown Greco Scritto quarries near Hasançavuslar and the Bigio marbles from the quarries of Belevi and of Mt Panayir are also presented. The paper, however, is mostly devoted to the identification of the Ephesos white marbles, whereas the problems posed by the Bigio and Greco Scritto varieties are deferred to future work. The samples have been characterized by isotopic analysis, EPR spectroscopy and additional petrographic and optical properties. Owing to the complex geology of the region, the white marbles of Ephesos exhibit rather inhomogeneous properties and are conveniently classified into four marble groups, which can be almost completely discriminated. The problem of distinguishing the Ephesos white marbles from other similar varieties has been dealt with by statistically comparing the Ephesos samples with all the similar, medium‐grained varieties present in the marble database. Although some of the four Ephesos groups can be reliably identified, general solution of the discriminant problem is difficult. Historical archaeological information is needed to reduce the complexity of the problem by eliminating unlikely provenances. In this way relevant problems, such as the discrimination between Ephesos and Proconnesos marbles, can be solved.     

Black limestones used in antiquity: the petrographic,isotopic and EPR database for provenance determination

We compiled a database of some of the most important black limestone quarries used in Roman times to be able to determine the provenance of ancient artifacts. For this purpose, we adopted a multimethod approach using the techniques commonly applied to study the provenance of white marbles: petrographic observations, carbon and oxygen isotope composition, electron paramagnetic resonance (EPR) spectra characteristics. Few black limestones were quarried and traded in ancient times; indeed, the importance of most of these quarries was restricted to the areas in which they were located. In this work, we selected a limited number of quarries that produced material believed to have been exported and used throughout the Mediterranean area; most of the quarries selected are found in northern Tunisia, i.e. the Roman proconsular Africa: Djebel Oust, Djebel Azeiz, Ain al Ksir and (presumably) Thala. The “Nero Chiota” limestone quarried on the Aegean island of Chios (Greece) was also included in the database.     

White marbles in Roman architecture: electron paramagnetic resonance identification and bootstrap assessment of the results

On the basis of a previously reported spectroscopic and petrographic database, a general procedure for determining the provenance of white marbles is presented and used to establish a classification rule for the marbles most used in Roman architecture. The rule, based on a data set including seven groups (Carrara, Naxos, Paros, Pentelicon, Proconnesus and Thasos, calcitic and dolomitic marbles) and 712 samples, uses quadratic discriminant analysis and a set of four spectroscopic and two petrographic variables, after logarithmic transformation. The performance of the rule, obtained by resubstitution, is 93.3%. Validation of this result, carried out using the bootstrap technique, after briefly introducing the method, indicates that the resubstitution bias is 1.4%, with a final unbiased performance of about 92%. The bootstrap result agrees satisfactorily with alternative bias estimates (jackknife, cross-validation). Problems connected with the assignment of sets of unknown samples and the methods used for distinguishing reliable from doubtful assignments are discussed.     

Determining the provenance of black limestone artifacts using petrography,isotopes and EPR techniques: the case of the monument of Bocco

This paper investigates the provenance of the black limestone of the monument known as Bocco, named after the king of Mauretania who presented Silla with this work as a sign of his submission to the power of Rome. A multi-method approach, comprising petrographic observations, carbon and oxygen isotope composition and electron paramagnetic resonance (EPR) spectra characteristics, was used for this purpose. The monument is part of a reconstruction of a rectangular base (approx. 8 × 2 m) which is likely to have served as the base for a bronze statuary group and is currently composed of six blocks of limestone whose exterior is engraved with a fine frieze. The results of the different analyses were compared with the data of a published database of the most important black limestone quarries exploited during Roman times in the Mediterranean area. The petrographic and physico-chemical parameters did not prove to be very effective in determining the provenance of the black limestone of Bocco; the isotopes result even raised some uncertainty as to whether all six blocks came from the same quarry, though this doubt was dispelled by the EPR data and, above all, by the evident petrographic similarity between them. Differences in the isotopic composition of the various blocks may be due to an in situ rock alteration process caused by interaction with an aqueous fluid. It was, however, impossible to determine with certainty the origin of the material of the stone artifact using the database available; it is possible that the black limestone of Bocco came from an ancient quarry that is likely to be located in north western Africa but has not yet been uncovered. An alternative hypothesis, according to which the limestone used for the monument of Bocco came from the ancient quarry of Ain el Ksir, one of the Tunisian quarries present in the database, is also discussed.     

Ancient ‘Black’ Decorative Stones and the Ephesian Origin of Sculptural bigio antico

Isotopic, EPR and petrographic studies aimed at establishing the provenance of sculptural bigio antico are reported, including 18 possible quarry sites and 21 ancient sculptures. The results demonstrate that the Belevi quarry, north‐east of Ephesos, above the more famous white marble quarry, was the major source of this marble. Nineteen artefacts belong to Belevi, whereas the precise origin of the remaining two is unknown, but must probably be found within the same area. The peculiar distribution of δ¹³C values both for Belevi and artefacts was crucial for inferring the correct provenance, subsequently confirmed by additional analytical and petrographic data.