PROVENANCE STUDIES OF CHALCOLITHIC OBSIDIAN ARTEFACTS FROM NEAR LAKE URMIA, NORTHWESTERN IRAN USING WDXRF ANALYSIS

In 2005–2006 we initiated a major archaeological survey and chemical characterization study to investigate the long-term use of obsidian along the eastern shores of Lake Urmia, northwestern Iran. Previous research in the area suggested that almost all archaeological obsidian found in this area originated from the Nemrut Daĝ source located in the Lake Van region of Anatolia (Turkey). More recent research on obsidian artefacts from the Lake Urmia region has identified a significant number of obsidian artefacts with compositions different from the sources near Lake Van. This suggests that the obsidian artefacts are from a yet to be identified geological source, but possibly one that was not too distant. In order to advance our knowledge of Iranian obsidians and eventually refine provenance criteria we analysed obsidian from 22 Chalcolithic sites and some source areas. The compositions of both obsidian source samples and artefacts were determined using wave length dispersive X-ray fluorescence spectrometry (WDXRF). This paper presents results from the trace elemental analysis of both geological and archaeological obsidians, providing important new data concerning the diachronic relationship between lithic technology and raw material in the north-west of Iran.     

GEOCHEMICAL PROVENANCE OF OBSIDIAN ARTEFACTS FROM THE MSA SITE OF PORC EPIC,ETHIOPIA*

Obsidian geochemical analysis has become an indispensable tool in archaeological research and has wide applications. However, in spite of the abundance of geological and archaeological obsidian, there has been virtually no such investigation in Ethiopia. Recent instrumental characterization of 31 obsidian artefacts from the Middle Stone Age site of Porc Epic and numerous obsidian geological sources show that artefacts were being transported from as far as 250 km away. We believe this has relevance to the understanding of the emergence of modern human behaviour.     

Age and exploitation of obsidian from the Medicine Lake Highland,California

The relationship between major and minor elements, trace element composition, and age of obsidian sources within a volcanic field, is of considerable interest for obsidian source and artifact research in the New and Old World. The present study investigates this relationship in the Medicine Lake Highland of western North America. Geological evidence had indicated a very young age for all obsidian sources in the Highland, yet archaeological evidence suggested that obsidian was utilized for several thousand years. X-ray fluorescence analysis distinguished the latest flow (Glass Mountain) from the Cougar Butte, Grasshopper Flat, and Lost Iron Wells sources. Data obtained from two nearby archaeological sites showed that obsidian from the latter two sources was used by c. 7500 bc, while Glass Mountain material was not used (or available) until after 1360± 240 bp. These findings indicate that inferences of an extremely recent age for all obsidian sources in the volcanic field were unwarranted. Further analysis of major and minor elements indicated different hydration rates for these sources. The results argue that significant geochemical variability, as well as age differences, can exist between obsidian sources within the same volcanic field.     

The Hungarian and Slovak sources of archaeological obsidian: an interim report on further fieldwork,with a note on tektites

This report describes the results of fieldwork carried out in the Zemplén Mountain area of north-eastern Hungary in 1975. The aim of this work was to locate and sample geological sources of obsidian which may have been used by prehistoric man. These sources are of increased importance since the work of Nandris (1975) showed that the Romanian “sources” do not produce workable obsidian. During the fieldwork three sources in Hungary were visited and sampled; one of these was the previously unlocated source of Csepegö Forräs. A number of other possible localities for geological obsidian are mentioned in 19th and 20th century geological and archaeological literature, and the present state of knowledge with regard to these is summarized. Further sources exist in central and in south-eastern Slovakia. These sources were not visited but material has been obtained from both areas. The central Slovak sources do not produce workable obsidian and are not therefore relevant to archaeological studies. Obsidian from three localities in south-eastern Slovakia is of good glassy quality and further fieldwork is now needed to check the validity of these localities as geological sources. Reference is made to obsidian sources in the western U.S.S.R., and the problem of the use of tektites in archaeological sites is discussed.The obsidian samples obtained during this work are currently being analyzed using neutron activation, in order to characterize the sources on the basis of their trace element analysis and thus to relate them to archaeological obsidian from central and eastern Europe.     

