Geomaterials in Cultural Heritage
The Geological Society of London Books Editorial Committee R. PANKHURST ( U K ) (CHIEF EDITOR)
Society Books Editors J. GREGORY (UK) J. GR1FF1THS (UK) J. HOWE ( U K ) P. LEAT ( U K ) N. ROBINS ( U K ) J. TURNER ( U K )
Society Books Advisors M. BROWN ( U S A ) R. GIERF- ( G e r m a n y ) J. GLUYAS ( U K ) D. STEAD ( C a n a d a ) R. STEPHENSON ( N e t h e r l a n d s ) S. TURNER ( A u s t r a l i a )
Geological Society books refereeing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society's Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society Book Editors ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Socie~.'. The referees' forms and comments must be available to the Society's Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. More information about submitting a proposal and producing a book for the society can be found on its web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways: MAGGETTI, M. & MESSIGA, B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257. SHORTLAND, A. J., HOPE, C. A. & TITE, M. S. 2006. Cobalt blue painted pottery from 18th Dynasty Egypt. In: MAGGETTI, M. & MESSIGA, B. (eds) Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 91 - 100.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 257
Geomaterials in Cultural Heritage EDITED BY MARINO MAGGETTI University of Fribourg, Switzerland and BRUNO MESSIGA Universit~ degli Studi di Pavia, Italy
2006 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society's fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society's international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the American Geological Institute (AGI), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists' Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies' publications at a discount. The Society's online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J 0BG: Tel. +44 (0)20 7434 9944: Fax +44 (0)20 7439 8975; E-mail: enquiries @geolsoc.org.uk. For information about the Society's meetings, consult Events on www.geoisoc.org.uk. To find out more about the Society's Corporate Affiliates Scheme, write to
[email protected].
Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN, UK (Orders: Tel. +44 (0)1225 445046 Fax +44 (0)1225 442836) Online bookshop: www.geolsoc.org.uk/bookshop
The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. (~) The Geological Society of London 2006. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London WlP 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/06/$15.00.
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. I S B N 1-86239-195-5 Typeset by Techset Composition, Salisbury, UK Printed by MPG Books Ltd, Bodmin, UK
Distributors USA AAPG Bookstore PO Box 979 Tulsa OK 74101-0979 USA Orders: Tel. +1 918 584-2555 Fax +1 918 560-2652 E-mail bookstore @aapg.org India Affiliated East-West Press Private Ltd Marketing Division G-I/16 Ansari Road, Darya Ganj New Delhi I l0 002 India Orders: Tel. +91 11 2327-9113/2326-4180 Fax +91 il 2326-0538 E-mail
[email protected] Japan Kanda Book Trading Company Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-shi, Tokyo 206-0034 Japan Orders: Tel. +81 (0)423 57-7650 Fax +81 (0)423 57-7651 E-mail geokanda@ ma.kcom.ne.jp
Contents
Preface
vii
MAGGETTI, M. Archaeometry: quo vadis?
SMITH, D. C. A review of the non-destructive identification of diverse geomaterials in the cultural heritage using different configurations of Raman spectroscopy
Pottery (BC) BASSO, E., BINDER, D., MESSIGA,B. & RICCARDI,M. P. The Neolithic pottery of Abri Pendimoun (Castellar, France): a petro-archaeometric study
33
LAVIANO,R. & MUNTONI,I. M. Provenance and technology of Apulian Neolithic pottery
49
MAGGETTI, M. & GALETTI,G. Late La Tbne pottery from western Switzerland: one regional or several local workshops?
63
MOMMSEN, H., BONANNO, A., CHETCUTI BONAVITA, K., KAKOULLI,I., MUSUMECI, M., SAGONA, C., SCHWEDT, A., VELLA, N. C. • ZACHARIAS,N. Characterization of Maltese pottery of the Late Neolithic, Bronze Age and Punic Period by neutron activation analysis
81
SHORTLAND,A. J., HOPE, C. A. & TITE, M. S. Cobalt blue painted pottery from 18th Dynasty Egypt
91
SHOVAL, S., BECK, P. & YADIN,E. The ceramic technology used in the manufacture of
101
Iron Age pottery from Galilee
SMITH, M. S. & TRINKLEY,M. B. Fibre-tempered pottery of the Stallings Island Culture from the Crescent site, Beaufort County, South Carolina: a mineralogical and petrographical study
119
Pottery (AD) BIANCHINI, G., MARROCCHINO,E., MORETTI, A. & VACCARO, C. Chemicalmineralogical characterization of historical bricks from Ferrara: an integrated bulk and micro-analytical approach
127
~OLAK, M., MAGGETTI,M. & GALETTI,G. Golden mica cooking pottery from
141
Grkeyfip (Manisa), Turkey
vi
CONTENTS
DELL'AQUILA, C., LAVIANO,R. & VURRO, F. Chemical and mineralogical investigations of majolicas (16th-19th centuries) from Laterza, southern Italy
151
VENDRELL-SAZ, M., MOLERA, J., ROQUI~,J. & PI~REZ-ARANTEGUI,J. Islamic and Hispano-Moresque (mfidejar) lead glazes in Spain: a technical approach
163
Glass
ARLETTI, R., CIARALLO, A., QUARTIERI, S., SABATINO, G. & VEZZALINI,G. Archaeometric analyses of game counters from Pompeii
175
ERAMO, G. Pre-industrial glassmaking in the Swiss Jura: the refractory earth for the glassworks of Derriere Sairoche (ct. Bern, 1699-1714)
187
FREESTONE, I. C. Glass production in Late Antiquity and the Early Islamic period: a geochemical perspective
201
MARCHESI, V., NEGRI, E., MESSIGA, B. & RICCARDI, M. P. Medieval stained glass windows from Pavia Carthusian monastery (northern Italy)
217
Stone
ANTONELLI, F., SANTI, P., RENZULLI,A. & BONAZZA, A. Petrographic features and thermal behaviour of the historically known 'pietra ollare' from the Italian Central Alps (Valchiavenna and Valmalenco)
229
BELLELLI, C., PEREYRA, F. X. & CARBALLIDO,M. Obsidian localization and circulation in northwestern Patagonia (Argentina): sources and archaeological record
241
D'AMIco, C. & STARNINI,E. Prehistoric polished stone artefacts in Italy: a petrographic and archaeological assessment
257
GANDIN, A., CAPEZZUOLI,E. & CIACCI,A. The stone of the inscribed Etruscan stelae
273
from the Valdelsa area (Siena, Italy)
MILLER, S., MCGIBBON, F. M., CALDWELL, D. H. & RUCKLEY, N. A. Geological tools to interpret Scottish medieval carved sculpture: combined petrological and magnetic susceptibility analysis
283
MORGENSTEIN, M. Geochemical and petrographic approaches to chert tool provenance studies: evidence from two western USA Holocene archaeological sites
307
QUARESIMA, R., GIAMPAOLO,C. SPERNANZONI, F. & VOLPE, R. Identification, characterization and weathering of the stones used in medieval religious architecture in L'Aquila (Italy)
323
Mortar
CARO, F., DI GIULIO, A. & MARMO, R. Textural analysis of ancient plasters and mortars: reliability of image analysis approaches
337
Index
347
Preface
The scientific study of monuments, as well as objects from excavations and museums, deals with their origin, technique, age and conservation. Such topics were addressed during the one-day topical symposium 'Geomaterials in Cultural Heritage' of the 32nd International Geological Congress held in Florence on 20-28 August 2005. We have edited this volume by assembling papers of participants of the Florence meeting, as well as invited contributions, to present a wide view of the interdisciplinary application of geoscience disciplines, and to reaffirm the important contribution of geosciences to solve problems concerning the study of complex materials such as minerals, rocks, glass, metals, mortar, plaster, slags and pottery. This interdisciplinary application of geosciences includes field geology, geophysics, microscopy, textural analysis, physical methods and geochemistry as fundamental support to disclose hidden information, retained by the ancient materials, such as the raw materials provenance, the firing technology, the ancient recipes and the alteration pathway. The volume
is dedicated to all scholars eager to undertake or to continue an exciting research activity. Many colleagues helped us in the review process and we thank C. D'Amico, F. Antonelli, M. Baxter, C. Belelli-Pereyra, G. Bianchini, G. Bigazzi, F. Car6, G. Eramo, I. Freestone, A. Gandin, K. Gherdan, B. Grob6ty, R. Heimann, A. Jornet, R. Laviano, L. Lazzarini, S. Miller, H. Mommsen, R. Quaresima, M. P. Riccardi, G. Schneider, V. Serneels, A. Shortland, S. Shoval, D. C. Smith, S. Smith, G. Thierrin-Michael, M. Tite, S. Trfimpler, M. Vendrell-Saz, G. Wagner and S. Wolf for their goodwill and rigorous review of the submissions. We acknowledge the efficient assistance and the exemplary editorial support of the Geological Society publishing staff (particularly Angharad Hills and Sally Oberst) and the remarkable technical help from Nicole Bruegger. Marino Maggetti Bruno Messiga
Archaeometry: quo
vadis?
MARINO MAGGETTI University of Fribourg, Department of Geosciences, Mineralogy and Petrography, Chemin du Musde 6, CH-1700 Fribourg, Switzerland (e-mail:
[email protected])
Abstract: First, a brief overview of the tasks and the historical development of archaeometry will be given. Although archaeometry is generally doing well, a few issues currently faced by this discipline will be outlined. These include: (1) funding for projects and research positions; (2) the appeal of archaeometry to a new generation of academics; (3) the standard of publications; (4) the safeguarding of and the immediate access to scientific data.
Scientific study of raw materials and products used in prehistoric and historical time involves an interdisciplinary collaboration between archaeology, art history, preservation of the cultural heritage, ethnography and science. This area of research, in which these disciplines overlap, is known as archaeometry or archaeological sciences. The term geomaterials includes rocks, soils, mortars, pigments, ceramics, glass and slags. Scientific analysis of these objects aims at answering the following questions: (1) Where does the raw material come from? (2) Where was the object manufactured? (3) How was it manufactured (technique)? (4) What was its purpose (function)? (5) When was it manufactured (dating)? Scientific analysis should not limit itself to the qualitative and quantitative description of the 'chai'ne oprratoire'. Rather, it should approach these questions in a holistic manner. This involves the socio-cultural environment in which the artefact was manufactured (household, workshop, etc.), distributed and used. Collaboration with archaeologists and art historians needs to show how and why a particular technique was introduced or a specific manufacturing process used. It also needs to clarify the intention behind a certain function and the choice of a particular trading structure. In the field of preservation, material-specific properties of unweathered objects must be compared with their decay products so as to work out restoration concepts within a framework of interdisciplinary collaboration.
Methods and history Experimental methods used in the field of archaeometry have been described in a number of papers (Aitken 1961, 1985, 1990, 1998; Brothwell & Higgs 1968; Berger 1970;
Tite 1972; Fleming 1976; Hrouda 1978; Goffer 1980; Riederer 1981a, 1987; Matteini & Moles 1984; Cuomo di Caprio 1985; Mommsen 1986; Parks 1986; Wagner & Van den Haute 1992; Taylor & Aitken 1997; Wagner 1998; Ciliberto & Spoto 2000; Barclay 2001; Brothwell & Pollard 2001; Martini et al. 2004). Apart from a multitude of papers published in journals or books, there are many specifically geomaterial-related monographs, as well as proceedings from conferences. We shall name only a few of these and limit them to three domains as examples, because a complete listing would go beyond the scope of this Introduction. Ceramics. Shepard 1956; Matson 1965; Brill 1971; Picon 1973; Rye & Evans 1976; Peacock 1977, 1982; Winter 1978; Drmians d'Archimbaud & Picon 1980; Arnold 1981, 1985, 1993; Howard & Morris 1981; Hughes 1981; Thompson 1982; Rice 1984, 1987, 1997; Kingery 1985, 1986a, b; Laubenheimer 1985, 1992, 1998; Empereur & Garlan 1986; Jones 1986; Kingery & Vandiver 1986; Riederer 1987; Atasoy & Raby 1989; Lenoir et al. 1989; McGovern & Notis 1989; Middleton & Freestone 1991 ; Noll 1991; Wilson 1991; Li Jiazhi & Chen Xianqiu 1992; Mrry 1992; Neff 1992; Pollard 1992; Failla 1993; Burragato et al. 1994; Guo Jinkum 1995; Lindahl & Stilborg 1995; Vendrell-Saz et aL 1995; Vicenzini 1995; Whitbread 1995; Frontini & Grassi 1996; Cumberpatch & Blinkhorn 1997; Drmians d'Archimbaud 1997; Freestone & Gaimster 1997; Gaimster 1997; Gibson & Woods 1997; Lang & Middleton 1997; Santoro Bianchi & Fabbri 1997; Druc 1998; Fabbri & Lega 1999; Levi 1999; Ruf et al. 1999; Velde & Druc 1999; Wood 1999; Henderson 2000; Rosen 2000; Shortland 2001; Veeckman et al. 2002; D'Albis 2003;
From: MAGGETTI,M. & MESSIGA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications,257, 1-8. 0305-8719/06/$15.00 (c) The Geological Society of London 2006.
2
M. MAGGETI'I
D'Anna et al. 2003; Di Pierro et al. 2003; Keblow Bernsted 2003; Bargossi et al. 2004; Gurt i Esparraguera et al. 2005; Livingstone Smith et al. 2005. Glass. Lucas 1921; Caley 1962; Sayre 1964; Ankner 1965; Berger 1970; Oppenheim et al. 1970; Besborodov 1975; Newton & Davison 1978; Frank 1982; Olin & Franklin 1982; Wertime & Wertime 1982; Kazmarzyck & Hedges 1983; Lambert 1984; Bhardwaj 1987; Bimson & Freestone 1987; Riederer 1987; Henderson 1989, 2000; Brill & Martin 1991; Foy & Sennequier 1991; Mendera 1991; Tait 1991; Vandiver et al. 1992; Lilyquist & Brill 1993; Foy 1995, 2001; Hook & Gaimster 1995; Pollard & Heron 1996; Kingery & McCray 1998; Seibel 1998; Nenna 2002; Veeckman et al. 2002; Foy & Nenna 2003; Steppuhn 2003; Wedepohl 2003; Bargossi et al. 2004.
Gnoli 1971; Young 1973; Winkler 1973; Pensabene 1985, 1994, 1998; Sieveking & Hart 1986; Torrence 1986; Riederer 1987; Fant 1988; Herz & Waelkens 1988; Trou & Podany 1990; Borghini 1992; Moens et al. 1992; Bradley 1993; Klemm & Klemm 1993; Moorey 1994; Maniatis et al. 1995; Cunliff & Renfrew 1997; Shackley 1997; Schvoerer 1999; Henderson 2000; Roux 2000; De Nuccio & Ungaro 2002; Herrmann et al. 2002; Lazzarini 2002, 2004; Kardulias & Yerkes 2003; Poupard & Richard 2003; Bargossi et al. 2004. The first scientific analyses of ceramics, metals and pigments started early; that is, at the beginning of the 19th century (Riederer 198 lb, 1987; Maggetti 1990, 1994a). The foundation of specialist laboratories at museums and universities, such as the Chemisches Laboratorium der k6niglichen Museen zu Berlin (RathgenForschungslabor, 1888), as well as the Research Laboratory for Archaeology and the History of Arts (1955) at the University of Oxford, were milestones in the development of archaeometry. The number of similar institutions, active working groups and professional societies has increased ever since. The publication of several archaeometric journals was initiated, along with a great number of conferences. Obviously, archaeometry is an encouragingly vital discipline, but is it free of problems? Stone.
Problems faced by archaeometry Fundamental aspects of the status of archaeometry have been discussed extensively by Tite (1991, 2004) and Jones (2004). It is therefore unnecessary to further comment on them here.
However, it appears appropriate to take on some of the points raised by Widemann (1982), Fabbri (1992), Vidale (1992) and Maggetti (1994b). They deal with the funding of projects and research positions, the appeal of archaeometry to young scientists, the quality of scientific publications and the immediate and efficient access to research data. Funding
Although interdisciplinary research is up-to-date and highly praised by all entities, people working in this sector do indeed face difficulties. For instance, it is not easy to obtain funding, because problems concern historical disciplines, whereas answers and methods pertain to the sciences. In the quest for funds, one may find that a scientific body either rejects a project because questions are regarded as of historical nature, or it may pass it on to an arts or humanistic body, which in turn also declines the project, considering it to be of scientific nature. Citation index
It is becoming more and more common for universities, departments and scientists to be judged by the number of scientific papers being published in journals belonging to the citation index. Many archaeometric publications, however, do not appear in such journals, a fact that must have a detrimental effect on the career and reputation of the scientist concerned, if they do not already hold a position in archaeometry. On the other hand, the archaeometric results should also be published in archaeologically relevant journals or books, to strengthen interdisciplinary collaboration. As a result, young, enthusiastic scientists will be discouraged from pursuing a career in archaeometry. Stable research p o s i t i o n s
In addition to the problem mentioned above, there are far too few permanent posts for trained archaeometrists. It is understandable that in times when jobs are cut everywhere, scientific disciplines do not appear willing to redefine a vacancy as an interdisciplinary lecturing and research position. Because the questions pertain to the field of archaeology, art history and the preservation of ancient monuments, it should be up to these disciplines to safeguard or create the appropriate posts. Without such new positions it is impossible to retain the interest of young scientists or to motivate them to undertake research in archaeometry.
ARCHAEOMETRY: QUO VADIS? 'Hobby'
Unfortunately, there are far too many people doing archaeometric research as a 'hobby'. Many of these part-time archaeometrists are not familiar with archaeometric literature and reinvent the wheel, so to speak, i.e. they tackle questions that have been solved a long time ago. Often archaeologically irrelevant questions are investigated, insufficient numbers of samples are analysed and it can happen that 'poverty of measurements are covered up with sophisticated data processing' (Widemann 1982). Many such papers appear in unrefereed journals or books and escape the quality filter. Databases
Often, results of archaeometric studies will not be found in relevant journals, such as Archaeometry, Geoarchaeology, Journal of Archaeological Sciences, Revue d'Arch¢ombtrie, Journal o f Cultural Heritage and Marmora. Instead, they are published in books or journals that are difficult to access. Those who need to consider the citation index will be more likely to publish in journals of their specific discipline. In these journals, however, archaeometric contributions tend to disappear and colleagues cannot find them. It is therefore understandable that far too many good papers will hardly be read. As a result, analyses pertaining to problems that have already been investigated tend to be repeated, a fact that should be avoided considering the limited financial resources within the archaeometry community. In the field of ceramics, for instance, many working groups possess a large collection of chemical data and chemical reference groups, which, for the reasons mentioned above, are not readily accessible to everybody or will be lost once the group ceases its activity. Consequently, it is very important to make these treasures accessible to all people involved in archaeometry. These days, the internet provides a powerful instrument to display results periodically on the homepage of the relevant team, as is done by the Freiburg archaeometric working group (www.unifr.ch/geoscience/mineralogy/ archmet). However, this applies not only to ceramics, but also to the other geomaterials. As a result, one could avoid duplicate working and save money as well as time.
Final remarks The previous discussion has highlighted some of the problems faced by archaeometrists. What does the future of archaeometry look like? It is beyond any doubt that archaeometric research,
3
such as precise dating, is indispensable to the fields of archaeology and art history. Also, it is impossible to make restorations without the appropriate and relevant research being done. This presents a huge opportunity for archaeometry, because, owing to its impact on tourism, the study and preservation of our cultural heritage is likely to receive sufficient financial support from governments. However, it is necessary for research in all archaeometric sectors to focus on archaeometric competence centres that provide sufficient long-term employment, knowledge and sound technical apparatus. This is the only way in which young, well-trained scientists will be willing to commit themselves to the fascinating field of interdisciplinary research.
References AITKEN, M. J. 1961. Physics and Archaeology. Clarendon Press, Oxford. AITKEN, M. J. 1985. Thermoluminescence Dating. Academic Press, London. AITKEN, M. J. 1990. Science-based Dating in Archaeology'. Longman, Harlow. AtTKEN, M. J. 1998. An Introduction to Optical Dating. Oxford University Press, Oxford. ANKNER, D. 1965. Chemische und physikalische Untersuchungen an vor- und friihgeschichtlichen Gliisern. Technische Beitr~ige zur Arch~iologie, II. ARNOLD, D. (ed.) 1981. Studien zur altiigyptischen Keramik. Philipp von Zabem, Mainz. ARNOLD, D. 1985. Ceramic Theory and Cultural Process. Cambridge University Press, Cambridge. ARNOLD, D. 1993. Ecology and Ceramic Production in an Andean Community. Cambridge University Press, Cambridge. ATASOY, N. & RABY, J. (eds) 1989. Iznik. The Pottery of Ottoman Turkey. Alexandria Press, London. BARCLAY, K. 2001. Scientific Analysis of Archaeological Ceramics. A Handbook of Resources. Oxbow Books, Oxford. BARGOSSI, G. M., FRANZIN1,M. & MESSIGA, B. 2004. A showcase of the Italian research in applied petrology. Periodico di Mineralogia, 73, Special Issue 3. BERGER, R. (ed.) 1970. Scientific Methods in Medieval Archaeology. University of California Press, Berkeley. BESBORODOV, M. A. 1975. Chemie und Technologie der antiken und mittelalterlichen Gliiser. Ph. von Zabern Mainz. BHARDWAJ, H. C. (ed.) 1987. Archaeometry of glass. Proceedings of the Archaeometry Session of the XI International Congress on Glass, New Delhi. The Indian Ceramic Society, Calcutta. BIMSON, M. & FREESTONE,I. C. 1987. Early Vitreous Materials. British Museum Occasional Papers, 56. BORGHINk G. (ed.) 1992. Marmi antichi. Leonardo De Luca, Rome.
4
M. MAGGETI'I
BRADLEY, R. 1993. Interpreting the Axe Trade. Production and Exchange in Neolithic Britain. Cambridge University Press, Cambridge. GRILL, R. H. (ed.) 1971. Science and Archaeology. MIT Press, Cambridge, MA. GRILL, R. H. & MARTIN, J. H. (eds) 1991. Scientific research in early Chinese glass. In: Proceedings of the Archaeometr 3' of Glass Sessions of the 1984 International Symposium on Glass, Beijing, September 7, 1984 with Supplementary Papers. Coming Museum of Glass, Coming, NY. BROTHWELL, D. & HIGGS, E. 1968. Science in Archaeology. Thames and Hudson, London. BROTHWELL, D. R. & POLLARD, A. M. (eds) 2001. Handbook of Archaeological Sciences. Wiley, Chichester. BURRAGATO, E., GRUBESSI, O. & LAZZARINI,L. (eds) 1994. Ist European Workshop on Archaeological Ceramics 1991. Universit/a degli Studi 'La Sapienza', Rome. CALEY, E. R. 1962. Analysis of Ancient Glasses 17901957. Coming Museum of Glass, Coming, NY. CILIBERTO, E. ~; SPOTO, G. (eds) 2000. Moden7 Analytical Methods in Art and Archaeology. Wiley, New York. CUMBERPATCH, C. G. (~; BLINKHORN, P. W. (eds) 1997. Not so much a Pot, More a Way of Life. Oxbow Monograph, 83. CUNLIFF, B. 8£ RENFREW, C. (eds) 1997. Science and Stonehenge. Proceedings of the British Academy, 92. CUOMO DI CAPRIO, N. 1985./.z~ ceramica in archeologia. Antiche tecniche di lavorazione e moderni metodi d'indagine. L'Erma di Bretschneider, La Fenice, Rome. D'ALBlS, A. 2003. Traitd de la porcelaine de Sbvres. Faton, Dijon. D'ANNA, A., DESBAT, A., GARCIA, D., SCHMITT, A. & VERHAEGE, F. 2003. La Cdramique. La Poterie du Ndolithique aux Temps modernes. Errance, Paris. DE NuccIo M. & UNGARO, L. 2002. I n~lrmi colorati della Roma imperiale. Marsilio, Venice. DI~MIANS D'ARCHIMBAUD,G. (ed.) 1997. La Cdramique mddidvale en Mdditerrande. Actes du Vlbme Congrbs de I'AIECM2, Aix-en-Provence 1 3 - 1 8 novembre 1995. Narration, Aix-en-Provence. DI~MIANSD'ARCHIMBAUD,G. & PICON, M. (eds) 1980. La Cdramique mddidvale en Mdditerrande occidentale X d - X V d sikcle. Actes du Colloque international no 584, Valbonne 11 au 14 septembre 1978. CNRS, Paris. Dt PIERRO, S., SERNEELS, V. (~ MAGGETTI, M. (eds) 2003. Ceramic in the Society. Proceedings of the 6th European Meeting on Ancient Ceramics, Fribourg, Switzerland, 3 - 6 October 2001. Department of Geosciences, University of Fribourg. DRUC, I. C. 1998. Ceramic Production and Distribution in the Chavin Sphere of Influence (NorthCentral Andes). British Archaeological Reports, International Series, 731. EMPEREUR, J.-Y. & GARLAN, Y. (eds) 1986. Rechepz~hes sur les Amphores grecques. Actes du Colloque International organisd par le Centre National de la Recherche Scientifique, l'Universitd
de Rennes H et I'Ecole Fran9aise d'Athknes (Athbnes, 10-12 septembre 1984). Diffusion de Brocard, Paris. FABBRI, B. 1992. Archeometria: chi, come e perch,:. In: SANTORO BIANCHI, S., FAILLA, A. & GUERMANDI, M. P. (eds) Archeometria della Ceramica, Problemi di Metodo. Int. Centro Ceramico, Bologna, 127-129. FABBRI, B. & LEGA, A. M. (eds) 1999. I rivestimenti e le decorazioni nei manufatti antichi: quali opportunitgt per I'archeometria. Atti della 3a giornata di archeometria della ceramica. Faenza, 30 marzo 1999. University Press, Bologna. FAILLA, A. (ed.) 1993. Archeometria della ceramica. Problemi di Metodo. Atti 80 SlMCER-Simposio hlternazionale di ceramica, Rimini, 10-12 novembre 1992. Int. Centro Ceramico, Bologna. FANT, J. C. (ed.) 1988. Ancient Marble Quarrying and Trade. Papers from a Colloquium held at the Annual Meeting of the Archaeological Institute of America, San Antonio, Texas, December, 1986. British Archaeological Reports, International Series, 453. FLEMING, S. J. 1976. Authenticity in Art: the Scientific Detection of Forger3.'. Institute of Physics, London. FoY, D. (ed.) 1995. Le Verre de l'Antiquitd tardive et du haut Moven Age. Procddd de l'Association Franfaise pour l'Archgologie du Verre, Huitikme Rencontre. Mus~e arch~ologique d~partemental du Val d'Oise, Guiry-en-Vexin. FoY, D. 2001. Le Verre mddidval et son Artisanat en France mdditerrandenne. CNRS, Paris. FoY, D. & NENNA, M.-D. (eds) 2003. Ateliers de Verriers de l'Antiquitd b la Pdriode prd-industrielle. Actes de la 4kme Rencontre de l'Association Franfaise pour l'Archdologie du Verre, Rouen 1989. Association Franqaise pour l'Archrologie du Verre, Rouen. FoY, D. ~ SENNEQUIER, G. (eds) 1991. Echanges et Commerce du Verre dans le Monde antique. Actes du Colloque de I'AFA V, Aix-en-Provence et Marseille, 7 - 9 juin 2001. Monique Mergoil, Montagnac. FRANK, S. 1982. Glass and Archaeology. Academic Press, London. FREESTONE, I. & GAIMSTER, D. (eds) 1997. Pottery in the Making. World Ceramic Traditions. British Museum Press, London. FREESTONE, I. & JOHNS, C. (eds) 1980. Current Research in Ceramics: Thin Section Studies. British Museum Occasional Papers 32. FRONTINi, P. & GRASSI, M. T. (eds) 1996. Indagini Archeometriche Relative Alla Ceramica a Vernice nera: nuovi dati sulla Provenienza e la Diffusione. Atti del Seminario hzternazionale di Studio, Milano, 2 2 - 2 3 novembre 1996. New Press, Como. GAIMSTER, D. 1997. German Stoneware 1200-1900. British Museum Press, London. GmSON, A. & WOODS, A. 1997. Prehistoric Pottery. for the Archaeologist, 2nd edn. Leicester University Press, London. GNOLI, R. 1971. Marmora Romana. Elefante, Rome. GOFFER, Z. 1980. Archaeological Chemistry. Wiley, New York.
ARCHAEOMETRY: QUO VADIS? Guo JINKUM (ed.) 1995. Science and Technology of Ancient Ceramics 3. Proceedings of the International Symposium on Ancient Ceramics (ISAC'95), Shanghai. Shanghai Science and Technology Press, Shanghai. GURT I ESPARRAGUERA,J. M., BUXEDAI GARRIGOS,J. & CAU ONTIVEROS, M. A. (eds) 2005. LRCW, Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean. Archaeology and Archaeometry. British Archaeological Reports, International Series, 1340. HENDERSON, J. (ed.) 1989. Scientific Analysis in Archaeology and its Interpretation. Oxford University Committee on Archaeology Monograph, 19. University of California, Los Angeles, Institute of Archaeology Research Tools, 5. HENDERSON, J. 2000. The Science and Archaeology of Materials. An Investigation of Inorganic Materials. Routledge, London. HERRMANN, J. J., HERZ, N. & NEWMANN, R. (eds) 2002. Interdisciplinary Studies on Ancient Stones. Proceedings of the 5th ASMOSIA Conference, Museum of Fine Arts, Boston, June 11-15, 1998. Archetype, London. HERZ, N. & WAELKENS, M. (eds) 1988. Classical Marble: Geochemistry, Technology, Trade. NATO ASI Series E, 153. HOOK, D. R. & GAIMSTER, D. R. M. (eds) 1995. Trade and Discovery. The Scientific Study of Artefacts from Post-Medieval Europe and Beyond. British Museum Occasional Papers, 109. HOWARD, H. & MORRIS, E. (eds) 1981. Production and Distribution: a Ceramic Viewpoint. British Archaeological Reports, 120. HROUDA, B. (ed.) 1978. Methoden der Archiiologie. Eine Einfiihrung in ihre naturwissenschaftlichen Techniken. Beck, Munich. HUGHES, M. J. (ed.) 1981. Scientific Studies in Ancient Ceramics. British Museum Occasional Papers, 19. JONES, A. 2004. Archaeometry and materiality: materials-based analysis in theory and practice. Archaeometry, 46(3), 327-338. JONES, R. E. 1986. Greek and Cypriot Pottery. A Review of Scientific Studies. British School at Athens, Fitch Laboratory Occasional Papers, 1. KARDULIAS, P. N. & YERKES, R. W. (eds) 2003. Written in Stone: the Multiple Dimensions of Lithic Analysis. Lexington Books, Lexington, MA. KAZMARZYCK, A. & HEDGES, R. E. M. 1983. Ancient Egyptian Faience. Aris and Phillips, Warminster. KEBLOW BERNSTED, A. M. 2003. Early Islamic Pottery. Materials and Techniques. Archetype, London. KINGEt~Y, W. D. (ed.) 1985. Ancient Technology to Modern Science. Ceramics and Civilisation, Vol. L American Ceramic Society, Columbus, OH. KINGERY, W. D. (ed.) 1986a. Technology and Style. Ceramics and Civilisation, Vol. II. American Ceramic Society, Columbus, OH. KINGERY, W. D. (ed.) 1986b. High Technology Ceramics. Past, Present and Future. Ceramics and Civilisation, VoL IlL American Ceramic Society, Westerville, OH.
5
KINGERY, W. D. & MCCRAY, P. (eds) 1998. The Prehistory and History of Glassmaking Technology. Ceramics and Civilization, Vol. VIII. American Ceramic Society, Columbus, OH. KINGERY, W. D. & VANDIVER, P. 1986. Ceramic Masterpieces--Art, Structure and Technology. Free Press, New York. KLEMM, D. & KLEMM, R. 1993. Steine und Steinbriiche im alten Aegypten. Springer, Berlin. LAMBERT, J. B. (ed.) 1984. Archaeological Chemistry III. American Chemical Society. Advances in Chemistry Series, 205. LANG, J. & MIDDLETON, A. (eds) 1997. Radiography of Cultural Material. Butterworth Heinemann, Oxford. LAUBENHEIMER, F. 1985. La Production des Amphores en Gaule Narbonnaise. Annales Litteraires de l'Universit~ de Besanqon, 327. LAUBENHEIMER, F. (ed.) 1992. Les Amphores en Gaule. Production et Circulation. Table Ronde lnternationale Metz, 4 - 6 octobre 1990. Annales Litt~raires de l'Universit6 de Besanqon, 474. LAUBENHEIMER, F. (ed.) 1998. Les Amphores en Gaule. Production et Circulation. Les Belles Lettres, Paris. LAZZARINI, L. (ed.) 2002. Interdisciplinary Studies on Ancient Stone. Proceedings of the Sixth International Conference (ASMOSIA VI), Venice, 1518 June 2000, Bottega d'Erasmo Aldo Ausilio, Padora. LAZZARINI, L. (ed.) 2004. Pietre e marmi antichi. Natura, caratterizzazione, origine, storia d'uso, diffusione, collezionismo. CEDAM, Padova. LENOIR, M., MANACORDA, O. & PANELLA, C. (eds) 1989. Amphores romaines et histoire ~conomique. Dix ans de recherche. Actes du colloque de Sienne (22-24 mai 1986) organis~ par l'Universith degli studi di Siena, l'Universitgt degli Studi di Roma-La Sapienza, le Centre national de la recherche scientifique (RCP 403) et l'Ecole franfaise de Rome. Ecole Fran~aise de Rome, Rome. LEVI, S. T. (ed.) 1999: Produzione e circolazione della ceramica nella Sibaritide protostorica. I. Impasto e doll Prima di Sibri 1. Insegna del Giglio, Firenze. LI JIAZHI & CHEN XIANQIU (eds) 1992. Science and Technology of Ancient Ceramics. Proceedings of the 1989 International Symposium, Shanghai. Shanghai Science and Technology Press, Shanghai. LILYQUIST, C. & BR1LL, R. H. (eds) 1993. Studies in Early Egyptian Glass. Metropolitan Museum of Art, New York. LINDAHL, A. & STILBORG, O. (eds) 1995. The Aim of Laboratory Analyses of Ceramics in Archaeology. Kungliche Vitterhets Historic och Antivitets Akademien Konferenser, 34. LIVINGSTONE SMITH, A., BOSQUET, D. & MARTINEAU R. (eds) 2005. Pottery Manufacturing Processes: Reconstitution and Interpretation. Acts of the XIVth UISPP Congress, University of Likge, Belgium, 2 - 8 September 2001. British Archaeological Reports, International Series, 1349. LUCAS, A. 1921. Ancient Egyptian Materials and Industries. Edward Arnold, London.
6
M. MAGGETTI
MAGGETTI, M. 1990. I1 contributo delle analisi chimiche alia conoscenza delle ceramiche antiche. In: MANNONI, T. ~ MOLINARI, A. (eds) Scienze in Archeologia. Insegna del Giglio, Firenze, 65-88. MAGGETT[, M. 1994a. Mineralogical and petrographical methods for the study of ancient pottery, h~: BURRAGATO, F., GRUBESSI, O. & LAZZARINI, L. (eds) Proceedings of the 1st European Workshop on Archaeological Ceramics. Dipartments Scienze della Terra, Universitg degli Studi di Roma 'La Sapienza', 23-35. MAGGETTI, M. 1994b. Is there still a future for archaeometry at universities? In: OLCESE, G. (ed.) Ceramica Romana e Archeometria: lo Stato degli studi. Insegna del Giglio, Firenze, 15-22. MANIAT1S, Y., HERZ, N. & BASIAKOS, Y. (eds) 1995. The Study of Marble and other Stones used in Antiquity. Transactions of the 3rd ASMOSIA Conference. Archetype, London. MARTINI, M., MILAZZO, M. & PIACENTINI, M. (eds) 2004. Physics Methods in A rcheometo'. Proceedings o f the International School o f Physics 'Enrico Fermi', Varenna on Lake Como. Villa Monastero, 1 7 - 2 7 June 2003. IOS Press, Amsterdam. MATSON, F. R. 1965. Ceramics and Man. Methuen, London. MATTEINI, M. & MOLES, A. 1984. Scienza e Restauro, Metodi di Indagine. Nardini, Firenze. McGOVERN, P. E. & NOTIS, M. D. (eds) 1989. Crosscraft and Cross-cultural Interactions in Ceramics. Ceramics and Civilization, Vol. IV. American Ceramic Society, Columbus, OH. MENDERA, M. 1991. Archeologia e storia della produzione del vetro preindustriale. Giglio, Firenze. MERY, S. (ed.) 1992. Sciences de la Terre et Cdramiques archdologiques. Expdrimentations, applications. Documents et travaux de l'institut g6ologique Albert-de-Lapparent, 16. MIDDLETON, A. & FREESTONE, I. C. (eds) 1991. Recent Developments in Ceramic Petrology. British Museum Occasional Papers, 81. MOENS, L., DE PAEPE, P. & WAELKENS, M. 1992. Multidisciplinary Research and Interlaboratorv Cooperation: Keys to a Successful Provenance Determination of White Marble. Acta Archaeologica Lovanensia Monographiae, 4. MOMMSEN, H. 1986. Archiiometrie. Teubner Studienbiicher. Teubner, Stuttgart. MOOREY, P. R. S. 1994. Ancient Mesopotamian Materials and Industries. Clarendon Press, Oxford. NEFF, H. (ed.) 1992. Chemical Characterization o f Ceramic Pastes in Archaeology. Monographs in World Archaeology, 7. NENNA, M. D. (ed.) 2002. La Route du Verre. Ateliers primaires et secondaires du Second Milldnaire av. J.-C. au Moyen Age. Travaux de la maison de l'orient M~diterran~en, 22. NEWTON, R. G. & DAVlSON, S. 1978. Conservation of Glass. Butterworth, London.
NOLL, W. 1991. Alte Keramiken und ihre Pigmente. Studien zu Material und Technologie. Schweizerbart, Stuttgart. OLIN, J. S. & FRANKLIN,A. D. (eds) 1982. Archaeological Ceramics. Smithsonian Institution Press, Washington, DC. OPPENHEIM, A. L., BRILL, R. H., BARAG, D. & VON SALDERN, A. (eds) 1970. Glass and Glassmaking in Ancient Mesopotamia. Coming Museum of Glass, Coming, NY. PARKS, P. A. 1986. Current Scientific Techniques in Archaeology. Seminar Press, London. PEACOCK, D. P. S. (ed.) 1977. Pottery and Early Commerce. Characterization and Trade in Roman and Later Ceramics. Academic Press, London. PEACOCK, D. P. S. 1982. Pottery in the Roman World. Longman, Harlow. PENSABENE, P. (ed.) 1985. Marmi antichi. Problemi d'impiego, di restauro e d'identificazione. "L'Erma" di Bretschneider, Rome. PENSABENE, P. 1994. Le vie del marmo. Itinerari osticusi 7, Rome. PENSABENE, P. (ed.) 1998. Marmi antichi II. Cave e tecnica di lavorazione. Provenienze e distribuzione. 'L'Erma' di Bretschneider, Rome. PICON, M. 1973. Introduction b l'Etude technique des Cdramiques sigilldes de Lezoux. Centre de Recherches Techniques Gallo-Romaines, Lyon. POLLARD, A. M. (ed.) 1992. New Developments in Archaeological Science. Proceedings of the British Academy, 77. POLLARD, A. M. & HERON, C. 1996. Archaeological Chemistry. Royal Society of Chemistry, Cambridge. POUPARD, L. & RICHARD, A. (eds) 2003. Marbres en Franche-Comtd. Actes des journdes d'dtudes Besanfon, 10-12 juin 1999. ASPRODIC, Besanqon. RICE, P. M. (ed.) 1984. Pots and Potters. Current Approaches in Ceramic Archaeology. Institute of Archaeology, University of California, Los Angeles, Monograph, XXIV. RICE, P. M. 1987. Potter3' Analysis: a Sourcebook. University of Chicago Press, Chicago, IL. RICE, P. M. (ed.) 1997. The Prehistory & History of Ceramic Kilns. Ceramics and Civilization, Vol. VII. American Ceramic Society, Westerville, OH. RIEDERER, J. 1981a. Kunstwerke chemisch betrachtet. Materialien. Analysen, Altersbestimmung. Springer, Berlin. RIEDERER, J. 1981b. Zum gegenw~irtigen Stand der naturwissenschaftlichen Untersuchung antiker Keramik. In: ARNOLD, D. (ed.) Studien zur altiig3ptischen Keramik. Ph. yon Zabern, Mainz. RIEDERER, J. 1987. Archiiologie und Chemie-Einblicke in die Vergangenheit. RathgenForschungslabor, Berlin. ROSEN, J. 2000./_zl Manufacture de Meillonnas (Ain). Etude d'une Fabrique de Cdramique rdgionale. 1760-1870 Temps modernes 2. Monique Mergoil, Montagnac. Roux, V. (ed.) 2000. Cornaline de l'Inde. Des Pratiques techniques de Cambay aux Technosystkmes
ARCHAEOMETRY: QUO VADIS? de I'Indus. Maison des Sciences de l'Homme, Paris. RUF, M. A., MAGGETTI, M. & VENDRELL-SAZ, M. (eds) 1999. Archaeological and Archaeometric Studies. Proceedings of the 4th European Meeting on Ancient Ceramics, Andorra 1995. Govern d'Andorra, Ministeri de Turismo i Cultura, Andorra. RYE, O. S. & EVANS, C. 1976. Traditional Pottery Techniques of Pakistan: Field and Laboratory Studies. Smithsonian Contributions to Anthropology, 21. SANTORO BIANCHI, S. & FABBRI, B. (eds) 1997. Il Contributo delle Analisi Archeometriche alldo Studio delle Ceramiche grezze e Comuni. I11rapporto forma/funzione/impasto. Atti della l a Giornata di Archeometria della Ceramica, Bologna, 28 febbraio 1997. University Press, Bologna. SAYRE, E. V. 1964. Some Ancient Glass Specimens with Compositions of Particular Archaeological Significance. Brookhaven National Laboratory, Brookhaven, NY. SCHVOERER, M. (ed.) 1999. Archdomatdriaux. Marbres et autres Roches. Actes de la IVb Confdrence internationale ASMOSIA IV, France, Bordeaux, Talence, 9-13 octobre 1995. Presses Universitaires de Bordeaux, Bordeaux. SEIBEL, F. 1998. Technologie und Fertigungstechniken rOmischer Glashiitten am Beispiel der Ausgrabungen im Hambacher Forst. Aktualistische Vergleiche und Modelle. Galda & Wilch, Berlin. SHACKLEY, M. S. (ed.) 1997. Method and Theory in Archaeological Volcanic Glass Studies. Advances in Archaeological and Museum Studies. Plenum, New York. SHEPARD, A. O. 1956. Ceramics for the Archeologist. Carnegie Institution of Washington. Washington, DC. SHORTLAND, A. J. (ed.) 2001. The Social Context of Technological Change. Oxbow Books, Oxford. SIEVEKING, G. & HART, M. B. (eds) 1986. The Scientific Study of Flint and Chert. Proceedings of the Fourth International Flint Symposium held at Brighton Polytechnic, 10-15 April 1983. Cambridge University Press, Cambridge. STEPPUHN, P. (ed.) 2003. Glashiitten im Gespriich. Berichte und Materialien vom 2. lnternationalen Symposium zur archiiologischen Erforschung mittelalterlicher und friihneuzeitlicher Glashiitten Europas. Schmidt-R6mhild, Ltibeck. TAIT, H. (ed.) 1991. Five Thousand Years of Glass. British Museum Press, London. TAYLOR, R. E. & AITKEN, M. J. (eds) 1997. Chronometric Dating in Archaeology. Plenum, New York. THOMPSON, I. 1982. Grog-Tempered 'Belgic' Pottery of South-Eastern England. British Archaeological Reports, 108. TITE, M. 1972. Methods of Physical Examination in Archaeology. Seminar Press, New York. TITE, M. S. 1991. Archaeological science--past achievements and future prospects. Archaeometry, 33(2), 139-151.