Sources of obsidian artefacts,exchange networks and landscape use in Auca Mahuida (Neuquén,north-western Patagonia)

The results of the X-ray fluorescence (XRF) analysis of 59 obsidian samples from 11 archaeological sites in the Auca Mahuida region of north-western Neuquén, Argentina, are present. They indicate that several obsidian sources were used; however, the intensities of their exploitation were variable. Strong differences appear between the Colorado River basin, characterized by a low variability of obsidian groups from northern Neuquén; the Auca volcano, with a low variability of obsidian groups, but from local sources located north and southwards of the study area; and along Bajo del Añelo, which presents a high variability of obsidian groups from several local and non-local sources. The pattern recorded fits different mechanisms of access to the sources and the conveyance of obsidian across the landscape. Two distinct paths of direct access are suggested for obsidian availability along the Colorado River in northern Neuquén and for Portada Covunco obsidian in central Neuquén. Additionally, the presence of obsidian from sources in southern Neuquén province (Cerro Las Planicies-Lago Lolog), located about 350 km from the study area, is suggested. While not yet conclusive, this possibility parsimoniously integrates the available geochemical and spatial information, allowing the existence of either long-distance transport or indirect access by exchange or similar mechanisms to be proposed.     

PROVENANCE OF OBSIDIAN EXCAVATED FROM LATE CHALCOLITHIC LEVELS AT THE SITES OF TELL HAMOUKAR AND TELL BRAK,SYRIA*

X‐ray fluorescence and laser ablation inductively coupled plasma mass spectrometry (LA–ICP–MS) analyses conducted on 40 obsidian samples from the Late Chalcolithic 2 levels at Tell Hamoukar and Tell Brak in north‐east Syria have shown trends towards the exploitation of obsidian sources in the eastern Taurus. While the Bingöl region appears to provide the majority of obsidian to both sites, there is also evidence of more minor exploitation of a source in the Lake Van area and an altogether unknown source (X). This paper presents the data acquired from the analyses of the archaeological obsidian and situates these results within their chronological and regional contexts.     

VARIETIES AND SOURCES OF ARTEFACTUAL OBSIDIAN IN THE MIDDLE STONE AGE OF THE MIDDLE AWASH,ETHIOPIA*

Recent investigation of the geochemical provenance of obsidian artefacts from spatially and temporally variable archaeological sites in Ethiopia has shown the diversity of geological sources and concomitant differences in the transport of raw material to archaeological sites, thus allowing reconstruction of the utilization of raw material by early hominids as well as recent humans. We recognize 30 compositionally different obsidians that were used at the Middle Stone Age (MSA) archaeological sites of Aduma (A8), Halibee Herto, Aladi Springs and Porc Epic in the Middle Awash region. Probable sources of nine of these obsidians are Adokoma, Ayelu, Ida'ale, Assebot, Asboli, Gira‐Ale and Kone. Many compositional types are confined to a single site, but others are shared between sites, although shared obsidians between sites more than 300 km apart are exceptional.     

Obsidian sources and distribution systems in Island Southeast Asia: new results and implications from geochemical research using LA-ICPMS

This paper discusses new evidence of long-distance interaction networks in Island Southeast Asia obtained from geochemical analyses using SEM-EDXA and LA-ICPMS of 101 obsidian samples from 25 locations including seven obsidian sources and 19 archaeological sites. Given that there are obsidian sources distributed throughout much of Island Southeast Asia, the potential for obsidian studies to provide greater understanding of patterns of mobility and exchange in the Pre-Neolithic and Neolithic periods would seem to be considerable. This potential, however, remains largely unrealised as obsidian sourcing has hitherto only been carried out intermittently in Island Southeast Asia using PIXE-PIGME, XRF and other methods.     

THE PROVENANCE OF OBSIDIAN ARTEFACTS FROM THE WĀDĪ ATH‐THAYYILAH 3 NEOLITHIC SITE (EASTERN YEMEN PLATEAU) BY LA–ICP–MS

The geological sources of obsidian in the Red Sea region provide the raw material used for the production of obsidian artefacts found in prehistoric sites on both sides of the Red Sea, as far afield as Egypt, the Persian Gulf and Mesopotamia. This paper presents the chemical characterization of five obsidian geological samples and 20 prehistoric artefacts from a systematically excavated Neolithic settlement in highland Yemen. The major element concentrations were determined by SEM–EDS analysis and the trace element concentrations were analysed by the LA–ICP–MS method, an almost non‐destructive technique capable of chemically characterizing the volcanic glass. A comparison of archaeological and geological determinations allows the provenance of the obsidian used for the Neolithic artefacts to be traced to definite sources in the volcanic district of the central Yemen Plateau.     