7
TITE, M. S. 2004. Archaeometry--an overview. In: MARTINI, M., MILAZZOM. & PIACENTINI,M. (eds) Physics Methods in Archaeometry. IOS Press, Amsterdam, 347-356. TORRENCE, R. 1986. Production and Exchange of Stone Tools. Cambridge University Press, Cambridge. TROU, M. & PODANY,J. (eds) 1990. Marble. Art Historical and Scientific Perspectives on Ancient Sculpture. Papers Delivered at a Symposium Organized by the Department of Antiquities and Antiques Conservation and Held at the J. Paul Getty Museum April 28-30, 1988. J. Paul Getty Museum, Malibu, CA. VANDIVER, P. B., DRUZIK, J. R., WHEELER, G. S. & FREESTONE, [. C. (eds) 1992. Materials in A n and Archaeology II1. Materials Research Society Symposium Proceedings, 267. VEECKMAN, J., JENNINGS, S., DUMORTIER, C., WHITEHOUSE, D. & VERHAEGHE, F. (eds) 2002. Majolica and Glass from Italy to Antwerp and Beyond. The Transfer of Technology in the 16thearly 17th Century. Stad Antwerpen, Antwerp. VELDE, B. & DRUC, I. C. 1999. Archaeologial Ceramic Materials. Springer, Berlin. VENDRELL-SAZ, M., PRADELL, T. & MOLERA I MAITE GALERA, J. (eds.) 1995. Estudis sobre cer&mica antiga. Studies on Ancient Ceramics. In: Proceedings of the 2nd European Meeting on Ancient Ceramics, Barcelona, 18-21 November 1993. VICENZINI, P. (ed.) 1995. The Ceramics Cultural Heritage. In: Proceedings of the International Symposium The Ceramics Heritage of the 8th CIMTEC-World Ceramics Congress and Forum on New Materials, Florence, Italy, June 28-July 2, 1994. Monographs in Materials and Society, 2. VIDALE, M. 1992. I givani siberiani. Archaeometria della ceramica antica e il problema delle nuove figure professionali. In: SANTORO BIANCHI, S., FA1LLA, A. & GUERMANDI,M. P. (eds).Archeometria della Ceramica, Problemi di Metodo. Int. Centro Ceramico, Bologna, 131 - 150. WAGNER, G. A. 1998. Age Determination of Young Rocks and Artefacts--Physical and Chemical Clocks in Quaternary Geology and Archaeology. Springer, Berlin. WAGNER, G. A. • VAN DEN HAUTE, P. 1992. Fission Track Dating. Enke, Stuttgart. WEDEPOHL, K. H. 2003. Glas in Antike und Mittelalter. Geschichte eines Werkstoffes. Schweizerbart, Stuttgart. WERTIME, T. A. & WERTIME, S. F. (eds) 1982. Early Pyrotechnology: the Evolution of the First Fireusing Industries. Smithsonian Institution Press, Washington, DC. WHITBREAD, I. K. 1995. Greek Transport Amphorae. A Petrological and Archaeological Study. British School at Athens, Fitch Laboratory Occasional Papers, 4. WIDEMANN, F. 1982. Why is archaeometry so boring for archaeologists? In: OLIN, J. & FRANKLIN, F. (eds)Archaeological Ceramics.
8
M. MAGGE'Iq'I
Smithsonian Institution Press, Washington, DC, 29-34. WILSON, T. (ed.) 1991. Italian Renaissance Potterr: Papers Written in Association with a Colloquium at the British Museum. British Museum Publications, London. WINKLER, E. M. 1973. Stone. Properties, Durability in Man's Environment. Springer, Berlin.
WINTER, A. 1978. Die antike Glanztonkeramik. Praktische Versuche. Philipp von Zabern, Mainz. WOOD, N. 1999. Chinese Glazes. Their Origins, Chemistry and Recreation. A&C Black, London. YOUNG, W. J. (ed.) 1973. Applications of Science in Examination of Works of Art. Proceedings of the Seminar, 15-19 June 1970. Museum of Fine Arts, Boston, MA.
A review of the non-destructive identification of diverse geomaterials in the cultural heritage using different configurations of Raman spectroscopy D A V I D C. SMITH Museum National d'Histoire Naturelle, Laboratoire LEME, USM0205, 61 Rue Buffon, 75005 Paris, France (e-mail:
[email protected]) Abstract: Non-destructive Raman microscopy (RM) applied to geomaterials in the cultural
heritage is reviewed by means of explaining selected examples representative of the different kinds of geomaterials that can be characterized and of the different kinds of analytical configuration that can be employed. To explain the versatility and considerable analytical potential of RM that result from its unique combination of capabilities, the first sections summarize the theory and practice of the method and its advantages and disadvantages. The most modern configurations (mobile RM (MRM) and ultra-mobile RM; micromapping and imaging; telescopy) are described. Applications in the new age of 'don't move it, don't even touch it' archaeometry have previously been classified into 10 domains, seven of which concern geomaterials: gems; rocks; ceramics; corroded metals; coloured vitreous materials; and mineral pigments on an inorganic or organic substrate. The representative examples here include all these domains and cover the time range from Prehistoric through Egyptian, Roman, Meso-American, Medieval, Chinese, Renaissance and Mogul cultures to modern colouring of glass and a contemporaneous simulation of submarine archaeology.
The analysis of geomaterials in the cultural heritage, to clarify the nature of the material employed, evaluate possible provenances, detect treatments or to recognize fakes, calls for a variety of techniques, depending upon the type of material available and the kind of information sought. Raman microscopy (RM) (one kind of Raman spectroscopy (RS)) has become an important technique in archaeometric studies in archaeology and art history since about 1996, and the pseudo-acronym 'ARCHAEORAMAN' was coined by Smith & Edwards (1998) to summarize this wide field of research activity. More recently the term 'mobile Raman microscopy' (MRM) (Smith 1999) was employed to analyse art works in situ inside museums by taking the laboratory to the object, rather than the object to the laboratory as in conventional 'immobile Raman microscopy' (IRM). Subsequently, the possibility of using MRM for subaquatic archaeology was evaluated positively (Smith 2003), and more recently Raman micromapping has been used to clarify the microstructural mineralogy of artworks (Smith 2004a) or of rocks susceptible to be the provenance thereof (Smith 2004b,c). The most recent development in RS is telescopy (Sharma et al. 2002, 2003)
for very remote studies (such as planetology); this approach has not yet been applied to archaeology, but it could be useful for analysing gemstones in shop windows from across the street, which brings us into the domain of 'Raman spying' (Smith 2005a), and 21st-century social science, which will not be pursued here. Future developments will no doubt soon include synthetic vocal replies for automated analysis (Smith 2005a). In 1986, during a review of RM applications to mineralogy in general, Smith (1987) argued that RM should be of great value to archaeometry, but no significant studies were known to the geological community at that time, except for some pioneering studies on gemstones and their microinclusions (Drlr-Dubois et al. 1981a,b, 1986a,b). In fact, some chemists and physicists had already begun analysing artworks (Delhaye et al. 1985; Guineau 1987), but only pigments, and only publishing in journals in fields other than geology or mineralogy, especially chemistry or art history; furthermore, they generally avoided mineralogical terminology by using chemical names such as mercury sulphide or colour names such as vermilion instead of mineral names such as cinnabar. At the
From: MAGGETTI,M. & MESSlOA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 9-32. 0305-8719/06/$15.00 ~:) The Geological Society of London 2006.
iPrehistoric i
i
!Roman
Ukraine
water
4 x micro vertical immobile in lab x
5 x micro vertical immobile in lab x air ~
6 x micro vertical immobile in lab x air x
~
r/"c3
, __
7 x micro vertical immobile in lab x air x
8 x micro vertical immobile in lab x air
9 x micro vertical immobile in lab x air '
10 x nlcro ~ imTdoite in lab x air trap
11 x moo ~ irm'ebile in lab x air trap ~ |
12 Illl x rricro ~ m3bile in situ fibres ar x
~ /
....
i
i i
Barbet (1999)
i .
icorr~ed ic°ins (Fe' i& °ther !Cu' Zn Pb iartefacts in ;i
Smith (2001, 2005b)
(2001), Bouchard & Sn~th /2005b)
I
~
o ~
i I
~
(1999b); (2004b); (2004a) Bouchard & Smth Smith (2005c) /2002/
iglass
ipgrnentson i or n stained!
i
ia
_
'in
i axehead,
! Meso-
iA~ncan
(1°J99a); Robin (2001b, Gendron Vemioles Smith & (1997) 2003) (1997a), (1997) Bouchard Smith /2000al /2005cl Notes to tabular part: x denotes "no"; bold type in the configurations highlights unusual features.
references
! i
ivrmo,
iI~
i i i
',METALLOIRAMAN
- -
i
i
....................................................... ! .....
i
ievaluation IofRMo n igemstones ~underwater i
'PETRO.
iRoman iintaglios' .France
',~A~N
!GEMI~ RAI~N
(1999, 2900); Smth /2005at
!l~ished
am~c~
.
.......................................................................................... i..................................... I. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i lCONO-
i~s, Ro~, i
[pigments in !pigmantsin i
i RAMAN
3 x micro vertical immobile in lab x air x
iFRESCO-
2 yes micro vertical immobile in lab x air
1 yes micro vertical immobile in lab x air x
micro-extraction macro/micro horizontal/vertical mobile/immobile in situ/in lab optical fibre head under micro-mapping
i
(2001a, 2005a); Srffth elal, /2003a/
~
i
,
13 x x x mobile in situ til~'es air
Snith (2002); Srffth t2005a/
L
14 x x x mobile in situ fibres air
i
blue
17 x x x rn:Yoile in situ fibres air x
a ~oodm slalue,
!i i
i
16 x x x m:3bile in situ fit:~es gl~ss x
-
'
! ~
!
' ,
.- _
'~ ........................ JR/IM~
(2001a,b) S r ~ & references Lmblanchel (unpub. data/
. . . . . (2003b); F~ndeau ~?th (2001); (2005a) Srdth /20(35a/
!
i~"Y
i
~................................. }.............................................
cave,
Ii~a i
18 x r r i ~ x macro/rricro x ~ ~ u/~ra-].l~l,f rn~nn'doile in situ in situ/in lab all in one opticalfibrelead air under x nicro-map~ng i I:~ristmc i iPg~ , ~
~'-- ....................... ~ ........................~ i n c ~ i n irnarl~ MIdl:le/~ , . , ~ i s o r ~ i stone J
15 x x x rrd)ile in situ fibres air x
NON-DESTRUCTIVE RAMAN SPECTROSCOPY international GEORAMAN-1996 conference in Nantes an attempt was made to bring ARCHAEORAMAN topics to the attention of the geological community and since then contributions on archaeology and art history became significant at every G E O R A M A N meeting (1999 in Valladolid; 2002 in Prague; 2004 in Honolulu) (see table 8 of Smith & Carabatos-Nedrlec (2001) for a list of archaeological or art historical topics presented at these meetings). Another series of international congresses on nondestructive analysis in the cultural heritage brought in RM at Antwerp in 2002, and this continued at Lecce in 2005. The meetings of ICAM (International Congress on Applied Mineralogy), IRUG (InfraRed users Group), GFSV (Groupe Franqais de la Spectroscopie Vibrationelle), and GMPCA (Groupe des M~thodes Pluridisciplinaires Contribuant h l'Archrologie), and others, have started to include RM, as have other more archaeological meetings (e.g. Smith et al. 2000). A separate series of international congresses on exclusively 'Raman Spectroscopy applied to Archaeology and Art History' ('ArtRaman') was started in London in 2001 and continued in Ghent in 2003 and in Paris in 2005. The literature on A R C H A E O R A M A N has thus increased enormously in a decade, but it is dissipated amongst journals in many disciplines. This paper cannot review all the literature; it thus focuses on explaining why RM is so useful and describes a series of examples of studies by the author's research group that are in two ways representative: of the different kinds of geomaterials that can be analysed, and of the different kinds of analytical configuration that can be employed (Fig. 1).
What is Raman spectroscopy? RS is an optical, hence physical, technique by which the wavelength of light is modified by interactions with interatomic vibrations
11
(e.g. Smith & Carabatos-Nrdelec 2001; Nasdala et al. 2004). The modified light is called Raman diffused light according to the 'Raman effect' discovered by Sir Chandrasekhara Venkata Raman in 1928, for which he received the Nobel Prize for Physics. Thus the technique does not analyse a single atom, as do a great number of chemical analytical techniques such as X-ray fluorescence, as at least two atoms are required. The vibrational energies involved are the same as those in infrared (IR) spectroscopy, such that the two techniques are often considered similar. They are indeed complementary, but are not really similar, because in IR spectroscopy photons are absorbed or reflected according to the various vibrational energies, whereas in RS, incoming photons lose some energy, which leaves a vibration mode more excited, and hence the outcoming photons have lost some energy, i.e. they have a higher wavelength, and hence a lower wavenumber (the reciprocal of wavelength) (Fig. 2). This is called Raman Stokes scattering. Raman AntiStokes scattering also occurs whereby a vibration mode gives up some energy to become less excited and the outcoming photons have gained energy, i.e. they have a lower wavelength, and hence a higher wavenumber; this effect is weaker and will be ignored here. Thus with Raman Stokes scattering a single kind of interatomic vibration causes a shift of the wavenumber of the incoming exciting light, usually from a laser (although Raman used sunlight) and necessarily monochromatic. The exciting wavelength (e.g. 514.5nm from an Ar + green laser or 632.8 nm from a H e - N e red laser) is placed at zero cm - l on the relative wavenumber scale such that the Raman band created occurs at a characteristic Raman shift (e.g. 465 cm -1 from the major vibration of quartz). Raman shifts are conventionally plotted as being positive, as a shift is an amount without direction, but in reality it should be plotted as - 4 6 5 cm -1, as
Fig. 1. Representative examples of ARCHAEORAMAN studies on geomaterials: configurations, domains and images. Tabular part: configurations listed horizontally; domains listed diagonally; examples placed in the appropriate case. Arrowed superimposed images demonstrate the following selected cases. (a) Raman spectra from the Meso-American stone axe-head in eclogite; from top to bottom: titanite, garnet, clinoamphibole, clinopyroxene (modified after Smith & Gendron 1997a). (b) A Domitian denier silver alloy coin with cuprite corrosion (modified after Bouchard & Smith 2005b). (e) Raman spectra of microcline under air, distilled water and water badly contaminated by animal or vegetable debris as a simulation of subaquatic archaeology (modified after Smith 2003). (d) A Meso-American corroded metal axe-head (modified after Bouchard & Smith 2005b). (e) An Egyptian inscribed commemorative scarab in polycrystalline enstatite established by Raman mapping with a RENISHAW® Invia® spectrometer (modified after Smith 2004a). (f) A Chinese sculptured pendant in jadeite-jade (modified after Smith 2005a). (g) A Medieval cloisonnr-gold style fibula encrusted with garnets (photo D. C. Smith'S). (h) A Teotihuac~in sculptured mask in marble with the DILOR® LabRaman® horizontal microscope (modified after Nasdala et al. 2004). (i) A Florentine table in stone marquetry being analysed vertically with a KAISER® Holoprobe® remote head through the thick protective plate glass (invisible here) (modified after Smith 2005a).
D. C. S M I T H elastic
inelastic
inelastic
scattering
scattering
scattering
Rayleigh diffusion
Raman Stokes diffusion
Raman Anti-Stokes diffusion
Ee + E i +E,
vibrational
virtual state
E~+E,
Ee
excited state
Es
~round state
energy levels
extra relations
input (i) energy
E,
E,
Ei
frequency
v, = E , / h
v, = E i / h
vi = E i ] h
= h. v i = c / Li
wavelength
k~ = c /v,
~.~= c / v ,
k, = c / v ,
= h . c / Ei
wavenumber (absolute) wavenumber (relative)
W, = 1 / k,
W i = 1 / k~ W, set at zero
W, = 1 / k i W, set at zero
= v, / c
change to sample change to light
none
(EcEg) gained
(E~-Eg) lost
none
(Ee-Eg) lost
(Ee-Eg) gained
output (o~ energy
Eo = Ei
Eo = Ei - ( E e-E~)
E o = E, + ( E e - E 0
frequency
vo = E o / h
vo= E o/ h
vo = E o / h
wavelength wavenumber (absolute)
k o = c / Vo Wo= I /~o
~.o = c / v o W o = ! /ko
ko = c / v o Wo= 1 /k o higher
comparison: input to output
none
lower
wavenumber (relative)
Wr = W, - W o
Wr = W o - W 1
Wr= W 0- W
= Raman shift
zero
negative
positive
c = speed o f light; h = Planck's constant Fig. 2. S c h e m e o f the different ways in which inter-atomic vibrational energy levels give rise to three types o f light scattering (diffusion): Rayleigh, R a m a n Stokes, R a m a n Anti-Stokes.
t h e a b s o l u t e w a v e n u m b e r is l o w e r t h a n t h a t o f t h e e x c i t i n g l i n e ( F i g . 2). A n i m p o r t a n t p o i n t is that Raman shifts are constant for any wavel e n g t h o f t h e e x c i t i n g l a s e r as t h e s h i f t s a r e f i x e d r e l a t i v e to t h a t w a v e l e n g t h a n d a r e l i n e a r in c m - ~ ; v e r y f e w e x c e p t i o n s to this r u l e o c c u r (e.g. t h e D b a n d o f g r a p h i t e ) . As there are many different kinds of v i b r a t i o n a l s y m m e t r y , e a c h w i t h its o w n e n e r g y
l e v e l (e.g. s y m m e t r i c s t r e t c h i n g , a n t i - s y m m e t r i c stretching, deformation, bending, rocking, w a g g i n g , t w i s t i n g ) , a n d all o f this f o r e a c h k i n d of combination of chemical elements (depending upon the Raman 'selection rules', which depend upon the crystal or molecular symmetry and also u p o n the n u m b e r o f c h e m i c a l e l e m e n t s present), there are several distinct Raman bands created (which occasionally overlap) such that
NON-DESTRUCTIVE RAMAN SPECTROSCOPY a spectrum is obtained (where the ordinate shows photon intensity, and the abscissa shows the wavenumber) (Fig. l a and c). According to these rules, some materials give only one band (e.g. diamond), simple carbonates and sulphates give fewer than 10 bands, silicates such as garnet give about 20, and more complex silicates such as micas and amphiboles may give more. Organic molecules may give rise to hundreds of bands. Spectra are variably plotted with the zero at the left or the right, but the zero is never plotted as this is where Rayleigh scattering occurs; this involves the restitution of the exciting light with the same wavelength (Fig. 2) (physically not the same, but effectively the same as simple reflection). Rayleigh scattering is very approximately 1012 times more efficient than Raman scattering and this important fact has several consequences: (1) a Raman spectrum cannot show the intensity at 0 c m - i as it would plot somewhere in interplanetary space; (2) it would burn the detector, or create a plasma from it, and has to be filtered out; (3) a 'Rayleigh tail' occurs in the 10100 cm -1 spectral range where the Rayleigh scattering intensity decreases to zero; (4) only about one photon in several billion incoming photons is subject to the Raman effect, so the development of RM necessitated strong laser sources and powerful detectors of very weak signals as well as coupling to a microscope (Dhamelincourt & Bisson 1977); (5) commonly 1 - 1 0 0 mW power is used to analyse a 1 Ixm sized portion of a sample or an art object; if the same power per ixm2 were applied over a 1 m 2 surface it would need 1012 times more power, i.e. 1 - 1 0 0 GW, which brings us to the scale of several nuclear power stations (and this ignores the third dimension and another 106). Thus we are dealing with an extremely powerful energy applied to an extremely small location to detect an extremely weak effect. It is important to appreciate that the intensity of a Raman band of a crystal depends, often strongly, on the orientation of its crystal symmetry with respect to the polarization of the laser (compare X-ray diffraction) such that in certain situations a Raman band may disappear completely. If it is not possible to rotate either the art work or the RM, one can introduce a half-wave plate and rotate it to see the missing band (Smith 1996). There are basically two ways of using RS. One approach uses RS to satisfy the chemist's, physicist's or mineral physicist's need to try to predict and calculate Raman phenomena and to extract thermodynamical data, often by measuring Raman spectra at high or low
13
temperature (T) and/or high or low pressure (P); this is not discussed further here. The second is to use 'Raman spectral fingerprinting' (Dhamelincourt & Bisson 1977; Smith 1987) to identify mineral or molecular species, as different species cannot give the same spectrum and the same species will always give the same spectrum (at the same P - T, if there are no differences in chemical composition, crystal structural order, etc.). This of course requires spectral databases; several now exist, but all are limited in scope (see White 1975; Griffith 1987; Guineau 1987; Pinet et al. 1992; Bell et al. 1997; Burgio & Clark 2001; Bouchard & Smith 2003, 2005a) and numerous others are in preparation as every Raman research group builds its own.
Why has Raman microscopy become so polyvalent and powerful? This is principally because of its great versatility owing to its unique combination of capabilities, as follows. (1) It characterizes simultaneously the physical structure and the chemical composition of an unknown species by comparison of its Raman spectrum with reference spectra (compare IR and X-ray diffraction (XRD)). This is extremely useful for distinguishing polymorphs such as quartz-moganitetridymite-cristobalite-coesite (SiO2), aragonitecalcite (CaCO3), sanidine-orthoclase-microcline (KA1SiO3), rutile-anatase-brookite (TiO2), etc., which cannot be done with any purely chemical technique. (2) It can do this with inorganic or organic material in different states or forms, such as crystalline, molecular, glassy or amorphous; whether solid, powdery, suspended, plastic, vitreous, liquid or gaseous; and whether pure or mixed. Apparently only IR can also do this. Mixed phases, such as in a pigment or in sub-micronsized mineral intergrowths in a rock, gem or ceramic, are commonly encountered in archaeometry. (3) The analysed volume may be on a micrometre scale, from about 0.5 p~m to about 50 l~m in surface diameter, commonly 1 - 2 txm, but the analysed object may have any size and different parts thereof may be systematically analysed. IR and XRD cannot do this except with a synchrotron (which must be the least mobile analytical apparatus). (4) No sample preparation whatsoever is required (no extracting, drilling, scraping, sawing, cutting, grinding, polishing, liquefaction, gasification, etc., nor a vacuum chamber,
14
D.C. SMITH
KBr pelleting, or other kind of processing) as the method is non-destructive; with an appropriate reflection configuration IR can also be non-destructive. This non-destructive property is true as long as one maintains a laser power sufficiently low to avoid damage; if, unfortunately, this is not achieved then a micron-sized volume of the analysed object may be 'burned' or otherwise disintegrated, but fortunately this will be invisible to the naked eye and harmless to most materials such as gemstones, although it could become dangerous for inflammable materials such as the paper of a priceless ancient book. (5) RM can provide micro-mapping or microimagery of textures of intergrown phases, of chemically zoned crystals or of physically deformed crystals. Other techniques can map structures, but IR is on a larger scale. (6) With the use of mobile optical fibres one can analyse any part of an artefact (including re-entrant angles such as under the arm of a statue or gemstones mounted inside a crown). (7) MRM may be carried out almost anywhere, such as in situ inside a museum display cabinet, a conservation or storage building, or on an archaeological site. (8) One can identify a phase under another transparent one, such as microinclusions inside a mineral, as well as pigments under glass, gems under plastic, or statues under water, so that submarine archaeology by MRM has become possible (Smith 2003). (9) One can obtain semi-quantitative chemical analysis of mineral solid-solutions by RM for example, by using the RAMANITA method devised by Smith & Pinet (1989), calibrated by Pinet & Smith (1993, 1994) and updated by Smith (2002b, 2004d, 2005b). The method is based on the time-consuming calibration of wavenumber shifts along each binary join (if natural or synthetic samples are available) and then within various choices of multivariant chemical space. All these possibilities and developments led Smith (2002a) to declare that 'The new age of "don't move it, don't even touch it" archaeometry has now arrived to allow remote non-destructive characterisation in all the domains of ARCHAEORAMAN and in situ almost anywhere'.
What disadvantages exist with Raman microscopy? As with all analytical techniques there are some disadvantages with RM, but they are small in number compared with the advantages. Very
few minerals give no Raman band at all because they have a high symmetry and a low number of different atoms in the unit cell (e.g. halite (NaCI)). Most pure metals give no Raman signal, partly for the preceding reason, and partly because of their high reflectivity; on the other hand, as soon as a metal is corroded to form oxides, hydroxides, carbonates, sulphates, chlorides, etc., RM works very well. Opaque or semi-opaque minerals absorb too much light and give either no Raman signal or a very weak one; manganese oxyhydroxides are a good example as they have been difficult to recognize in pigments; however, with more recent instrumentation one can now obtain Raman spectra from many of these phases (Ospitali & Smith 2005). Some materials are rather photosensitive and need low laser power to avoid instantaneous dehydration (e.g. iron hydroxides and lead hydroxides). The detector picks up not only the Raman signal but also various kinds of 'parasite' signals, such as laser lines from the laser source that have not been sufficiently well filtered, cosmic rays, daylight, incandescent room light, Hg and Ne emissions in common neon 'fluorescent tube' lamps, photoluminescence (PL) from chemical impurities in the sample or in the optical trajectory (e.g. the infamous 843 c m - I band from the Olympus × 50 objective), or fluorescence. These sources can be attenuated by laser filters, by reducing daylight or room light, or by changing the exciting laser wavelength such that photoluminescence lines occur elsewhere in the spectrum. Background fluorescence, which gives a very high baseline that partially or totally obscures the Raman spectrum, is no doubt the worst problem, but its true cause is not always obvious. It is known that it can come from electronic transitions in imperfectly crystallized minerals, from some nanocrystalline materials such as clays, and from mixed organic materials (living or dead). If the parasite does not interfere in the same spectral range as relevant Raman bands then the problem is avoided. Waiting a few minutes before acquiring spectra usually reduces the baseline, perhaps as a result of some annealing by heating. Analysing under water is beneficial (Smith et al. 1999a; Smith 2003). Changing the exciting laser wavelength often (but not always) creates drastic improvements. Pulsing the laser is an excellent antidote but it is not easy to acquire the necessary configuration. Interchanging a troublesome optical component (e.g. filter, mirror, objective) in the RM system with one of a different kind will cure the problem in some cases.
NON-DESTRUCTIVE RAMAN SPECTROSCOPY Raman spectra frequently need some amount of spectral treatment if we are to be able to exploit the data by spectral fingerprinting (on the other hand, treatments are usually avoided for thermodynamical studies as one must not modify the raw data upon which certain calculations are based). First, the 'baseline correction' tries to make the background line horizontal, regardless of the cause of it not being flat (fluorescence, luminescence); this procedure can dramatically increase signal-to-noise visibility. Subtracting an oblique straight line is acceptable if the baseline has a sub-linear steep slope, but often it is necessary to subtract a polynomial 'best line' curve calculated from selected landmarks on a distinctly curved baseline. Automatic correction can be disastrous as the computer program may confuse wide Raman bands with an undulating baseline. More than a × 2 polynomial can produce major distortions and, in any case, it is not necessary to achieve a perfectly flat baseline. Second, one may eliminate known parasite peaks or known detector defects by 'rubbing out' with the computer mouse instead of a piece of rubber. An automatic 'peak elimination' procedure may be useful for eliminating narrow cosmic rays that are distinctly narrower than Raman bands, but it needs to be used with care. Third, 'smoothing' by averaging all intensities over a selected small wavenumber zone is very useful to make real Raman bands more visible by eliminating the basic zigzags of the irreducible background flutter, but must not be done over zones too wide or real Raman bands will become too diluted in intensity or separate nearby bands (doublets) may become fused together. With these three treatments one can frequently transform apparently hopeless spectra into perfectly exploitable ones, and this is because the basic information exists in the raw spectrum and it just needs to be rendered visible. A variety of more sophisticated computerized techniques exist, such as spectral combination, peak-fitting, Fourier transforms and 3D-plotting, but they will not be dealt with further here.
Classifications of Raman microscopic studies of the cultural heritage To demonstrate applications of RM to the cultural heritage it is convenient to classify the examples according to some criteria. Here the cultural period (Prehistoric, Roman, Medieval, Renaissance, etc.) is not used as this paper is more mineralogical-technological than archaeological. The type of material analysed can be a
15
Table 1. The 10 domains of ARCHAEORAMAN, updated from Smith (1999, 2002a) (1) GEMMORAMAN from 'gems': gemstones (rough, cut or mounted), cameos, corals, intaglios, jewellery, collection stones, etc. (2) CERAMIRAMAN from 'ceramics': brick, china, earthenware, faience, glass, porcelain, pottery, slags, tiles, other vitrified minerals, etc. (3) PETRORAMAN from 'petros' for rocks: axeheads, building columns, ceremonial stones, inlaid rock, millstones, mosaics, necklaces, sculptures, etc. (4) METALLORAMAN from 'corroded metals': corroded bracelets, coins, cutlery, necklaces, statues, swords, tools, etc. (5) RESINORAMAN from 'resin' as an example of a non-cellular organic material composed of only a few different molecules or of amorphous hydrocarbons without a growth texture: amber, glue, gum, oil, putty, wax, bitumen, lignite, coal, etc. (6) TISSUERAMAN from 'tissue' as an example of cellular organic molecules or biominerals with a growth texture: bone, claw, cotton, feather, fur, hair, hoof, horn, ivory, leather, linen, nail, papyrus, parchment, silk, skin, teeth, wool, wood, etc. (7) FRESCORAMAN from 'fresco' as an example of pigments/inks/dyes on or in an inorganic substrate: brick, ceramic, plaster, stone, stucco, etc. (8) ICONORAMAN from 'icon' as an example of pigments/inks/dyes on or in an organic substrate: bone, canvas, paper, skin, textile, wood, etc. (9) VITRORAMAN from the 'vitreous' state: pigments on or in enamel, glass or glaze, etc. (10) ENVIRORAMAN from 'environmental' deterioration of any of these materials by climate, burial or immersion: original materials, corrosive agents involved, intermediate and final products
useful criterion, and this was used by Smith & Edwards (1998) as there may be different analytical protocols for different materials; Table 1 lists the 10 domains of research activity as updated by Smith (2002a). The analytical configuration employed is also relevant (macro or micro; vertical or horizontal microscope; optical fibres or not; mobile or immobile; in situ or in a laboratory; under air, glass, mineral, or water). Figure 1 plots the seven domains relevant to geomaterials against combinations of analytical configurations and lists the studies (by the author's research group) that are mentioned here as being representative of research in A R C H A E O R A M A N in general.
16
D.C. SMITH
RM analysis of pigments, whether inorganic or organic materials on inorganic (FRESCORAMAN) or organic (ICONORAMAN) substrates has dominated ARCHAEORAMAN from the early works of Delhaye et al. (1985) and Guineau (1987) to the production of minicatalogues of Raman spectra of pigments (Bell et al. 1997; Burgio & Clark 2001), and from applications to prehistoric rock art (e.g. Bouchard 1998, 2001; Edwards et al. 1998; Smith et aL 1999a,b; Smith & Bouchard 2000a) via Roman art (e.g. Smith & Barbet 1999) through various periods of the last millennium (e.g. Rull-Perez et al. 1999; Withnall 1999; Rull-Perez 2001) to modem art (e.g. Vandenabeele et al. 2000). The biomaterials domains RESINORAMAN and TISSUERAMAN have been mainly limited to specialists in biology and/or organic chemistry (e.g. the early works of Edwards et ai. 1996a,b,c; Brody et al. 1998). Turning to geomaterials, the earliest known work was on GEMMORAMAN (D~lr-Dubois et al. 1978). The advantages for gemmology are considerable, as RM can be employed for several different purposes: to verify the nature of the gemstone itself, to examine for treatments (e.g. heating, resin impregnation, pigmentation), to explore solid or fluid microinclusions, or to detect synthetic and imitation stones. Certain aspects of gemmology have been studied in detail by RM by Lasnier (1989) and Maestrati (1989), and the first catalogue of the Raman spectra of gemstones was published by Pinet et al. (1992); more recent studies have been made notably by Coupry & Brissaud (1996), Schmetzer et al. (1997), Smith & Robin (1997), Smith & Bouchard (2000b), Kiefert et al. (2001) and Smith (2001a, 2005a). Apart from extremely few early works (Coupry et al. 1993; Macquet 1994; Wang et al. 1995), 1997 saw the effective beginning of RM studies in the remaining four geomaterial domains, in particular: (1) PETRORAMAN of jade and eclogite by Smith & Gendron (1997a,b) or of sculptured polished ceremonial rocks by Smith & Bouchard (2000b) and Smith (2005a); (2) CERAMIRAMAN of vitrified forts by Smith & Vernioles (1997), of the minerals constituting pottery by Fry et al. (1998) or of the pigments in glazes by Colomban & Treppoz (2001), Colomban et al. (2001) and Liem et al. (2000, 2002); (3) METALLORAMAN on corroded metal coins and various archaeological metals (Fig. l b and d) by McCann et al. (1999), Bouchard & Smith (2000a,b, 2001, 2005a,b), Bouchard (2001), Di Lonardo et al. (2002), Frost et al. (2002a,b),
Smith & Bouchard (2002) and Martens et al. (2003); (4) VITRORAMAN on the minerals colouring stained glass by Edwards & Tait (1998), Smith et al. (1999c) and Bouchard & Smith (2002). ENVIRORAMAN studies are less common (e.g. Seaward & Edwards 1998). The RM spectral catalogues of Bouchard & Smith (2003, 2005a) included minerals of relevance to prehistoric paintings, corroded metals and stained glass. Probably at least 80% of all ARCHAEORAMAN publications to date concern pigments. Apparently over 90% of all RM analysts are physicists or chemists, which is logical given the physico-chemical basis of the technique. Thus, like botanists and zoologists, geologists of one kind or another (e.g. crystal chemists, mineralogists or petrographers) engaged in archaeometry via RM make up a very small community worldwide. However, each specialist brings his own particular competence and, similar to the need for an experienced botanist to identify a kind of tree, geologists are clearly necessary when studying natural rock artefacts from the cultural heritage (and solid-solutions, microinclusions, transformations, etc. in their constituent minerals, and their possible provenance in one or other geological unit). It was argued by Smith & Edwards (1998) that ARCHAEORAMAN studies really require three co-authors, a spectroscopist for the analysis, a natural scientist for the species of the natural raw material, and a social scientist for the artefact (form and cultural context). Individual scientists can often manage to adequately cover two of these disciplines, but to cover all three properly (or all five if one separates geology, botany and zoology) would be utopia, surely requiring a born-again Leonardo da Vinci. The following sections, organized by analytical configuration, focus on the geomaterials applications listed in Table 1.
Representative examples of RM applications I m m o b i l e a n a l y s i s in a l a b o r a t o r y : u n d e r air with a vertical microscope
This is the standard method of performing archaeometry with RM, either by placing on the microscope stage micro-samples extracted from a cultural item (i.e. not strictly non-destructive in this case) or by placing the whole item under the microscope if it is small enough to squeeze between the objective and the stage, or by taking away the stage. Methods.
NON-DESTRUCTIVE RAMAN SPECTROSCOPY
i
Wavenumber (cm -I)
500
1000
~
1500
Wavenumber ( cm- I )
400
17
600
800
1000
(c)
....
oo
,ooo
'z°° (d)
Fig. 3. Raman spectra of selected subjects. (a) Raman spectra of pigments from the Roman tomb at Kertsch (Ukraine): red minium (BJMI55yy, bottom left); blue cuprorivaite (BHCV03zz, top left); black carbon (BHCA21hh, right) (modified from Smith & Barbet 1999). Int, intensity. Eight-digit codenames are the computer spectra filenames. (b) Raman spectra of minerals from the new type of jadeite-jade from Guatemala: from top to bottom: jadeite alone (AHCP03 mm); jadeite + quartz (key peak at 468 cm- 1, AJQZ05 mm); jadeite + rutile (key peak at 445 cm- ~, AGCP22 mm); jadeite + titanite (key peak at 543 cm- 1, AHUN 16 ram). Some of the key peaks of jadeite are present in all spectra: 203, 373,698, 986, 1039 cm -1 (modified from Gendron et al. 2002). (e) Raman spectra of Cu-hydroxysulphates, from top to bottom: archaeological brochantite (DGCU 17je); standard brochantite (BSCUO6je); archaeological antlerite (CRCU08je); standard antlerite (BOCU08je). Some bands are at the same wavenumber in all spectra but there are significant shifts between the two species, notably the SO]- vibration just below 1000 cm-l (modified from Bouchard & Smith 2005b). (d) Raman spectra of the interior of two modem glasses: colourless 'verre cord616' (top, BUVE071f) showing intense bands revealing a high Na content (573 cm-i) and a tectosilicate Si-O arrangement (1100 cm- 1); red 'verre antique' (bottom, BQCO04jv) dominated by the bands at 195 cm-1 (CdSe) & 288 c m - l (CdS) characteristic of CdSo.g5Seo.55(modified from Bouchard & Smith 2005b).
A D I L O R ® X Y ® spectrometer belonging to the M u s e u m National d'Histoire Naturelle (MNHN) was employed.
Pigments: Roman wall-paintings. Black, red and blue are the major colours in decorations on a wall-painted Roman tomb at Kertsch, Ukraine. Micro-samples more or less invisible to the naked eye were extracted by the archaeologist A. Barbet and submitted to RM examination. It was easy to focus the 1 - 2 tzm diameter laser beam onto any selected mineral grain or part of a composite micro-assemblage to determine its mineral constitution (Smith & Barbet 1999). In this way it was found that the black is semi-amorphous carbon (C) (Fig. 3a);
this is a very c o m m o n phase in all cultures (often called 'carbon black', but such varietal names are not always used with precision) and it was probably the first pigment ever used by mankind. The blue pigmentation derived from cuprorivaite (CaCuSi4Olo) (Fig. 3a), which is the key constituent in the pigment called 'Egyptian Blue' and which was widely used in the Roman Empire. The red turned out to be m i n i u m (Pb304) (Fig. 3a); although k n o w n elsewhere in the R o m a n Empire it was not previously k n o w n as far NE as Kertsch.
Pigments: Prehistoric cave paintings. Although RM work on pigments had begun in the mid1980s, it was not until the late 1990s that R M
18
D.C. SMITH
analysis of Prehistoric pigments from surface rock art (Edwards et al. 1998; Smith et al. 1999b) or cave wall-paintings (Smith et al. 1999a; Smith & Bouchard 2000a; Bouchard 2001) was attempted. Prehistoric pigments are, in general, far more difficult to determine than pigments from historical times. This is not only because they tend to give an enormous fluorescence but also because the three most common phases used, other than carbon black, each have an additional problem. Thus yellow goethite (a-FeO(OH)) rapidly dehydrates to form red hematite (a-Fe203) even at very low laser power; red hematite strongly absorbs a green laser beam, overheats and decomposes into a black dot that might contain magnetite (Fe304); black MnxOvOHz phases absorb so much light that they give particularly bad Raman spectra. In the case of the limestone caves Pergouset, Les Merveilles and Les Fieux, in the Quercy district, Lot, France, it was possible to identify on various drawings (lines, dots, negative hands, etc.) predominant hematite with minor goethite in the red colours, and carbon in most black parts. Some other black parts were not of carbon and did not give a Raman signal until the micro-fragments were covered with water to keep them cool (Smith et aL 1999a). The Raman signal obtained resembled that of bixbyite (Mn203), a rare species in nature. This raised the question of the possible creation of bixbyite by heating some other MnxOyOHz phase, either by prehistoric man or by the laser beam during the analysis. Using more recent Raman apparatus, better spectra from some MnxOyOH: phases have been obtained both from samples in the MNHN mineral collection (Ospitali & Smith 2005) and from other limestone caves in Quercy (Roucadour and Combe N~gre 1) (Ospitali et al. 2005). Thus it is now easier to distinguish carbon from MnxOyOHz, which helps enormously in deciding which drawings to sacrifice for carbon isotope dating. A spectrum of an interesting orange microphase was obtained at Pergouset, which is neither goethite nor hematite because of a strong band at precisely 400 cm -j that lies between the values for well-crystallized goethite or wellcrystallized hematite; it was called 'disordered goethite' as it shared several bands with goethite (Smith et al. 1999a) and was probably created by prehistoric man heating 'yellow ochre' (a mixture coloured by goethite). G e m s t o n e s : R o m a n intaglios. Gemstone identification is one of the applications where RM excels. Three small intaglios were excavated
from a Roman site at Lut~ce (Paris) by the archaeologist S. Robin. When they were studied on a microscope stage it was rapidly established by RM that they were all composed of quartz (SiO2) (Smith & Robin 1997). The texture under the microscope indicated polycrystalline quartz, i.e. chalcedony, but this mineral has a great number of varieties. Two intaglios were apple-green in colour and it was first thought that they were of chrysoprase, a variety coloured green by Ni. Subsequently, some other green chalcedonies in other rocks were shown to be green because of Cr and have been called Cr-onyx. Because RM does not detect trace elements, as about 1 atomic % of an element is necessary to create a detectable spectral difference, the naming of the mineral variety of these intaglios could not be established with confidence, but the mineral species was unequivocal. One of them had a small mineral inclusion, which turned out to be zircon (ZrSiO4). The third intaglio was metallic blue under reflected light but bordeaux red under transmitted light; RM showed that this was also of quartz; its variety name could be jasper or sard.