New Data on the Exploitation of Obsidian in the Southern Caucasus (Armenia,Georgia) and Eastern Turkey,Part 1: Source Characterization

A large analytical programme involving both obsidian source characterization and obsidian artefact sourcing was initiated recently within the framework of the French archaeological mission ‘Caucasus’. The results will be presented in two parts: the first part, this paper, deals with the presentation and characterization of obsidian outcrops in the southern Caucasus, while the second presents some results obtained from a selection of artefacts originating from different Armenian sites dated to between the Upper Palaeolithic and the Late Bronze Age. The same analytical method, LA–ICP–MS (laser ablation inductively coupled plasma mass spectrometry), has been used to characterize all the studied samples (both geological and archaeological). This method is more and more widely used to determine the elemental composition of obsidian artefacts, as it causes minimal damage to the studied objects. We present in this first part new geochemical analyses on geological obsidians originating from the southern Caucasus (Armenia, Georgia) and eastern Turkey. These data enhance our knowledge of the obsidian sources in these regions. A simple methodology, based on the use of three diagrams, is proposed to easily differentiate the deposits and to study the early exploitation of this material in the southern Caucasus.     

Long-distance provenance for obsidian artifacts of Mesoamerica Preclassic and Early Classic periods found in the Los Naranjos Archaeological Park (Honduras)

Los Naranjos is one of the most important pre-Columbian human settlements of Honduras related to the south-easternmost border of the Mayan civilization. Although the archaeological site mostly spans from 850 BC to 1250 AD, the present obsidian study was only focused on the Preclassic and Early Classic periods (Jaral, 800–400 BC and Edén, 400 BC–550 AD) where undamaged blades and/or retouched obsidian flakes are rare. In this way, the INAA analyses of 17 obsidian samples, compared with major-trace elements data of Honduran and Guatemalan obsidian sources, are mostly representative of waste flakes. Lithic artifacts of Los Naranjos such as sandstones, basalts, and quartzites come from local geological outcrops; whereas, obsidian provenance has to be searched from sources which are located within a radius up to 300 km far away. San Luis, La Esperanza, and Güinope obsidian sources are located in Honduras while the three most exploited Highland Guatemalan obsidian outcrops, which have been dominating long-distance trade in the Maya area mostly for the Classic-Postclassic periods, are San Martin de Jilotepeque, El Chayal, and Ixtepeque. An Ixtepeque provenance, for all the investigated obsidian samples of Preclassic and Early Classic periods found in the Los Naranjos Archaeological Park, was established, thus emphasizing a long-distance source (180 km). This also confirms that Ixtepeque represents the most important provenance of the obsidian artifacts found in archaeological sites of Western and Central-Western Honduras. The possible role played by some of the most important rivers of Guatemala and Honduras as waterway networks of transport was finally pointed out. New INAA chemical data from the Honduran obsidian source of La Esperanza (“Los Hoyos”, 4 samples) are also reported in this paper.     

The application of Mössbauer spectroscopy to the characterisation of western mediterranean obsidian

Iron-57 Mössbauer absorption spectra have been measured for samples of obsidian from known geological flows and from archaeological site material from the western Mediterranean region. Of the four main sources available to prehistoric man it is possible to distinguish Sardinian (SA) and Pantellerian obsidian from Lipari obsidian on the basis of differences in the local atomic surroundings of iron atoms, as determined from the Mössbauer spectra. There is, however, some overlap between Lipari and Pontine Island obsidians. The Gabellotto flow on Lipari is readily identified through the presence of magnetite inclusions. The ratio of ferric to ferrous ions is found to be much higher in the surface layers (< 60 μm) than in the bulk obsidian as detected using Mössbauer backscattering.     