Rocks: M e s o - A m e r i c a n axe. A Meso-American polished axe-head from Cozumel Island, Mexico, now in the collection of the Musre de l'Homme, Paris (Gendron 1998), had previously been classified as a 'greenstone', which literally means a green rock that has not been identified. This one contained at least two reddish minerals as well as two greenish minerals. With RM four kinds of Raman spectra were obtained and identified as clinopyroxene ((Na,Ca)(A1, Fe3+,Mg,Fe2+)Si206) (green), 3clinoam~hibole ((D,K,Na)(Na,Ca)z(A1,Fe ,Mg,Fe )5 (Si,AI)sOzz(OH)2) (darker green), garnet ((Mg,Mn,Fe~+,Ca)3(A1,Cr,Fe3+)2Si3012) (red) and titanite (CaTiSiOs) (brown) (Fig. la) (Smith & Gendron 1997a). The positions of the T - O - T bands of the clinopyroxene and the SiO4 bands of the garnet implied considerable proportions of respectively jadeite (NaA1Si2Or) and pyrope (Mg3A12Si30~2) in solid-solution, based on the semi-quantitative analytical method RAMANITA (mentioned above) (Smith 2005b). These two species indicated an eclogite, a rock type in which clinoamphibole and titanite often occur (Smith 1988). The kinds of clinoamphibole cannot be established as there are over 50 amphibole end-members and relatively few published data on their Raman spectra. Eclogite does not occur geologically on Cozumel Island, thus proving its
NON-DESTRUCTIVE RAMAN SPECTROSCOPY transport from afar, possibly from Guatemala (McBirney et al. 1987). Rocks: jades. A second axe head, from Guatemala but of uncertain provenance, was shown to be a true jadeite-jade by comparison with the Raman spectrum of a Burmese jadeite-jade (Smith & Gendron 1997a). Indeed, RM is undoubtedly the best technique for rapidly and non-destructively distinguishing the three types of jade: jadeite-jade (clinopyroxene); nephrite jade (clinoamphibole close to the tremoliteactinolite series (Ca2(Mg,Fe)5(Si)8022(OH)2)) and 'tourist jade' (anything else) (Smith 2005c). Thanks to RM, a fiver pebble subsequently collected by the archaeologist F. Gendron was shown to be a new sub-type of jade (quartzjadeitite) composed also of rutile (TiO2) and titanite (CaTiSiOs) (Fig. 3b) formed at higher pressure than usual Meso-American jade (albite-jadeitite) (Smith & Gendron 1997b; Gendron et al. 2002) from a new locality, on the south side of the Motagua River Valley, whereas all previous findings of geological jade had come from the north side (Harlow 1994). The light greyish-green 'type' jadeite in the MNHN mineral collection, which is itself a Neolithic jade axe whose provenance was most probably in the Western Italian Alps, as well as a strong green 'chromo-jadeite' from Burma both gave typical spectra of jadeite (NaA1Si206) with > 9 0 mol% Jd characterized by the S i - O Si Raman band at 701 -t- 2 c m - l (Gendron et al. 2002; Smith 2005c). The singlet (OH) Raman vibration of nephrite at c. 3673 cm-1 is very useful proof of the presence of nephrite jade, when found in addition to the lower wavenumber of the S i - O - S i stretching vibration close to 675 c m - l , which is much lower than that in jadeite. The nephrite jade nature of a series of artefacts, mainly polished flat bracelets or rings, but also some geological source rocks, all from China, was analysed by Smith et al. (2003b). Many were found to be of nephrite, but one of the six source rocks was a serpentine (Mg3SizOs(OH)4),and three artefacts were not nephrite but either calcite or quartz. A few darker artefacts revealed only a weak band at about 675 cm-~ suggestive of nephrite. Two probable tourist jades from SE Asia were also examined: a supposed sculptured 'lilac jade' was only quartz with a colour between that of amethyst and 'rose quartz', and a green and white bracelet of supposed jade turned out to be of calcite (Smith 2005a). Ceramics: vitrified wall. Enigmatic vitrified forts are known throughout the c. 1000 BC to
19
c. AD 1000 Celtic world from Portugal to Sweden, and especially in Ireland and Scotland (Ralston 1983; Buchsenschutz et al. 1998; Kresten et al. 1998). They have in common the fact that stone building blocks at the lower levels are often found to have been fused together by melting. Whether fused for defence, by attack or for religious reasons, a second major archaeological problem is to elucidate how such high temperatures were achieved, and over long surfaces (e.g. 100 m) and sometimes several centimetres depth. A few fragments of vitrified wall were collected by the archaeologist J. Vernioles from the vitrified base of the frequently rebuilt fort at St. Suzanne, Mayenne, France. Amongst glass, some crystals were shown by RM to be of e~-cristobalite (SiO2), which is supposed to require a temperature of 1470 °C if created by cooling from [3-cristobalite (Smith & Vernioles 1997). There is considerable doubt over the real temperature achieved, as the literature on this topic is poor and sometimes contradictory, and the presence of other elements such as A1 or Na could reduce this temperature; furthermore, polymorphic and order-disorder phenomena are also relevant, as metastable forms of c~- and [3-tridymite and or- and [3-cristobalite can exist. Nevertheless, the temperature must have been high (at some other localities quartzite has been melted (P. Kresten, pers. comm.) and pure quartz melts at 1713 °C). This enigma, strangely unheard of by many archaeologists, is likely to remain a mystery for some time. Chemists have confirmed that wood smouldering during rain could produce gaseous unsaturated hydrocarbons (e.g. acetylene), which could migrate and burst into flame at extremely high temperature, but the energy available would not be sufficient to penetrate deep into the rock wall. Accumulated lightning strikes over a few millennia provide an alternative possible explanation, otherwise it might be necessary to invoke UFOs (unidentified flying objects)! Interestingly, identical Raman spectra were obtained (Smith & Vernioles 1997) from c~-cristobalite in glass in 'Libyan Desert Glass', which is believed to have been formed by some kind of extra terrestrial impact event. Corroded metals: copper coins. Coins constitute one of the most obvious kinds of metal artefact of the cultural heritage and their size is ideal for being placed under a fixed microscope objective. Coins from different periods and composed of various metals (Fe, Cu, Zn, Pb, Ag, A1, Ni, Sn) were thus examined by Bouchard (2001) and Bouchard & Smith (2001, 2005b). As mentioned above, the pure metal or even many alloys do not
20
D.C. SMITH
give a Raman signal, but their corrosion products do. Care must be exercised in interpretation, as the metal in an identified corrosion product may not be a major constituent of the original coin because of 'preferential corrosion'; thus Cu salts are often found on Ag coins that contain a small amount of Cu (Fig. l b). Hence the main purpose is to recognize the kind of corrosion process that has occurred, so as to help restorers and curators decide on the appropriate method to treat and conserve the coins (or tools, weapons or statues, etc.) (Fig. ld). Concerning copper, the products observed by RM on coins and other artefacts of various ages included Cu-oxides (cuprite (Cu20), tenorite (CuO)), Cu-hydroxycarbonates (azurite (Cu3(CO3)2(OH)2), malachite (Cu2CO3(OH)2)), Cu-hydroxychlorides (atacamite (Cu2CI(OH)3), clinoatacamite (Cu2CI(OH)3)) and Cuhydroxysulphates (antlerite (Cu3SO4(OH)4), brochantite (Cu4SO4(OH)6))(Fig. 3c). Particular attention was paid to the RM distinction of the Cu-hydroxychlorides by Bouchard (2001) and Bouchard & Smith (2005b), as clinoatacamite has only recently been recognized by the International Mineralogical Association (IMA) (Jambor et al. 1996) and in earlier works this mineral species may have inadvertently been called paratacamite, which is a C u - Z n solidsolution ((Cu,Zn)2CI(OH)3)). Corroded metals: lead plates and an iron ingot. A fragment of a strongly corroded Roman sarcophagus in lead is archived in the MNHN mineral collection and is labelled 'cotunnite', which thus indicates corrosion by chloride. A R M study found no cotunnite (PbC12) nor any other chloride, but only a mixture of several Pbhydroxycarbonates: plumbonacrite (Pblo(CO3)6 O(OH)6 ), hydrocerussite (Pb3(CO3)z(OH)2) and cerussite (PbCO3) (Bouchard 2001; Bouchard & Smith 2005b). Another similar plate revealed only the two oxides litharge (PbO) and massicot (PbO), hence only the valency Pb 2+ (as no minium (Pb304) or plattnerite (PbO2) was detected) and no chloride, carbonate or hydroxide. A Roman ingot brought up from a shipwreck at Sainte-Marie-de-la-Mer off the French coast was examined (Bouchard 2001; Bouchard & Smith 2005b). The minerals found on the highly corroded surface included the Fe-oxide maghemite (~/-Fe203) and the Fe-oxyhydroxides akaganrite, goethite and lepidocrocite (all (FeO(OH))). There were also RM spectral indications of the presence of the ion FeCI42-, and it is known that in akaganrite some (OH)may be replaced by C I - , especially in marine environments (Arnould-Pernot et al. 1994).
Stained glass: experimental, modem and archaeological. Glass can be coloured in various ways. The colour may derive from a single chemical element dissolved in trace amounts inside the glass, in which case there is no longer any crystalline mineral phase left to provide a Raman spectrum. Alternatively, there may be micro- or nano-crystalline inclusions, which can provide a Raman spectrum. However, the most common situation in stained glass in church windows (apart from unheated superficial paint) is coloured reaction products formed after pigment minerals (with or without fluxes such as minium and silica to produce a P b - S i - O glass) had been spread on the surface of the glass and then heated; this produces several distinct phenomena: dissolution of some original material into the glass; migration of certain elements from the glass onto the surface (especially alkalis and alkaline earths); intercrystalline reaction between the applied pigments with or without involvement of the glass; relict original pigment; or glass that did not react at all. A project involving the study of commercially available mineral pigments (whose precise chemical composition is not provided by the manufacturers), experimentation to create stained glass and to study the reaction products, and analysis of real archaeological stained glass from the 13th to 20th centuries was described by Smith et al. (1999c) and Bouchard (2001). The experimentation showed that there is not so much chemical reaction between the original glass and the mixture placed on top as multiple reactions within the mixture. Blue stain caused by superficial cobalt aluminate 'smalt' or 'cobalt blue' (CoO.nAl203) was easily recognized by characteristic strong Raman bands along with relict initial corundum (A1203). In a green superficial experimental stain on glass the complex Raman spectrum revealed a considerable number of intermixed phases: principally blue smalt with orange crocoite (PbCrO4) to create the average green colour by 'colour subtraction' (i.e. the opposite of the 'colour addition' rules that apply to RGB computer screens) along with minor relict green eskolaite (Cr203) and red minium (Pb304), which had created the crocoite by the oxidizing reaction 6Cr203 + 4 P b 3 0 4 + 702 = 12PbCrO4. A modern commercial deep red glass gave an interesting strong Raman spectrum from the interior of the glass (Fig. 3d); it was possible to identify bands of CdS and CdSe typical of a CdS-CdSe solidsolution (Bouchard 2001), and even deduce the S/(S + Se) proportion to be about 45 atomic % on the basis of a Raman shift calibration made by Schreder & Kiefer (2001).
NON-DESTRUCTIVE RAMAN SPECTROSCOPY The most common mineral pigment found in real archaeological stained glass from earlier periods is hematite (e.g. 13th century from Mans; 16th-17th century from the Mus6e Carnavalet, Paris; 18th-19tb centuries from Strasbourg), but 19th-century glass from Mans and Strasbourg revealed respectively smalt and crocoite. Minerals created by environmental alteration of stained glass included calcite (CaCO3) and gypsum (CaSO4.2H20), and in contact with Pb structural supports a mixture of lead carbonates was found (Bouchard 2001).
immobile analysis in a laboratory: under air with micro-mapping Micro-inclusions in Guatemalan jade. The rutile-quartz-jadeitite from Guatemala mentioned above was examined by Raman micro-mapping with a RENISHAW ® INVIA ® spectrometer to gain more information on the nature of the quartz-jadeite contacts (Smith 2004b,c, 2005d). It was already established from Raman point analysis that the Jd content of the clinopyroxene is highest in the grain cores (c. 95 mol%) and diminishes sharply at the grain boundaries (sometimes 1-2mo1% lower, sometimes 1 0 20 mol% lower), and that the quartz occurs as
(a)
I~
21
micron-sized inclusions in the clinopyroxene grain cores (Gendron et al. 2002). Micromapping of a 50 txm x 90 ~ m surface with a motorized step of 0.4 Ixm acquired over 20 000 complete spectra overnight. These data were then treated and presented in different ways; for example, the integrated area of the main band of quartz (Fig. 4a) or of the T - O - T band of the clinopyroxene was used to reveal the distribution of the presence and absence of the quartz microinclusions, and the Raman wavenumber shift of the T - O - T band was used to reveal the tool% Jd distribution in detail (Fig. 4b). The latter map summarizes the collision of the North American Plate with the Caribbean Plate, subduction and exhumation, all in a 50 ~ m x 90 Ixm surface.
Crystal orientation in an Egyptian scarab. An inscribed Ancient Egyptian commemorative scarab was supposed to be made of enstatite ((Mg,Fe)SiO3 with Mg > Fe) (Fig. le). Despite a strong fluorescence, possibly due to patina formed over several millennia, it was possible to confirm from the Raman spectra obtained by placing the scarab on a Raman microscope stage that it does contain enstatite, and so far no other mineral has been found except minor
(b)
E~
-j -~z:7
"0.4
>0.4
>0.4
>0.4
>0.4
Glauconitic pellets Quartz Micas K-feldspar Plagioclase Sparry calcite Calcite Aplite Sandstone Glauconitic limestone Oolitic limestone Micritic limestone Jasper Fe-oxides
: ! ..... :. : ,
S
>0.4 |
Cc
>0.4 ]
0.1 40%); XX, high (30-40%); X, moderate(15-30%); ++, low (5-15%); +, scarce (3-5%); -, rare (
NEOLITHIC POTTERY OF ABRI PENDIMOUN The use of glauconitic material in 95% of the pottery (Fig. 3a) indicates that raw materials and the production technology remained constant through the Middle and Late Neolithic. Current literature data (Nung/isser & Maggetti 1978; Nung/isser et al. 1985, 1992; Maggetti 1994; Martineau et al. 2000; Di Pierro 2003) highlight that the temper was carefully selected from outcrops even quite distant from the archaeological site. In particular, in Prehistoric times it was common practice to add granitic rocks to ceramic mixtures. These lithologies are easily crushed when heated and rapidly quenched in water (Nung/isser et al. 1992), as thermal shock causes 'self-crushing'. Another aspect regarding the functionality of mixtures is linked to the use of temper mixtures with white rocks, either silicates or carbonates. The technological aspects linked to the use of carbonate rocks are well documented in literature, whereas issues linked to the addition of silicate rocks have been less tackled. Experimental data presented by Kilikoglou et al. (1995, 1998) demonstrate that concentrations of quartz temper > 20 vol% result in a decreasing fracture strength. Did the addition of granites or aplites actually fulfil a technological requirement (for instance, for quartz) or was it simply dictated by tradition? This remains an open issue. We would like to thank the Earth Science Department (University of Siena) for the SEM analyses. We also wish to thank J.-C. l~challier, who has provided his geological survey data. A special acknowledgement is due to F. Crepaldi, who deals with archaeological aspects of VBQ I-Chassey interactions. The authors would like to express their gratitude to the anonymous referees, who improved the first draft of this paper.
References ARNOLD, D. 1985. Ceramic Theory and Cultural Process. Cambridge University Press, Cambridge. ARNOLD, D. 2001. Linking society with the compositional analyses of pottery: a model from comparative ethnography. In: LIVINGSTONE SMITH, A., BOSQUET,D. & MART1NEAU,R. (eds) Pottery Manufacturing Process: Reconstitution and Interpretation. British Archaeological Reports International Series. Archaeopress, Oxford, 1349, 1-12. BAGOLINI, B. & PEDROTTI,A. 1998. L'Italie septentrionale. Atlas du Ndolithique europgen. L'Europe occidentale, 2A, Etudes et Recherches Archdologiques de l'Universit~ de Li6ge, 46, 233-341. BARRAL, L. 1958. L'homme cardial de Castellar. Bulletin du Musde d'Anthropologie Prdhistorique de Monaco, 5, 135-164. BASSO, E. 2004. lndagini mineralogico-petrografiche applicate allo studio di reperti archeologici di eta neolitica del piacentino e del sito archeologico di Abri Pendimoun (Alpi Marittime--Francia). PhD thesis, Universit~i degli Studi di Pavia.
47
BERNAB() BREA, L. 1946. Gli scavi nella Caverna delle Arene Candide, 1. The Istituto di Studi Liguri, Bordighera. BERNABO BREA, L. 1956. Gli scavi nella Caverna delle Arene Candide, 2. Istituto di Studi Liguri, Bordighera. BINDER, D. 1990. N~olithique moyen et sup~rieur dans l'aire liguro-provenqale: le cas de Giribaldi (Nice, Alpes-Maritimes, France). In: GUILAINE,J. & GUTHERZ, X. (eds) Premibres communautds paysannes. Autour de Jean Areal, Montpellier, 147-161. BINDER, D. 1998. Silex blond et complexit6 des assemblages lithiques dans le N6olithique liguroprovenqale. Rencontres mdridionales de Prdhistoire rdcente, Juan-les-Pins. Association pour la Promotion et la Diffusion des Connaissances Archaeologiques, Juan-les-Pins. Aries, 111-128. BINDER, D. 2000. Mesolithic and Neolithic interaction in southern France and northern Italy: new data and current hypotheses. In: PRICE, T. D. (ed.) Europe's First Farmers. Cambridge University Press, Cambridge, 117-143. BINDER, D. 2002. Castellar--Abri Pendimoun (Alpes Maritimes). Deuxibme rapport interm6diaire, Campagne de 2001. Centre d'~tudes Pr~histoire, Antiquitd Moyen Age (CEPAM-CNRS). BINDER, D. & PERLI~S,C. 1990. Strat6gies de gestion des outillages lithiques au N6olithique. Paldo, 2, 257-283. BINDER, D., BROCHIER,J.-E., DUDAY,H., HELMER,D., MARINVAL,P., THII~BAULT,S. & WATTEZ,J. 1993. L'Abri Pendimoun ~ Castellar: nouvelles donn~es sur le complexe culturel de la c6ramique imprim6e m6diterran6enne dans son contexte stratigraphique. Gallia Prdhistoire, 35, 177- 251. BINDER, D., GASSIN, B. & St~NI~PART, I. 1994. l~16ments pour la caract~risation des productions c~ramiques n6olithiques dans le Sud de la France. L'exemple de Giribaldi. Terre Cuite et Socidtd. La Ciramique, Document Technique, Economique, Culturel. Association pour la Promotion et la Diffusion des Connaissances Archaeologiques, Juan-les-Pins. Aries, 255-267. BIRIS, K. T. 1998. Stones, numbers---history? The utilization of lithic raw materials in the Middle and Late Neolithic. Journal of Anthropological Archaeology, 17(1), 1-18. CONSTANTIN, C. & COURTOIS, L. 1985. Le mat6riau c6ramique comme caract~ristique culturelle. L'exemple du d6graissant pendant le N~olithique dans le Bassin Parisien. Etudes des Cdramiques en Archdologie. Documents et travaux Institut G~ologique Albert-de-Lapparont, 9, 19- 26. COURTIN, J. 1976. La baume Fontbr~goua (Salernes, Var). 9e Congrds de l'Union lnternationale des Sciences Prdhistoriques et Protohistoriques, 21-29. COURTOIS, L. 1973. Ph6nom~nes de r6g6n~ration aprbs cuisson de certaines cdramiques anciennes. Comptes Rendus de e'Academic des Sciences, Serie D, 276, 2931-2933. COURTOIS, L. 1976. Examen au microscope pgtrographique des cdramiques archdologiques. Centre de recherches arch6ologiques, Valbonne.
48
E. BASSO ET AL.
CREPALDI, F. 2001. Le Chassfen en Ligurie. Bulletin de la Socidtg Pr~historique Franqaise, 98(3), 485 -494. CREPALDI, F. 2002. Tecnologia e tipologia degli aspetti di tradizione chasseana in Italia settentrionale. In: FERRARI, A. & VISENTINI, P. (eds) II declino del mondo neolitico. Ricerche in Italia centrosettentrionale fra aspetti peninsulari, occidentali e nord-alpini. Quaderni del Museo Archeologico del Friuli Occidentale, Pordenone, 4, 157-166. CRISCI, G. M., RICQ-DEBOUARD,M., LANZAFRAME,U. & DE FRANCESCO, A. M. 1994. Les obsidiennes du Midi de la France. Nouvelle m&hode d'analyse et provenance de l'ensemble des obsidiennes n6olithiques du Midi de la France. Gallia Prghistoire, XXXVI, 299-309. D'AMICO, C. 2005. Neolithic 'greenstone' axe blades from North-western Italy across Europe: a first petrographic comparison. A rchaeometo', 47(2), 235- 252. D'AMIcO, C. & STARNINI, E. 2005. Prehistoric polished stone artefacts in Italy: a petrographic and archaeological assessment. In: MAGGETTI, M. & MESSIGA, B. (eds) Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 257-272. DICKINSON, W. R. & SHUTLER, R. J. 2000. Implications of petrographic temper analysis for Oceanian prehistory. Journal of World Prehistoo', 14(3), 203-266. DI PIERRO, S. 2003. Ceramic production technology and provenance during the Final Neolithic: the Portalban settlement, Neuchatel lake, Switzerland. Revue d'Archdomdtrie, 27, 75-93. FABBRI, B., GUALTIERI, S. & SANTORO, S. 1997. L'alternativa chamotte/calcite nella ceramica grezza: prove tecniche, la Giornata di Archeometria della Ceramica, Bologna. University Press, Bologna, Imola, 183-190. GEZE, B. 1968. Carte g~ologiques de France, feuille Menton-Nice, 973. Bureau de Recherches G6ologiques et Mini~res, Orleans. HEIMANN, R. B. & MAGGETTI, M. 1981. Experiments on Simulated Burial of Calcareous Terra Sigillata (Mineralogical Change). Preliminao, Results. British Museum Occasional Papers, 19, 163-177. HELLER-KALLAI, L., MILOSLAWSK1, I. & AIZENSHTAT, Z. 1987. Volatile products of clay mineral pyrolysis revealed by their effect on calcite. Clay Minerals, 22, 339-348. HOARD, R. J., O'BRIEN, M. J., GHAZAVYKHORASGANY, M. & GOPALARATNAM, V. S. 1995. A materialsscience approach to understanding limestone-tempered pottery from the Midwestem United States. Journal of Archaeological Science, 22, 823-832. KILIKOGLOU, V., VEKINIS, G. & MANIATIS, Y. 1995. Toughening of ceramic earthenwares by quartz inclusions: an ancient art revisited. Acta Metallurgica et Materialia, 43, 2959-2965. KILIKOGLOU, V., VEKINIS, G., MANIATIS, Y. & DAY, P. M. 1998. Mechanical performance of quartz-tempered ceramics. Part I: strength and toughness. Archaeomet©', 40, 261-279. LETSCH, J. & NOLL, W. 1983. Phase formation in several ceramic subsystems at 600 ':C-1000 °C as a function of oxygen fugacity. Ceramic Forum-
International Berichte der Deutschen Keramischen Gesellschaft, 60(7), 259- 267. MAGGETTI, M. 1994. Mineralogical and petrographic methods for the study of ancient pottery. In: 1st European Workshop on Archaeological Ceramics, Universith degli Studi di Roma La Sapienza, Rome, 23-35. MARTINEAU, R. 2000. Poterie, techniques et socidtds. Etudes analytiques et expdrimentales b Chalain et b Clain,aux (Jura), entre 3200 et 2900 av. J.-C. PhD thesis, Universit~ de Franche-Comt6, UFR des Sciences de 1' Homme, du Langage et de la Soci&& MARTINEAU, R., CONVERTINI, F. & BOULLIER, A. 2000. Provenances et exploitations des terres poterie des sites de chalain (Jura), aux 31e et 30e si~cles avant J.-C. Bulletin de la Socidtg Pr~historique Franqaise, 97(1), 57-71. NUNGASSER, W. & MAGGETTI, M. 1978. Mineralogisch-petrographische Untersuchung der neolithischen T6pferware von Burg~ischisee. Bulletin de la Socidtd Fribourgeoise des Sciences naturelles, 67(2), 152-173. NUNG,~.SSER, W., MAGGETTI, M. & STOCKLI, W. E. 1985. Neolitische Keramik von Twann--Mineralogische und Petrographische Untersuchungen. Jahrbuch der Schweizerischen Gesellschaft fiir Urund und Friihgeschichte, 68, 7-39. NL'NGASSER, W., MAGGETTI, M. & GALETTI, G. 1992. Analyse der Scherbensubstanz mit Mikroskop und R6ntgenlicht. /n: BILL, J., NUNGASSER, W. & GALETTI, G. (eds) Liechtensteinische Keramikfunde der Eisenzeit. Jahrbuch der Historischen Vereins ftir das Ffirstentum Liechtenstein, 91,
119-165. ODETTI, G. 1991. I1 Neolitico medio ligure e le influenze chasseane, ldentitg du Chassden, Actes du Colloque btternational de Nemours 1989, M6moires du Musge de Pr6histoire d'Ile-de-France, 4, 59-68. PETERS, T. J. & IBERG, R. 1978. Mineralogical changes during firing of Ca-rich brick clays. American Ceramic Society Bulletin, 57, 503-506. PORAT, N. 1989. Petrography of pottery from Southern Israel and Sinai. In: MIROSCHEDJI (ed.) L'urbanisation de la Palestine b l'age du Bronze Ancien. British Archaeological Reports. International series, Oxford, 527, 169-188. RICQ-DE BOUARD, M. 1996. Pdtrographie et socidtds ndolithiques en France mdditerrandenne. Monographie du CRA, 16, CNRS. SHOVAL, S., GAFT, M., BECK, P. & KIRSH, Y. 1993. Thermal behaviour of limestone and monocrystalline calcite tempers during firing and their use in ancient vessels. Journal of Thermal Analysis, 40, 263-273. STARNINI, E. & VOYTEK, B. 1997. The Neolithic chipped stone artefacts from the Bernabb BreaCardini excavations. Arene Candide: a functional and environmental assessment of the Holocene sequence, Memorie dell'lstituto Italiano di Paletnologia Italiana, 5, 349-426. TITE, M. S., KILIKOGLOU, V. & VEKINIS, G. 2001. Strength, toughness and thermal shock resistance of ancient ceramics, and their influence on technological choice. Archaeometo', 43(3), 301-324.
Provenance and technology of Apulian Neolithic pottery R O C C O L A V I A N O I & I T A L O M. M U N T O N I 2'3
1Dipartimento Geomineralogico, University of Bari, Via E. Orabona 4, 70125 Bari, Italy (e-mail: rocco.laviano @geomin, uniba, it) 2Museo delle Origini, Universith degli Studi di Roma 'La Sapienza' University, Piazzale Aldo Moro 5, 00185, Rome, Italy 3Department of Archaeometry, Science Faculty, University of Bari, Via E. Orabona 4, 70125 Bari, Italy Abstract: Apulia is the best-represented region in Italy as far as archaeometric analyses of Neolithic pottery are concerned. Cross-checked use of petrological (optical microscopy), mineralogical (X-ray powder diffraction) and chemical analyses (X-ray fluorescence) have been performed, in the Dipartimento Geomineralogico of Bari University, on 375 Early to Late Neolithic (from the seventh to the fourth millennium Bc) pottery samples from the Tavoliere and Murge areas. A correlated analysis of 134 samples of the main clayey deposits of the two areas was also conducted. Generally local clays were used and, in some cases, the exploitation of a range of different local fabrics has been verified. In Middle Neolithic sites, the use of non-local clay, probably imported, has been also determined. Few finished pots were actually exchanged at an inter-site scale during the Neolithic. Preparation of raw materials has shown different choices followed by ancient potters. Clays are usually more or less refined and the use of mineral temper such as sand, quartz, calcite and grog has been found. The maximum temperature reached during firing is usually between 600-700 and 850 °C. For some Middle Neolithic fine painted pottery higher temperatures have been suggested (between 850 and 1050 °C), revealing a better firing control and the use of kilns.
Archaeometric analysis of Italian Neolithic pottery, developed only in the last 20 years, has assumed a particular importance in relation to innovations in pottery production and the emergence of productive economies and structured settlements from the end of the seventh millennium Bc. On a regional scale Apulia is the best represented, in terms of both sampled sites (about 45 settlements) and analysed fragments (more than 600). The chronological range and the archaeological facies of the Neolithic period, from Early to Late, have been entirely covered. Many Italian (from Milan, Florence, Genoa and Bari University) and foreign teams (from the UK and Canada) have analysed Apulian Neolithic pottery (Muntoni 2002a). Within the framework of Matson's concept of Ceramic Ecology (Matson 1965; Kolb 1989), the emphasis here will be on research strategies, the relation between goals and methods, and on the sampling techniques of such studies. The ideal development should be the shift from pottery as a simply finished product, whose origin is to be understood, towards an ecological
and sociocultural frame of reference in which physical and chemical data are placed. In light of awareness of both the potentialities and limits of analytical techniques, the status of technological and archaeometric studies of Apulian Neolithic pottery will be briefly reviewed (Muntoni 2002b). The number of analysed samples per site is normally small: the number of fragments varies between one and 49 sherds per site, sampled from surveyed or excavated materials. Of the various archaeometric analytical methods, the most widespread technique is petrological examination on thin-sections optical microscopy (OM), often employed alone or in conjunction with other mineralogical methods, such as X-ray powder diffraction (PXRD) and/or thermal analyses (thermogravimetric analysis (TGA); differential thermal analysis (DTA)). Chemical instrumental neutron activation analysis (INAA), inductively coupled plasma-atomic emission spectrometry (ICP-AES) or morpho-chemical (scanning electron microscopy plus energydispersive spectrometry (SEM-EDS)) analyses
From: MAGGETTI,M. & MESSIGA,B. (eds) 2006. Geomaterialsin CulturalHeritage. Geological Society, London, Special Publications, 257, 49~52. 0305-8719/06/$15.00 © The Geological Society of London 2006.
50
R. LAVIANO & I. M. MUNTONI
are even now rarely used in Apulian Neolithic pottery studies,
arrangements involved in the production, use and distribution of ceramics.
Our approach to archaeometric analyses
Analytical methods
Since 1990 a new approach to the archaeometric analysis of Apulian Neolithic pottery has been applied in the Dipartimento Geomineralogico of Bail University, in collaboration with different archaeological teams from the 'La Sapienza' University, Rome, and the Soprintendenza per i Beni Archeologici della Puglia and, more recently, from the Dipartimento di Beni Culturali e Scienze del Linguaggio of Bari University. A crucial problem in the study of prehistoric ceramics is the fabric variability. The use of appropriate sampling strategies, number of samples and archaeometric techniques is essential if meaningful technological studies are to be done. To verify adequately the variability and heterogeneity within and between pastes, archaeologists had to construct rigorous and explicit sampling schemes that could encompass the full range of variability. Cross-checked use of mineralogical and chemical analyses was then employed: complete combinations of petrological (OM), mineralogical (PXRD) and chemical (X-ray fluorescence (XRF)) analyses have been performed on 375 archaeological samples, all analysed in the Dipartimento Geomineralogico of Bail University (Table 1). Twenty-three settlements, located in the Tavoliere plain and in the Murge plateau along the Adriatic coast, were sampled (Fig. lb). A correlated analysis of 134 pelitic and clayey samples was also conducted (Laviano & Muntoni 2004). Only the analysis of clay sources can give more information on which clay source was used, and why, and what kind of problems were encountered by the ancient potters in the modification of the materials. 'Any investigation of ceramic technology at an archaeological site or region is most fruitfully conducted within the perspective of ceramic ecology: it is essential to search out deposits and sample them so as to compare the pottery of interest with the properties of local clays' (Bishop et al. 1982, p. 319). As all the archaeometrical data are separately published or are still in press, in this paper we aim for the first time at a geographical and chronological synthesis of our results. The reconstruction of the working sequence used in pottery manufacturing (from raw material provenance, to preparation of bodies and firing techniques) should help to obtain insights into the potter's role as an active and controlling agent in the procedure of a specific pottery manufacture, and into the pottery economics or the socio-economic
Mineralogical studies were carried out by PXRD using a Philips diffractometer (PW 1710) with Ni-filtered Cu K,~ radiation and employing NaF as internal standard. Petrological observation was made on thin sections, with a polarized light microscope (OM). Modal analysis was carried out using a Swift & S. Point Counter on 2500-4500 points for each sample (according to their wall thickness), with a line distance of 0.05 mm and a lateral step of 0.2 mm. Major and trace element determination was performed by XRF, using a Philips PW 1480/10 spectrometer (Cr anticathode for major and minor elements, Rh anticathode for Rb, Sr, Y, Zr, Nb and W anticathode for Ce, La, Ba, Ni, Cr, V), following the analytical techniques outlined by Franzini et al. (1972, 1975) and Leoni & Saitta (1976). About 5 g of representative powder for each sample was subjected to XRF. Two reference standards (AGV-1 of the USGS, USA and NIM-G of NIM, South Africa) were used to check the accuracy of the analytical data. Loss on ignition was determined by heating the samples at 1000 °C for 12 h; then PXRD patterns of the same previously heated samples, for the identification of mineralogical changes, were recorded at room temperature.
The geological context The Tavoliere plain
The Tavoliere plain (Fig. 1a), the most extensive one in Southern Italy, is a Mesozoic-Palaeogene limestone depression filled with marine deposits of Plio-Pleistocene silty clay (Bradanic cycle), often overlain by post-Calabrian marine sands (terraced marine deposits), Upper Pleistocene (terraced alluvial deposits) and Holocene alluvial and lacustrine deposits of continental origin. Marine Plio-Pleistocene clays of the Bradanic cycle, also named the Argille Subappennine, crop out along the western margin of the Tavoliere plain. The depth of the outcrops may vary from a few metres to 350 m. The clays consist of silty clay or clayey silt, with little sand, and have (Balenzano et al. 1977; Dondi et al. 1992) a very similar mineralogical composition (clay minerals, carbonates, quartz and feldspars). The clay minerals are a mixture of 2 M illite, Mgbearing smectite, Fe-bearing chlorite, kaolinite and randomly interstratified illite-smectite with 30-70% montmorillonite-like layers. Natural
51
APULIAN POTTERY
Table 1. Sampled Neolithic settlements of Tavoliere and Murge areas Code
Samples Excavation
Tavoliere Monte Aquilone Masseria Valente Coppa Nevigata Masseria Candelaro Masseria Santa Tecchia Podere 96 Masseria Cascavilla Capo di Lupo Masseria Centonze Masseria Mischitelli Casello Amendola Total
MA MV CN MC ST P96 MCS CL MCZ MM CA
15 13 8 61 11 7 6 5 3 3 2 134
X X X X X
Murge Balsignano Pulo di Molfetta Ciccotto Madonna delle Grazie Torre delle Monache Santa Barbara Setteponti Cala Colombo Grotta della Tartaruga Grotta Scanzano Masseria Chiancudda Cala Scizzo Total
BALS PU CC MG TM SB SP CCL GT GS CH CS
30 59 33 6 6 43 15 10 10 9 1 19 241
X X X
non-plastic material consists of carbonates (calcite, as bioclastic or detrital granules, and dolomite), quartz and feldspars (orthoclase, microcline and Na-plagioclase). Very different Holocene alluvial clays can be found on the coastal plain, deposited by the numerous ancient rivers and streams. The clay composition is very variable depending on the erosion of different clayey and arenaceousmarly deposits (Cassano et al. 1995b; Eramo et al. 2004). Alluvial clays have volcanic minerals and rock fragments as a distinctive feature. Heavy minerals are represented by dominant diopside-augite pyroxene, magnetite, biotite and garnet, together with debris of volcanic glass. Marine and alluvial clays are characterized by the relative abundance of SiO2, A1203, CaO, Fe203, K20 and MgO. Because of their mainly calcareous composition (up to 17 wt% CaO), both groups can be classified as marly clays. In general, clay fractions ( < 2 Ixm) have a lower CaO content than the whole specimens, whereas the A1203 and Fe203 concentration is higher.
Survey
Early Neolithic
Middle Neolithic
X X X X X X X X X X
X X X X X X
X X
X X X X X
Late Neolithic
X X
X X X X X X X
The Murge plateau
The large geologically homogeneous Murge plateau (Fig. 1a) is formed by the limestone formations of Calcare di Bari and Calcare di Altamura, with terra rossa deposits present in the sequence. Terra rossa are silty-clayey continental sedimentary deposits, very poor in carbonate (Dell'Anna 1967; Dell'Anna & Garavelli 1968; Dell'Anna et al. 1973) composed of dominant clay minerals (illite and kaolinite) and Fe-oxides or hydroxides, with subordinate quantities of quartz, feldspars, micas and pyroxenes. SiO2, A1203 and Fe203 are the main oxides, both in the clay fraction and in the whole specimen. Marine Plio-Pleistocene silty clays of the Bradanic cycle (Argille Subappennine) crop out extensively along the western margin of the Murge plateau (Dell'Anna & Laviano 1991) and locally in the Rutigliano area (Dell'Anna 1969; Moresi 1990).
Raw material provenance Our studies indicate that generally local clays were used for Neolithic pottery production, in
52
R. LAVIANO & I. M. MUNTONI (a)
o. . . . . . . . . .
50 km i~,,. N
-%
:----" ( :i4~-~.~ -.
Adriatic
"\
[~
Alluvial and lacustrine deposits (Holocene) Terraced alluvial deposits (Upper Pleistocene)
[]
Terraced marine deposits (Upper-Middle Pleistocene)
[]
Bradano Units (Pliocene-Lower Pleistocene) Calcarenites (Miocene)
I~ [~
Calcarenites (Eocene) Limestone (Mesozoic)
-
~,
.
[]
[]
sea
,
.
Apenninic Chain Units (from Cretaceous to Pliocene)
~;:-: :
Ionian sea
(b)
50 km v o l i e r e
\--"-'--~r-,~. h! D , 3CCL
Fig. 1. (a) Geological and geomorphological map of Apulia (from Caldara et al. 1990, fig. 5). (b) Location of sampled Neolithic settlements (*) of Tavoliere and Murge areas.
APULIAN POTTERY
53
both the Tavoliere and Murge areas, with some differences between the two areas.
Early Neolithic (6200-5500 BC) of the Tavotiere plain Plio-Pleistocene silty clay and Holocene alluvial deposits of continental origin were exploited in the Tavoliere plain (Cassano et al. 2004). Archaeometric data from eight neighbouring Early Neolithic villages, located in a small area (150 km 2) of the plain near the Adriatic coast, show that ceramic production seems fairly homogeneous (raw material supply, grain-size variability and firing techniques) in these villages, although some degree of variation in grain-size composition of natural non-plastic material is also present between different wares (Cassano et al. 1995b; Muntoni & Laviano 2005). As far as the ecological and technological aspects are concerned, the same alluvial deposits, as shown by the well-preserved mineralogical component, from three river valleys were exploited. This hypothesis is also sustained by the presence in the pottery of clasts of pyroxenes and volcanic glass typical of alluvial clays. Behavioural and technical similarities in Early Neolithic pottery technology are thus confirmed. The concentration of chemical elements is very useful to identify a strong affinity between Early Neolithic samples that can be all defined as 'Ca-rich'. A very few finished pots were considered as outliers (Fig. 2a), probably exchanged at an inter-site scale during Early Neolithic. The ternary diagram (Fig. 3a) also shows a good chemical correspondence between pottery samples and alluvial clay deposits from the Tavoliere plain. The difference of CaO content between some pottery samples and alluvial clays (Fig. 3b) is due to the compositional variability of alluvial deposits and to the addition of calcareous sand in some samples of coarse ware, mainly from Monte Aquilone village. Three principal criteria have been used to identify temper: (1) the hiatal distribution of the inclusions; (2) a proportion of inclusions >30%; (3) the relative abundance of micritic bioclasts (mainly bivalves), limestone clasts and fine-grained calcite. XRF concentrations have also shown very fine local distinctions (Fig. 2b) between sites located in different alluvial basins (Muntoni & Laviano 2005). Therefore, villages collected their clays in areas around the settlement itself in the nearest alluvial basin. Thus groups that had the same taste and behavioural choices
300
IOO08
~ Sr
300
400 2 0 0 ~ 0 0 ~ / / " lO0
Zr
(b)
1000
4
o
o
F
~
,oo
1000~ , ~ J
.
~ .
.
.
~
~
..... .I
~
300
Fig. 2. (a) Crv. Sr v. Zr plot (ppm) for all Early Neolithic pottery samples (+, n ----73) from the Tavoliere area (note the cluster and the few samples (labelled) that could be considered as outliers). (b), Ba v. Sr v. Zr plot (ppm) for all Early Neolithic pottery samples from the Tavoliere area and the samples (marked) from sites located in two alluvial basins (D, samples from CN and MV villages; V, samples from CL, MM and MCS villages).
were completely autonomous as far as raw material supply and pottery manufacture are concerned.
Early Neolithic (6200-5500 BC) of the Murge plateau The exploitation of different local deposits in the same Early Neolithic site has been verified with an extended range of analysed samples, as in the large karst doline of Molfetta (locally known as 'pulo') or in the village of Balsignano, both located near the Adriatic coast of the Murge plateau (Muntoni 2003).
R. LAVIANO & I. M. MUNTONI
54
(a)
Si02 ._ ;,,,
, ,,
kj
/'
\
. -
-~/ C a - r i c h "
./
\
~.~
-.// "-\ /.' , ~" /.
CaO+MgO!
~"'Gh
\
.
\,
,./.,~ C a - p o o r ,,,.
'\
" ,.
!, An
.
, ]",
',
~',.
..........
[ ~ ~-
,
,,, •
,
'
'
0
(b)
~'~
',
25
5O
75
AI203 100
60.00 + +
5000
4-
t,++t -,*+ ~
+
40.0O
+
.~
30.00
+
,4+
2000
10.00
0.00 0.00
500
10.00
15.00
20.00
25 00
30.00
CaO
Fig. 3. (a) (CaO + MgO)-A1203-SiO2 wt% diagram (Di, diopside; Gh, gehlenite; An, anorthite); (b) S i O 2 v . CaO plot (wt%). Early Neolithic pottery samples (+, n = 73) from the Tavoliere area and the fields (A, whole sample; B, clay fraction) of alluvial deposits (n = 7).
Three paste groups with different dominant mineralogical constituents (predominant quartz and calcite, with little K-feldspar and plagioclase) were found in the Early Neolithic strata of Pulo di Molfetta (Laviano & Muntoni 2003). Raw material variability is evident mainly on a synchronous level: three pastes (Cal fabric, Qtz + Cal fabric and Qtz fabric) were used in the same village and in the same archaeological horizon. The dominant clastic constituents are all compatible with the geological deposits that crop out in the site hinterland. The differences in mineralogy and grain-size distribution (detected by PXRD and OM) between Qtz + Cal and Qtz fabrics could be explained if we hypothesize that Neolithic potters collected different clays in more areas around the settlement itself. The large amount of carbonate
fossil fragments (mainly molluscs and rare benthonic Foraminifera), which characterize the Cal-rich fabric, is due to an intentional addition of a calcareous temper. In this case two principal criteria have been used to identify temper: (1) a proportion of inclusions >_30%; (2) the exclusive presence of carbonate fossil fragments (from 2 0 0 - 4 0 0 p,m to 1.5-2 mm). XRF analyses are consistent with mineralogical data (obtained by PXRD and OM). SiO2, CaO and AI203 are the dominant oxides, with some variations in CaO percentage. The ternary diagram show a clear differentiation of the Cal-rich samples, the Qtz + Cal-rich samples being characterized by relatively more balanced amounts of the three main oxides, whereas the Qtz-rich samples are characterized by higher quantities of Qtz (Fig. 4a). This last group shows a good overlap
APULIAN POTI'ERY
55
(a)
\
Di
\
x,\\\
,/ ×\
~
"
"7
\
/
./
,.~ "-/ ~'~ .... /
'
'
0
(b)
/ /
. /
,
......... ~ \
\
'"'
,
"X
~
............ ::~ \
\
/'"" \
/
\ ,
'
\
•
G
\
\\\X \ \
-./
CaO+MgO(
V
//
!
\
,
\ ,
1 ' ' ' '
r
25
50
'
X" \
// /'
'\,
XX
,
xe-
\ ,
,
,,
\
I
'
/
o
''
75
/AI203 100
50 A
~ ........ t
-
t
roe
30 Nb
25
20
T
!
T-
15
!
4 I
J
i
;
q
5
o
[a r'l
'
5
10
15
', -
+-
~
"
' " ai -
IArgitle diRutigliano ~............. . . . . . . . . . . . .
L 0
a
I Z3.-=,.-.-~-,~_ --
20
i
'~
25
30
I
t
!