PARTICLE‐INDUCED X‐RAY EMISSION (PIXE) ANALYSIS OF OBSIDIAN FROM TEOTIHUACAN

A collection of 50 archaeological obsidian samples studied in the framework of the Preciudadela Project (Teotihuacan, Mexico) has been analysed using particle‐induced X‐ray emission (PIXE) with the external beam line of the Accélérateur Grand Louvre d'Analyse Elémentaire facility (C2RMF, Paris) and of the Instituto de Física (UNAM, Mexico). This work addresses the provenance of these obsidian samples, with the purpose of determining if they come from the obsidian sources exploited by Teotihuacans (mainly Otumba and Sierra de Pachuca), from other sources, or arrived via commercial exchanges with other regions. For that, the elemental compositions derived from the PIXE spectra have been compared with data published in the literature on the basis of instrumental neutron activation analysis. From the concentrations of selected key elements (Na, K, Mn, Fe, Zn, Rb, Sr, Zr), it was possible to unambiguously assign the provenance of most samples. Many originate from two major sources, namely Sierra de Pachuca (Hidalgo) and Otumba (Mexico), which were the main obsidian deposits used by the Teotihuacans. However, some samples exhibit a compositional fingerprint matching other provenances, i.e., Paredón (Puebla) and Zacualtipan (Hidalgo).     

Increasing data (INAA) on Ecuadorian obsidian artifacts: preliminary provenance and a clue for pre-Columbian eastward trade

In this work we carried out INAA major (Na, K, Ca and Fe %) and trace (ppm) elements (plus Mn by FAAS analysis) of 15 obsidian samples (waste flakes) coming from an unknown archaeological site (¹⁴C-AMS age of 1425 AD) located on the south-eastern flank of the back-arc Sumaco volcano (to the east of the Cordillera Real) and from two already known pre-Columbian archaeological localities: La Florida (Quito) and Milan (Cayambe). Literature compositional data of the Ecuadorian obsidian outcrops provide some constraints on the provenance of the analyzed waste flakes, even though different methods of analyses make comparisons a difficult task. Concerning the obsidian artifacts of La Florida and Milan, they come from the well known Sierra de Guamanì obsidian sources (Cordillera Real). By contrast, the obsidian fragments of the Sumaco settlement show some compositional characters compatible with obsidian erratic pebbles recently discovered in some river banks of the Amazonian foothills draining the easternmost flanks of the Antisana volcano in the Cordillera Real as well. In this way, the obsidian artifacts found at the Sumaco site reinforce the opinion that Ecuadorian source inventory is not yet exhaustive. Although the Antisana volcano seems to be the best candidate to find out additional primary outcrops of obsidian sources, it cannot be also excluded that sub-Andean and Amazonian people directly took advantage from obsidian secondary sources (e.g. river banks), rather than procurements from primary outcrops in the Cordillera Real. The new archaeological findings at the Sumaco volcano are really of paramount importance in tracing the ancient routes of a possible obsidian eastward trade toward the Amazonian region.     

THE GEOCHEMISTRY OF THE MAJOR SOURCES OF ARCHAEOLOGICAL OBSIDIAN ON HOKKAIDO ISLAND (JAPAN): SHIRATAKI AND OKETO

The geochemical compositions for obsidian from two of the most important sources on the Japanese island of Hokkaido, Shirataki and Oketo, are presented. This work represents the first systematic study of obsidian geochemistry on Hokkaido from the view of modern methodological standards. The study was performed with the help of neutron activation analysis to determine the concentrations for 28 elements. The results obtained allow us to subdivide both sources into two geochemical groups (Shirataki‐A and ‐B; and Oketo‐A and ‐B), with each representing an individual sub‐source. Obsidian from both Shirataki and Oketo sources is identified at archaeological sites located on Hokkaido, on the neighbouring Sakhalin Island and Kurile Islands, and in the lower course of the Amur River basin. The distance of obsidian transport during the Upper Palaeolithic was up to ～250 km, and in the following Neolithic and Palaeometal periods up to ～1200 km. This testifies to the wide distribution of Hokkaido obsidian to archaeological complexes in North‐East Asia and its active transport/exchange in prehistory. The data presented here should be used as a reference for the obsidian geochemistry of Shirataki and Oketo sources from now on.     

Characterization of archaeological obsidians from Lagartero, Chiapas Mexico by PIXE

Obsidian samples from the archaeological site of Lagartero, Chiapas, Mexico and other samples from Mexican and Guatemalan sources were analyzed by proton induced X-ray emission. Statistical treatments such as principal-component analyses were applied to the data set. Obsidians from Lagartero were identified as coming from two obsidian sources in Guatemala and one in Mexico, thus indicating that there was contact between people of Lagartero and other Mayans or Mesoamericans.     