, 35
40
45
50
Y
Fig. 4. (a) (CaO + MgO)-A1203-SiO2 wt% diagram (Di, diopside; Gh, gehlenite; An, anorthite); (b) Nb v. Y plot (ppm). Early Neolithic Cal-rich (×), Qtz + Cal-rich (El) and Qtz-rich (A) pottery samples (n = 47) from Pulo di Molfetta and the fields (whole sample) of Argille di Rutigliano (n = 5) and terra rossa (n = 25).
with the chemical composition of the terra rossa and the Qtz + Ca1 group with the Argille di Rutigliano of the Murge plateau. Also, some trace elements, such as niobium (Nb) and yttrium (Y), are important for distinguishing the three groups of pottery (Fig. 4b). In a few cases, different paste types can be actually related to typological groups of vessel form or archaeological classes, as in the Early Neolithic village of Balsignano (Modugno). At this site (Muntoni 2003) all Neolithic fragments were macroscopically analysed and attributed to the paste groups. Only four basic vessel forms were distinguished in the analysed materials: dishesplates, bowls, large jars and collared jars. The occurrence of bowls and large jars in pastes is different (Muntoni 2003, table 60). Bowls are more concentrated (77%) in Q t z + Cal-rich fabrics, with few carbonate clasts: the high
quantity of quartz inclusions was probably reserved for pots generally used for serving or display. Large jars are more concentrated (60%) in Cal-rich fabrics, with carbonate rock and fossils: the higher proportion of carbonate inclusions, mainly in coarse wares, could suggest their use as cooking and/or storage vessels. In low-fired cooking vessels, which are heated and cooled during use, because the thermal expansion of calcite is similar to that of average fired clays (Bronitsky & Hamer 1986; Fabbri et al. 1997), stresses owing to differential expansion of the clay matrix and temper are usually minimal.
Middle Neolithic (5500-4400 or 4200 Bc) of Tavoliere and Murge Middle Neolithic pottery shows a substantial shift in the whole production sequence; in
56
R. LAVIANO & I. M. MUNTONI
particular, the systematic exploitation of Marine Plio-Pleistocene silty clays, even in sites where they could be considered non-local materials, has been determined in both the Tavoliere (Cassano et al. 1995b; Muntoni 1999) and the Murge areas (Muntoni 2003; Muntoni et al. 2006). All red and/or brown painted pots (figulina), which are typical of the Middle Neolithic of Southern Italy, show a very fine paste texture, with a fairly fat clay matrix: of the sheet silicates only mica crystals are recognizable. Non-plastic inclusions are homogeneous fine-grained (< 150-200 ~m) quartz and scarce carbonate fossils (mainly benthonic Foraminifera). Clasts of feldspars and iron oxides or hydroxides are also present. XRF analyses showed that all samples are Ca-rich. The ceramic diagram (Fig. 5) shows an almost complete overlap of the samples in the central part, corresponding to the chemical composition of Plio-Pleistocene Apulian silty clays. A few samples from the village of Masseria Candelaro in the Tavoliere area are characterized by very high quantities of calcite, whereas others from the northern Murge sites of Pulo di Molfetta and Grotta Scanzano have lower quantities. The former shows a more Cal-rich fabric with carbonate fossil fragments (mainly molluscs, such as bivalves and gastropods), whereas the latter has a Qtz-rich fabric with very few carbonate inclusions. Also, trace
element concentrations confirm the homogeneity between samples. This finding is in agreement with the geochemical homogeneity of the PlioPleistocene Apulian clays. Nevertheless, some trace elements, such as Ba, Sr and Zr (Fig. 6), are important for distinguishing sub-groups of pottery related to their geographical setting. Mineralogical and chemical data clearly show in the two areas the exploitation of the PlioPleistocene silty clay, which in some cases crops out more than 30 km from the sites. The use of specific clay-beds shows a more complex clay supply activity, involving perhaps the whole group of people. Such an activity might be distinct from individual and domestic tasks, and may suggest that local production was no longer domestic. In Middle Neolithic societies, pottery production, mainly of fine painted ware, probably evolved from a domestic mode of production to an incipientspecialization stage (Rice 1981; Van der Leeuw 1984). This stage would include an increasing standardization of paste composition, reflecting greater exploitation of particular kinds of clays. In addition, a greater skill is more evident in manufacturing and firing (up to 1000 C ) technology. Middle Neolithic black household pots, analysed in two Murge settlements (Setteponti and Santa Barbara), show a silicate matrix with angular to sub-angular coarse-grained alabastrine
SiO2&, ./ "v~.. i
)...-"
"-/ !
', Ca-rich ,,f.
,>,'
~
.// ,"
* \ ,
•- .. ............
0
'
""> ),~ i
, '\
~
.
25
,,
\
~/"'~' •/ ,
r
\ ; An
,
,I
Ca-poor x
',i f
\
~,,
C a O + M a O
./
.
.
.
'.f-
:~ ...........
"'/Gh
~ .....
50
O
~__. !
L.f"
75
AI20 3 t00
"
Fig. 5. (CaO + MgO)-A1203-SiO 2 wt% diagram (Di, diopside; Gh, gehlenite; An, anorthite). Middle Neolithic fine painted pottery samples (n = 127) from the Tavoliere plain (•), the Bradanic trough (grey dots), the northwestern (x) and southeastern (D) areas of the Murge plateau, and the fields (whole sample) of Plio-Pleistocene silty clays (n = 89).
APULIAN POTTERY
(a)7°°] Ba
Non-calcareous clay (terra rossa) could be used as a raw material, tempered with crushed alabastrine limestone clasts.
/
! 600~
o
i
Late Neolithic (4400 or 4 2 0 0 - 4 0 0 0 Bc)
o
of the Murge plateau
oo
6oo! o
o o**
•
P* " 8 o
*
o~
.
400.
" ~.+.
o
o
oe.~..-
, c~°.
300 o
o
o
•
o @@
200 200
(b) 700-
400
600
800
10'00 12'00 14'00Sr
Ba
600-
500.
•
o
°
o
i
*
*
• o~
° •
4004
j.: •'.., • i
•
}
° O
]
• '4~'0
0,
111¢* O0 * °°
. *. +
GO
300~
•@~.
"z~" •
°
,,
i "; 200
57
"
•
6o
8'0
16o 1to 1~,o 16o 180 260 220 Zr
Fig. 6. (a), Bav. Sr plot (ppm) for Middle Neolithic fine painted pottery samples (small dots) and the samples from the Bradanic trough (grey dots) and the southeastern area of the Murge plateau (R); (b) Ba v. Zr plot (ppm) for the Middle Neolithic fine painted pottery samples (small crosses) and the samples from the Tavoliere plain (e) and the Bradanic trough (grey dots).
limestone clasts; quartz and Fe-oxides or hydroxides as aggregates and pisoliths were also observed as natural non-plastic inclusions. PXRD analyses confirm the presence of predominant calcite, accompanied by variable amounts of quartz and feldspar. XRF analyses are consistent with the mineralogical data: SiO2, C a • and A1203 are the dominant oxides (Geniola et al. 2005; Muntoni et al. 2006).
Late Neolithic pottery (black and plain household wares) was sampled only from the Cala Scizzo cave (Bail province), located on the Adriatic coast of the Murge plateau (Geniola et al. 2005). As far as the ecological and technological aspects are concerned, archaeometric data suggest the use of two different (noncalcareous and calcareous) clays for production for the two archaeological wares. Raw material variability is again evident (as in the Early Neolithic pottery of the same area) mainly on a synchronous level. Petrological and mineralogical data, and chemical concentrations of SiO2, A1203 and C a • allow the identification of two material groups, with different dominant clastic constituents. In the black household pottery, the silicaterich matrix is dominant and non-plastic inclusions are coarse-grained quartz and feldspar clasts, with no carbonate rocks. These samples are also distinguished by the highest AlzO3 and SiO2 values (Fig. 7a) related to clay matrix abundance. Terra rossa, which is very poor in carbonate (Dell'Anna 1967; Dell'Anna & Garavelli 1968; Dell'Anna et al. 1973), could be used as a raw material for this ware. Plain household pottery shows an abundant sheet silicate matrix, in which only the micas are recognizable; nonplastic inclusions are homogeneous fine-grained minerals such as quartz and carbonate fossils (planktonic Foraminifera). This ware is also distinguished by similar amounts of the three main oxides. The Argille di Rutigliano, marly clays cropping out not far from the considered site (Moresi 1990), are consistent with the sheet silicate clay matrix of plain household pottery (Fig. 7a). As regards trace elements, rare earth elements (REE), such as lanthanum (La) and cerium (Ce), clearly show only two clusters (Fig. 7b): the former contains plain household pottery and the latter black household samples.
Preparation of bodies Preparation of raw materials has shown the different choices followed by ancient potters in the preparation of bodies. Clays are usually more or less refined, and in some cases the use of mineral temper (such as fossiliferous sand, alabastrine limestone and calcite), grog or
58
R. LAVIANO & I. M. MUNTONI (a)
c>Si02~
J
'
~/'~
~'Argille di Rutigt!ano
'
,/
\\\
/
-,/ / AI~
/ ,-
\ \
"/
CaO+MgOI
L
,
~
0
(b)
V Gh \
,
i \
~
\\
/ \\
\
\
, I 25
'
'
"
; '\'\
;
/
\
, \
~
~'~-
\
, J 50
'
~
'
/."~,. \
, \
'
/
\
, 1 75
'v" '
'
~
'
IAI203 100
2oo
180 160 140 120 Ce
too
8o
_ _~
.......... " ~ - - -
~ _ . _ ~
6o ~.~Argille
di Rutigliano
40 2o 0
J
10
1
20
30
40
50
80
70
80
90
100
La
Fig. 7, (a) (CaO + MgO)-A12Os-SiO2 wt% diagram (Di, diopside; Gh, gehlenite; An, anorthite; (b) Ce v. La plot (ppm). Late Neolithic plain household (x) and black household (O) pottery samples (n = 19) from Cala Scizzo, and the fields (whole sample) of Argille di Rutigliano (n = 5) and terra rossa (n = 25).
vegetable material has been found, with variations in its incidence from sample to sample, from site to site and from area to area. The Tavoliere samples (Cassano et al. 2004) showed that clays were generally used as they naturally occurred, as has been proven by the mineralogical component, which compares well with that of local clays, although some degree of variation in grain size and percentage occurs. Microscopic observation also revealed that fossiliferous sand or crushed sparry calcite was occasionally (n = 18) employed as temper in coarse ware in only three Early and Middle Neolithic settlements (Monte Aquilone, Santa Tecchia and Masseria Candelaro). In the Murge area ancient potters followed different choices in the preparation of pastes.
Analytical data have shown that the main degree of variation was in paste preparation and grain-size (from 5 0 - 2 5 0 to 5 0 - 6 0 0 p~m) composition. Some Early Neolithic Qtz-rich samples from Pulo di Molfetta and Ciccotto (Muntoni 2003) are characterized by the presence of grog fragments (n = 9, mainly from Pulo) or by the presence of curvilinear and long pores probably derived from vegetable material burnt during firing (n = 5, from Pulo). The large amount of fossils, which characterized the Cal-rich Early Neolithic coarse wares of Pulo di Molfetta and Balsignano, is due to an intentional addition of a calcareous temper to terra rossa. The presence of this particular ware in two different Neolithic villages could also indicate the same appreciable intra-group
APULIAN POTTERY choice as a response to functional or/and social constraints. Only Middle Neolithic black household pots from the Murge settlements of Setteponti and Santa Barbara (about 90 km apart) are systematically tempered by angular to subangular coarse-grained alabastrine limestone clasts (Geniola et al. 2005; Muntoni et al. 2006). In this case three principal criteria have been used to identify temper: (1) the hiatal distribution of the inclusions; (2) a proportion of inclusions _>30%; (3) the angular outlines of the grains. Such data could give some positive insight into the inter- and intra-group organization of the many Middle Neolithic communities of the Murge plateau, who shared so many other common behavioural features (Cassano 1993).
Firing techniques Mineralogical and petrological data gave some insight concerning pottery firing temperatures: the maximum temperatures reached during firing have been inferred to be usually between 600-700 and 850 °C in both the considered areas. As kaolinite, which is a common component of the Apulian clays, has disappeared in every sample, one could suggest that temperatures exceeding 600 °C were always reached. Early Neolithic pots are usually light buff coloured; pottery was probably fired in pits and several factors determined the success or outcome of the firing process (for some experimental tests, see Cassano et al. 1995a). For Early Neolithic firing technology, mineralogical data show that maximum temperatures did not exceed 800 °C. However, on the basis of the presence in X-ray patterns (Table 2) of very weak peaks of some clay minerals (illite plus muscovite and minor quantities of smectite), one could argue that some Early Neolithic samples were fired at temperatures that could have reached 600-700 °C. The high peaks of primary calcite in Early Neolithic Cal-rich samples (Table 2) show that these coarse pots also were fired at a temperature not exceeding 700 °C. Other samples, as the lower amount of clay minerals, together with the presence of calcite, shows (Table 2), could have been fired at higher temperatures, but still not over 800 °C. In Middle and Late Neolithic villages, black burnished and plain red and/or brown painted pottery are associated in archaeological contexts. For these pots, involving different firing structures and techniques, greater efforts were made by ancient potters to control the amount of oxygen that entered the firing structures, to produce an oxidizing or reducing atmosphere.
59
Mineralogical and chemical data show that Middle-Late Neolithic black household and Late Neolithic plain household pots were fired at a temperature not exceeding 600-800 °C. For Middle Neolithic fine painted pottery (the so-called f i g u l i n a ) higher temperatures have been suggested, revealing a better firing control (temperature, rate of heating and oxidizing atmosphere) and the use of kilns. The absence of a dark core and the low birefringence of the matrix confirm a high degree of sintering. Such samples show gehlenite and pyroxene neoformations, only apparent by PXRD analysis, whereas clay minerals are absent (Table 2). In the same group one can see decreasing amounts of calcite, which, as microscopic observation on thin section shows, was recrystallized in pottery pores. On the basis of such data one can suggest that temperatures between 850 and 1050 °C were obtained. In some settlements (Masseria Candelaro in the Tavoliere plain and Pulo di Molfetta and Ciccotto in the Murge plateau) a clear differentiation between Early and Middle Neolithic ceramics (the latter fired at higher temperatures) has been found (Cassano et al. 2004; Muntoni 2003). Unfortunately, in the Tavoliere and Murge areas very few fire structures have been identified, probably because of a lack of extensive archaeological excavations, and no direct connection with ceramic firing (rather then with baking or roasting) can yet be established. PXRD analyses of representative samples heat-treated at 1000 °C (Table 2) confirm that pyroxene and gehlenite synthesis is dependent on calcite and clay mineral abundance (Maggetti 1982). The presence of a high quantity of neoformed pyroxene and gehlenite was detected in Qtz + Cal-rich samples; their concentration increases in Middle Neolithic fine painted pots and in Early Neolithic Qtz ÷ Cal-rich samples whereas they are initially absent. In Qtz-rich samples, found only in Early (Pulo di Molfetta and Ciccotto) and Late Neolithic (Cala Scizzo) sites of the Murge region, characterized by very low quantities of calcite, only hematite is the secondary product of firing. Only in Cal-rich heat-treated samples, characterized by coarsegrained carbonate fossils (Early Neolithic Pulo di Molfetta and Balsignano coarse ware) and alabastrine limestone clasts (Middle-Late Neolithic black household ware), was the co-occurrence of diopside, gehlenite and hematite observed. CaO and a neoformed calcium silicate very similar to larnite were also detected, probably formed as a result of the high quantities of CaCO3 in the paste.
FP FP FP FP FP FP FP FP FP FP BH BH BH BH BH
BH BH BH PH PH PH
Middle Neolithic PU30 PU53 MC23 MC24 MC30 SP04 GT02 GS09 CC08 CC 18 SPI4 SPI5 SP18 SB01 SB 11
Late Neolithic CSI0 CS02 CS01 CSll CSI2 CSI7
X XX tr X X
X X X X XX
X
X X tr tr X tr
XX XX XX XX XX XX XX
Sm
tr
tr X X tr
tr X XXX XX XX X XX
tr
tr tr
X X X XX XX XX
XX X XX X XX XX XXX
I11 + Ms
XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX
XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXX XXXX XXXX XXXXX XXXX
XXXX XXXX XXXXX XXXXX XXXX XXXXX XXXXX XX XX XX XXX XXXXX XXXXX XXX XXXXX XXXXX XXXXX
Qtz
XXX X X XX X XX
XXXX XXX XXX XX XX XXX XX XXX XXXX XXXX X X X X XXX
X X XX X X XXX XXX tr tr tr tr XX XX XXXX XXXX XXX XXXX
Feld
tr XXXX XXXXX XXXX
X X X X tr tr XX XX XXXXX XXXXX XXXXX XXXXX XXXXX
tr
XXXX XXXXX XX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX
Cal
X tr
XX XX XX X XX XX XX XX XX XX
Px
X XXX XX XX XX XX XX XX XX XX
Gh
X tr tr
tr X tr X tr X
Hem
XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX
XXXXX XXXXX XXXXX XXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX XXXXX
XXXX XXXX XXXXX XXX XXXX XXXXX XXXXX XX XX XX XXX XXX XXXX XXX XXXXX XXXXX XXXXX
Qtz
XXXXX XX XX XXX XXX XXX
XXXXX XXXX X X XXX XX XX XXX XXXX XXX XX XXX XX XXX XX
X XX XXX XXX XX XXX XXXX tr tr tr tr XX XXX XX XXXX XXX XXX
Feld
X X XXX XXX XXX
XX XX XXX XX XXX XX XXX XX XXX XXX XX XX XX X XX
XX XXX XX XXX XXX XXX XXX X X X X XX XXX XX XX X
Px
Gh
XX XX XX tr
XXXX XX XXX
XX X tr tr tr X tr X X tr X X tr XX XX
XX XX XX
Hem
tr
X XX XXXX XXXXX XX XX XX XX X X XX XX XXX X X
X XX X XX XX XX XX XXXX XXXX XXX XXX XXX XX XXXX
heat-treated at 1000 C
Sm, smectite; I11, illite; Ms, muscovite; Qtz, quartz; Feld, k-feldspar and plagioclase; Cal, calcite; Px, pyroxene; Gh, gehlenite; Hem, hematite (symbols as in Kretz 1983); number of X is in relationship with mineralogical phase abundance; tr, traces. Archaeological wares: C, coarse; B, burnished; P, painted; FP, fine painted; BH, black household; PH, plain household.
C B P P P C C C C C C C P C B C B
Ware
as-received
Mineralogical composition (by PXRD) of representative pottery, samples
EarlyNeolithic CN03 MCI2 MM1 MA8 MV8 PU02 PU26 PUI6 PUI9 BALS24 BALS30 MCSI MA4 MAI0 PU01 PU05 CC27
T a b l e 2.
¢~ o
APULIAN POTTERY
61
Concluding remarks
References
The review of Apulian Neolithic settlements providing archaeometric pottery data has shown that ceramic studies employing laboratorybased techniques remain one of the most active areas of research in Italian archaeometry. A complete combination of petrological (OM), mineralogical (PXRD) and chemical (XRF) analyses is more informative and capable of greater discriminatory power. The correlated analysis of clay sources is most fruitfully conducted within the perspective of gathering information about which clay source was used, and why, and what kind of problems were encountered by the ancient potters in the modification of the materials. The main patterns of variation of Neolithic pottery production, from the seventh to the fourth millennium Bc, may be summarized in relation to provenance of raw materials and/or finished pottery artefacts, preparation of raw materials and firing techniques. Our studies indicate that generally local clays, the PlioPleistocene silty clay of the Bradanic cycle (Argille Subappennine and Argille di Rutigliano) and silty-clayey continental sedimentary deposits (terra rossa), were used for Neolithic pottery production. In some cases the exploitation of a range of different local fabrics has been verified. In Middle Neolithic sites the systematic use of Plio-Pleistocene silty clay has been also determined, even in some sites where such clay cannot be strictly considered as a local raw material (sometimes cropping out more than 30 km from the sites). Few finished pots were actually exchanged at an inter-site scale during the Neolithic. In the preparation of raw materials, different choices were followed by ancient potters in the preparation of pastes. Clays are usually more or less refined and, in some cases, the use of mineral temper, grog or vegetable material has been found, with variations in their incidence from sample to sample and from site to site. Firing techniques have been also considered; the maximum temperatures reached during firing are usually between 6 0 0 - 7 0 0 and 850 °C. For Middle Neolithic fine painted pottery higher temperatures have been suggested (between 850 and 1050 °C), revealing a better firing control (temperature and atmosphere) and the use of kilns. The different sources of ceramic variation, their relative frequency and the potter's role as a controlling agent in pottery manufacture will be further explored in the next stage of our research.