Characterization of New Zealand obsidian using PXRF

New Zealand has some of the most active areas of rhyolitic volcanism in the world and this has produced numerous obsidian sources in the northern half of the North Island. In total archaeologists have recognized 27 named locations from which obsidian can be obtained scattered across 4 geological source regions. Shortly after colonization in the late 13th century AD Polynesian settlers began transporting this material some thousands of kilometers throughout the country and across the sea in small quantities to distant neighbors in the Kermadecs and Chatham islands. Although considerable research has been conducted on obsidian sourcing in New Zealand the complexity of geochemical source discrimination and the lack of a practical method of non-destructive geochemical analysis has hindered progress. We present the results of our use of PXRF to provide geochemical data on New Zealand obsidian sources and to compare the use of discriminant analysis and classification tree analysis to discriminate among sources and attribute archaeological samples to sources. Our research suggests that classification tree analysis is superior to discriminant analysis in sourcing studies. A large case study using an important settlement phase site (S11/20) from the Auckland region demonstrates the utility of the methods and the results support a model of high degrees of mobility and interaction during the early settlement of New Zealand.     

The distribution and provenance of archaeological obsidian in central and eastern Europe

The sources of archaeological obsidian in central and eastern Europe are briefly described and analyses of 48 samples from 10 of these sources in northeast Hungary and southeast Slovakia are reported. Instrumental Neutron Activation Analysis was used to determine 16 trace elements and two major elements. Principal Components Analysis supported by Discriminant Analysis showed seven analytical groups in these data. A total of 270 pieces of archaeological obsidian were assigned by Discriminant Analysis to three of the Carpathian source groups defined, the remaining four source groups not being represented in the archaeological record. The three source groups used are: (1) Szöllöske and Málá Toron?a in Slovakia (designated group Carpathian 1); (2) Csepegö Forrás, Tolcsva area, Olaszliszka and Erdöbénye in Hungary (Carpathian 2a); and (3) Erdöbénye (Carpathian 2b). Carpathian 2a and 2b type obsidians are both found at the re-deposited source of Erdöbénye. Carpathian obsidian was used most widely in Hungary, Slovakia and Romania, and also reached south to the Danube in Yugoslavia, west to Moravia, Austria and to the Adriatic near Trieste, and north to Poland. Carpathian 2a obsidian was used in the Aurignacian period, Carpathian 1 in the Gravettian and Mesolithic, and Carpathian 1, 2a and 2b in the Neolithic, when Carpathian 1 predominated and obsidian use was at its most intensive. Only Carpathian I type has been identified in the Copper and Bronze Ages. There is no evidence at present for any overlap between the Carpathian obsidian distribution and the distributions of the Near Eastern or Aegean sources, but there is an overlap with Mediterranean obsidian at the Neolithic site of Grotta Tartaruga in northeast Italy where Liparian and Carpathian 1 material were identified. The distribution of obsidian from the Carpathian sources is considered in terms of linear supply routes. Based on limited available evidence the supply zone is significantly smaller and the rate of fall-off with distance slightly lower than that reported for Near Eastern obsidians.     

Non-destructive PXRF analysis of museum-curated obsidian from the Near East

Non-destructive pXRF has the potential to expand sample populations of archaeological provenancing studies by facilitating access to museum collections of artefacts. In this study, we use museum-curated obsidian for the sites of Tell Brak, Mersin-Yümüktepe and Tell Arpachiyah and geological obsidian samples from central Turkey to demonstrate that non-destructive pXRF is comparable to other techniques in differentiating between Near Eastern obsidian sources. Secondly, we use multivariate analysis to evaluate whether non-destructive pXRF is sufficiently accurate and precise to enable comparison with legacy datasets from previous analysis using PIXE and ICP-MS. Multivariate analyses show that instrumental offsets between the results from pXRF and other analytic techniques are less than source-to-source variation typical for most Anatolian obsidian outcrops and pXRF-identified geochemical groups are directly comparable to legacy datasets. Non-destructive pXRF analysis indicates that during the Chalcolithic, East Göllü da? and Nenezi da? continued to be the major sources of obsidian to communities in central Anatolia and the Bingöl range and mountains surrounding Lake Van supplied sites east of the Levant. Variations in the analysed assemblages suggest a number of obsidian exchange networks existed during this period.