BALENZANO, F., DELL'ANNA, L. & DI P1ERRO, M. 1977. Ricerche mineralogiche, chimiche e granulometriche su argille subappennine della Daunia. Geologia Applicata e Idrogeologia, XII(II), 33 -55. BISHOP, R. L., RANDS, R. L. & HOLLEY, G. R. 1982. Ceramic compositional analysis in archaeological perspective. In: SCmFFER, M. B. (ed.) Advances in Archaeological Method and Theory, 5. Academic Press, New York, 275-330. BRONITSKY, G. & HAMER, R. 1986. Experiments in ceramic technology: the effect of various tempering materials on impact and thermal-shock resistance. American Antiquity, 51, 89-101. CALDARA, M., FATIGUSO, R., GARGANESE, V. & PENNETTA, L. 1990. Bibliografia geologica della Puglia. SAFRA, Bari. CASSANO, S. M. 1993. La facies Serra d'Alto: intensificazione delle attivit~ produttive e aspetti del rituale. Origini, XVII, 221-253. CASSANO,S. M., EYGUN,E., GARIDEL,Y. & MUNTON1, I. M. 1995a. Pottery making in southern Italy Neolithic: an experimental study. In: VENDRELL-SAZ, M., PRADELL,T., MOLERA,J. & GARCIA,M. (eds) Estudis sobre cerhrnica antiga. Actes del simposi sobre ceramica antiga. Universitat de Barcelona, Barcelona, 11- 16. CASSANO, S. M., LAVIANO, R. & MUNTONI, I. M. 1995b. Pottery technology of early Neolithic communities of Coppa Nevigata and Masseria Candelaro (Foggia, Southern Italy). In: FABBRI, B. (ed.) The Cultural Ceramic Heritage European Ceramic Society Fourth Conference. Gruppo Editoriale Faenza, Faenza, 14, 137-148. CASSANO, S. M., ERAMO, G., LAVIANO, R. & MUNTONI, I. M. 2004. Analisi archeometriche delle ceramiche. In: CASSANO, S. M. & MANFREDINI, A. (eds) Masseria Candelaro. Vita quotidiana e mondo ideologico in una communitgt neolitica del Tavoliere. Claudio Grenzi, Foggia, 221-249. DELL'ANNA, L. 1967. Ricerche su alcune terre rosse della Regione Pugliese. Periodico di Mineralogia, XXXVI(2), 539-592. DELL'ANNA, L. 1969. Ricerche mineralogiche e chimiche sulle 'Argille di Rutigliano'. Periodico di Mineralogia, XXXVIII(3), 515-577. DELL'ANNA, L. & GARAVELLI,C. L. 1968. Su alcune 'terre rosse' della Puglia settentrionale. Grafiche Rossi, Bari. DELL'ANNA, L. & LAV1ANO,R. 1991. Mineralogical and chemical classification of Pleistocene clays from the Lucanian Basin (Southern Italy) for the use in the Italian tile industry. Applied Clay Science, 6, 233-243. DELL'ANNA, L., DI PIERRO, M. & QUAGLIARELLAASCIANO, F. 1973. Le 'terre rosse' delle Grotte di Castellana (Bari). Periodico di Mineralogia, XLII(1-2), 23-67. DONDI, M., FABBRi, B. & LAVIANO,R. 1992. Characteristic of the clays utilized in the brick industry in
62
R. LAVIANO & I. M. MUNTONI
Apulia and Basilicata (Southern Italy). Mineralogica Petrographica Acta, XXXV(A), 181 - 191. ERAMO, G., LAVIANO,R., MUNTONI, I. M. & VOLPE, G. 2004. Late Roman cooking pottery from the Tavoliere area (Southern Italy): raw materials and technological aspects. Journal of Cultural Heritage, 5, 157-165. FABBRI, B., GUALTIERI, S. & SANTORO, S. 1997. L'alternativa chamotte/calcite nella ceramica grezza: prove tecniche. In: Santoro BIANCHI, S. & FABBRI, B. (eds) II contributo delle analisi archeometriche allo studio delle ceramiche grezze e comuni. II rapporto forma/funzione/impasto. University Press Bologna, Imola, 183-190. FRANZINI, M., LEONI, L. & SAITTA, M. 1972. A simple method to evaluate the matrix effects in X-ray fluorescence analysis. X-Ray Spectrometr3', 1, 151-154. FRANZINI, M., LEONI, L. & SA1TTA, M. 1975. Revisione di una metodologia analitica per fluorescenza X, basata sulla correzione completa degli effetti di matrice. Rendiconti della Societh ltaliana di Mineralogia e Petrologia, 31,356-378. GENIOLA, A., LAV1ANO, R. & MUNTONI, I. M. 2005. Pottery production in Late Neolithic cult sites of Santa Barbara and Cala Scizzo (Apulia, Southeast Italy). In: PRUD~.NCIO, M. I., D1AS, M. I. & WAERENBORGH, J. C. (eds) Understanding people through their potter3'. Proceedings of the 7th European meeting on Ancient Ceramics, Lisbon, 27-31 Ottobre 2003, Instituto Portugu6s de Arqueologia, Lisbon, Trabaihos de Arqueologia, 42, 89-101. KOLB, C. C. 1989. Ceramic Ecology, 1988. Current Research on Ceramic Material. British Archaeological Reports, Oxford, 513. KRETZ, R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277-279. LAVIANO, R. & MUNTON1, I. M. 2003. Early and Middle Neolithic pottery production at 'Pulo di Molfetta' (Apulia, Italy): social, chronological and functional implications of raw materials variability, hT: Dt PIERRO, S., SERNEELS, V. & MAGGETTI, M. (eds) Ceramic in the Society. Proceedings of the 6th European Meeting on Ancient Ceramics. University of Fribourg, Fribourg, 163-173. LAVIANO, R. & MUNTONI, I. M. 2004. Le argiile e l'archeometria delle ceramiche. Scelte tecnologiche delle comunita neolitiche in Puglia ricostruite con modeme metodiche analitiche. In: INGRAVALLO, E. (ed.) 11 fare e il suo senso. Dai cacciatori paleo-mesolitici agli agricoltori neolitici. Congedo, Galatina, 113-164. LEONI, L. & SMTTA, M. 1976. Determination of yttrium and niobium on standard silicate rocks by
X-ray fluorescence analysis. X-Ray Spectrometr3", 5, 29-3O. MAGGETTI, M. 1982. Phase analysis and its significance for technology and origin, hi: OLIN, J. S. & FRANKLIN, A. D. (eds) Archaeological Ceramics. Smithsonian Institution Press, Washington, DC, 121-133. MATSON, F. R. (ed.) 1965 Ceramics and Man. Aldine, Chicago, IL. MORESl, M. 1990. Genesi ed evoluzione di depositi argillosi pleistocenici in Puglia. Mineralogica Petrographica Acta, XXXIII, 283-295. MUNTONI, I. M. 1999. From ceramic production to vessel use: a multi-level approach to the Neolithic communities of the Tavoliere (Southern Italy). In: OWEN, L. R. & PORR, M. (eds) Ethno-Analogy and the Reconstruction of Prehistoric Artefact Use and Production. Mo Vince, Tfibingen, 237-254. ML'NTONi, I. M. 2002a. The application of archaeometric analyses to the study of Italian Neolithic pottery: some methodological considerations. In: Archaeometrv in Europe in the Third Millennium. Atti del Com'egno h~ternazionale, Roma, 29-30 Marzo 2001. Accademia Nazionale dei Lincei, Roma. 203-213. ML'NTONI, I. M. 2002b. Le analisi archeometriche di ceramiche neolitiche in Italia: storia degli studi, strategie di campionamento, tecniche analitiche e obiettivi delle ricerche. Origini, XXIV, 165-234. ML'NTONI, I. M. 2003. Modellare l'argilla. Vasai del Neolitico antico e medio nelle Murge pugliesi. Istituto Italiano di Preistoria e Protostoria, Firenze. MUNTONI, I. M. & LAVIANO, R. 2005. La produzione ceramica nel Neolitico antico del Tavoliere (FG): verso un modello di interazione tra le diverse comunit~ di viilaggio, hi: FABBRI,B., GUALTIERI,S. & VOLPE, G. (eds) Tecnologia di lavorazione e impieghi dei manufatti. Atti della 7" Giornata di Archeometria della Ceramica, Lucera, 10-11 Aprile 2003. Edipuglia, Bari, 61-69. MUNTONL I. M., LAVlANO, R. & RADINA, F. 2006. Materie prime e tecnologia di produzione della ceramica 'Serra d'Alto' helle Murge pugliesi. In: FABBRI, B., GUALTIERI, S. & ROMITO, M (eds) l_xt ceramica in Italia quando I'Italia non c'era. Atti della 8" Giornata di Archeometria della Ceramica, Vietri sul Mare, 27-28 Aprile 2004. Edipuglia, Bari, (in press). RICE, P. M. 1981. Evolution of specialized pottery production: a trial model. Current Anthropology, 22(3), 219-240. VAN DER LEEUW, S. E. 1984. Dust to dust: a transformational view of the ceramic cycle. In: VAN DER LEEUW, S. E. & PRITCHARD, A. C. (eds) The Matt3' Dimensions of Potter3,. University of Amsterdam Press, Amsterdam, 707-774.
Late La T~ne pottery from western Switzerland: one regional or several local workshops? MARINO MAGGETTI & GIULIO GALETTI
University of Fribourg, Department of Geosciences, Mineralogy and Petrography, Chemin du Musde 6, CH-1700 Fribourg, Switzerland (e-mail:
[email protected])
Abstract: A total of 203 pieces of fine ceramic and four clays from seven sites of western Switzerland (Bern, Gen6ve, Grotte du Four, La T~ne, Matin, Saint-Triphon-Massongex and Yverdon) were studied chemically and mineralogically to determine if there was local production at each site and if trade links existed between the sites. Firing wasters from Bern and the region of Gen~ve indicate local ceramic production. The sherds are often contaminated with secondary phosphorus and, in the case of Bern, copper. Most of the fine ceramic is CaO-poor, contrasting with the CaO-rich clays. Based on the chromium and nickel concentrations, it can be subdivided into two distinct groups. The majority of the sherd populations from Gen~ve, Saint-Triphon and Massongex, as well as a few specimens from Bern, La T~ne and Yverdon, have high Cr and Ni values. The remaining sherds have low Cr and Ni concentrations. The analyses show that: (1) the fine ceramic from each of the seven sites forms an often inhomogeneous and widely dispersed group, distinct from the others; consequently, it is most probably a local or regional product; (2) ceramic import is probable for one piece from Grotte du Four (provenance Yverdon); (3) the Late La T~ne fine ceramic was manufactured mainly from silicate or silicate-carbonate, fat to lean clays.
Late La T6ne (LT D, c. 150-30 BC) ceramics from the NW of Switzerland (locations: BaselGasfabrik, Basel-Mtinsterhtigel, Sissach-Brtihl, Waldenburg-Gerstelfluh) were analysed by Maggetti & Galetti (1981) and Maggetti et al. (1988). They showed that the ceramic of Basel-Gasfabrik and of the pottery centre of Sissach-Brtihl is a CaO-poor fine ware, which forms two well-defined, homogeneous chemical reference groups. It must therefore be assumed that pottery was produced not only in SissachBriihl, but also in Basel-Gasfabtik. The similar chemical composition of some objects from Basel-Mtinsterhtigel and Waldenburg-Gerstelfluh to these reference groups suggests local ceramic trade. This study focuses on 203 La T~ne fine ceramic fragments from eight other sites in western and southwestern Switzerland, i.e. Bern, Gen~ve, Grotte du Four, La T6ne, Matin, Saint-Triphon, Massongex and Yverdon (Fig. 1). For further comparison, the four clays excavated at the production sites were included in the study. Based on firing wasters, local production was indicated for the sites of Bern and Gen~ve only. As a result, it was necessary to determine if the remaining samples originated from these two production sites, or if they formed different chemically, mineralogically and petrographically distinguishable groups. A
significant enough difference among the seven provenances would point to local ceramic production, i.e. manufacture at different places simultaneously, rather than large isolated workshops in the area of present-day western Switzerland, with regional and national trade.
Samples A total of 203 grey to light grey fine ceramic samples were provided by the archaeologists involved in the project. They belong typologically mainly to bottles and bowls and were made with a wheel. Dating and archaeological literature are as follows.
Bern BE. Middle to Late La T~ne, LT C 1 - L T D2 (c. 250-c. 30 Bc): Mtiller-Beck (19631964), St/ihli (1977), Bacher (1989), Mtiller (1990, 1996), Kohler (1991). Genbve GE. Middle to Late La T~ne, LT C 2 LT D2 (c. 200-c. 30 ~c); Paunier (1981), Kaenel (1990), Bonnet (1997). Grotte du Four (Boudry NE). Middle to Late La T~ne, LT C 2 - L T D2 (c. 2 0 0 - c . 30 BC); Kaenel ( 1991). La Tkne (Marin-Epagnier NE). Middle La T~ne, LT C (c. 2 5 0 - c . 150 BC); Schwab (1989), MUller (1990), Egloff (1991).
From: MAGGETTI,M. & MESSlGA,B. (eds) 2006. Geomaterials in CulturalHeritage. Geological Society, London, Special Publications, 257, 63-80. 0305-8719/06/$15.00 © The Geological Society of London 2006.
64
M. MAGGETrI & G. GALETTI structure types was made, based on the matrix (fat or lean). Clay pellets were not considered in the determination of grain sizes, because they dissolve during soaking.
X-ray diffraction (XRD).
A total of 199 fine ceramic samples were X-rayed under standard conditions (Cu K~, 3-65°20, operating conditions 30 kV and 40 nA) on a Siemens Kristalloflex D500 generator.
!
X-ray fluorescence analysis.
B~lrn Yverdon
Triphon
Fig. 1. Location of the studied sites. Gen~ve = Annecy (France), Dardagny-Brive, Gen~ve, Meinier, Momex and Vandoeuvre-Pressy.
Marin (Marin-Epagnier NE). Late La T~ne, LT DI (c. 150-c. 80 BC); Arnold (1992). Massongex VS. Late La T6ne, LT D (c. 150c. 30 Bc): Haldimann et al. (1991), Curdy et al. (1997). Saint-Triphon (Ollon VD). Late La T6ne, LT D (c. 150-c. 30 BC); Kaenel (1990). Yverdon (Yverdon-les-Bains VD). Middle to Late La T~ne, LT C 2 - L T D2 (c. 200-c. 30 BC); Curdy et al. (1992, 1995), Brunetti (2005). Methods
Powder preparation. For chemical analysis and X-ray diffraction (XRD), 4 - 5 g per sample were ground to a fine powder in a tungsten carbide mill after abrasion of possibly contaminated surface areas. Optical microscopy. Thin sections were prepared when sufficient original material was present. After microscopic analysis, an approximate classification of the 97 samples into
Determination of major elements (Si, Ti, AI, Fe, Mn, Mg, Ca, Na, K, P) and trace elements (Ba, Cr, Cu, Ga, Nb, Ni, Pb, Rb, St, Th, V, Y, Zn, Zr) was performed on all samples. Circular tablets (40 mm diameter) of glassy material (calcined powder) were used for the major elements. Circular tablets of pressed powders (non-calcined) were used for the trace elements. The preparation of the glass tablets was carried out as follows: after calcination of the powdered sample for l h at 1000 ~C, 1.2 g of this calcined material was mixed with 5.7 g of lithium tetraborate and 0.3 g of lithium fluoride. This mixture was then placed in a P t - A u crucible and melted at 1150 °C before being poured into a preheated mould and cooled with compressed air. The preparation of the pressed tablets was carried out as follows: 2.5-3.0 g of initial powder were mixed with < 0 . 3 m l of a moviol-saturated aqueous solution. This was then added to a 32 mm mould and subjected to a pressure of 6 tons for at least 1 rain. The tablet was carefully removed and placed on a bed of 7 g of boric acid in a 40 mm mould. This was again subjected to a pressure 6 tons for at least 1 min. The resulting tablet was dried in a vacuum for 24 h at room temperature. Analytical measurements were performed using a Philips PW 1400 X-ray spectrometer with Cr anticathode. The conversion of the measured values to weight percentage concentrations utilized standardization curves established on reference samples (e.g. USGS, NIM, ANRT). The results of the measurements of the major elements of the matrix were corrected with Philips alpha coefficients.
FeO. Determination of FeO was by the 2.2 dipyridilic method (Lange & Vejdelek 1980), using a Philips Pye-Unicam PU 8650 at 528 nm. Statistics.
Multivariate analyses were performed using SPSS 11.0, neglecting P205 and Cu (contamination effects) as well as Pb and Th (too many blank values) and FeO. For cluster analysis, the raw data (20 elements), average linkage and Ward linkage, and z-scores were
LATE LA TENE POTTERY used. For the factor and the discriminant analyses log-transformed data (20 elements) were used.
Chemical contamination In contrast to the clays, which can contain a maximum of c. 0.2 wt% P205 (Koritnig 1978), ancient ceramics are often characterized by distinctively higher concentrations. In most cases, this can be interpreted as a phenomenon relating to the contamination through migrating P-rich solutions during burial (see Collomb & Maggetti 1996). However, with a concentration of up to 11 wt%, most ceramic pieces from Bern greatly exceed this value. To minimize this secondary phosphorus contamination, all analyses shown in Table 1 were recalculated to 0 wt% P205, and standardized to a total of 100 wt%. These values were used in the following paragraphs. Some Bernese samples show an increased copper content up to a maximum of 2540 ppm. This is due to the use of a Cu-bearing marker for the annotation of these objects (i.e. BE 20, 27, 58, 66, 74, 85 and 92). Copper was therefore not taken into account for the multivariate statistics. In addition to phosphorus, other elements may have been mobile, a possibility that is virtually impossible to confirm. The discussion is therefore based on the assumption that no elements other than P and Cu were affected by secondary processes.
Pottery from the production sites of Bern and Gen~ve Bern A new reference group. A total of 87 samples were analysed (Tables 1 and 2), comprising 85 fine ceramic sherds from two neighbouring excavation sites, Tiefenau/Heiligkreuz and Engemeistergut, as well as two clays, BE 40 from the first excavation site, mixed with sherds, and BE 52 from a trench close to the first excavation site and corresponding to the clayey substratum of the Celtic settlement. The identification of four kiln wasters (BE 38, 44, 49, 73) indicates local ceramic production. The products show a wide scatter in their CaO content, but the bulk of the Bernese fine ceramic is relatively CaOpoor, with a blurred transition to CaO-richer specimens (Fig. 2a). Both clays are very CaOrich and belong therefore to marls rather than to clays (Table 1). Some pottery samples stand out markedly from the main body with regard to one or several chemical parameters (Fig. 2a-d). These are BE 28 (low Zr and Y contents), BE 68 (highest CaO with lowest A1203
65
and Zr values), BE 53 (lowest Na20, MgO and MnO, and highest TiO2, Y and Zr values), and BE 26, 48, 71 (increased Cr and Ni contents). There may be several possible explanations for the lack of compatibility between outliers and the fine ceramic group. (1) The outlier may be of local origin but the amount of sampled material per sherd is insufficient and not representative of a single object; or the outlier may have been affected by secondary contamination during use or burial; or there may be coincidental fluctuation of the chemical composition of the otherwise homogeneous raw materials, i.e. fluctuations that were not eliminated during clay preparation. (2) The outlier is not of local production. For fine ceramic, the population investigated here, the amount of material selected for a single analysis, i.e. 4 - 5 g, is sufficient (Schneider 1989). Also, the contamination hypothesis can be ruled out because elements such as Zr,Y, Mg, etc. are normally not affected by secondary processes. What about the last two possibilities? The use of an illitic-chloritic clay as raw material can be inferred from the negative A1203-SIO2 (Fig. 3a) and the positive A1203TiO2 (Fig. 3b) and AlzO3-FezO3totcorrelations (Fig. 3c). The amounts of calcite vary (Fig. 2a). This substantiates results obtained by XRD phase analysis, revealing relict primary clay phases such as quartz, illite, chlorite, plagioclase and potash feldspar (+_calcite) in the samples with the lowest firing temperatures of c. 500650 °C. Microscopic examination allows for allocation of the samples to three fabrics, as follows. Fabric 1. Very fat, silicate to silicatecarbonate matrix. Small amounts of mostly silicate non-plastic fragments with a maximum grain diameter of 2.52 mm (BE 7, 8, 10, 14, 15, 19-22, 24, 27-30, 34, 35, 39, 41-43, 46-48). Fabric 2. Lean, silicate matrix. Relatively large amounts of silicate non-plastic fragments (sandstone) with a maximum grain diameter of 1.08 mm and a serial granulometry (BE 13, 17, 26, 45). Fabric 3. Carbonate-silicate matrix. Nonplastic constituents with a maximum diameter of 2.23 mm (clays BE 40, 52). CaO-poor as well as CaO-rich (e.g. BE 41 with 13.6 wt% CaO) fine ceramic samples have been allocated to fabric 1, despite distinct variations in CaO, because the matrix does not differ significantly between the specimens. It can be concluded that geologically similar material, with a considerable fluctuation in microcrystalline calcites, has been used in the production of ceramics belonging to fabric 1. The assumption that these raw materials share a similar geological history
66
M. MAGGETTI & G. GALETTI
Table 1. Chemical analyses (wt% ) No.
SiO2 TiO 2 AI203 Fe203* MnO MgO
CaO
Na,O K20 P205
Total
FeO
LOI
H20-
63.77 50.58 52.22 59.05 56.37 60.01 63.18 56.30 51.84 51.41 57.15 52.99 64.51 55.50 62.11 51.78 60.60 47.62 56.33 57.70 59.77 57.20 62.21 45.95 51.24 54.40 51.42 53.12 53.54 49.88 51.53 53.35 47.53 56.70 56.35 44.70 50.77 47.79 62.31 57.78 56.28 60.31 58.34 55.39 64.80 49.12 59.09 63.14 60.54 58.47 47.07 53.67 53.20 55.31 50.79 45.57 54.81 61.59 49.38 54.40 61.18 51.20 52.45 57.51
3.45 12.6 7.31 0.90 7.29 2.93 1.43 2.14 2.01 8.25 3.67 5.08 1.49 2.05 1.18 4.97 2.16 3.58 2.16 1.94 0.96 1.87 1.80 4.90 7.09 !.88 11.50 3.52 1.85 3.01 3.20 2.70 3.96 5.05 16.50 12.60 3.85 6.59 2.82 1.42 2.79 1.67 2.31 2.20 1.40 23.70 2.41 1.36 5.41 9.10 4.46 3.15 1.82 2.13 3.50 9.43 3.87 1.74 5.57 3.48 1.41 18.40 8.53 2.40
1.18 0.96 0.85 0.92 1.02 1.18 1.09 0.99 0.89 0.93 0.92 0.81 1.03 0.93 1.06 0.76 1.13 0.62 1.00 1.25 0.91 0.98 1.30 0.52 0.75 1.02 0.91 0.77 0.78 0.70 0.76 0.80 0.86 0.88 0.96 0.75 0.78 0.56 1.13 0.95 1.22 1.22 1.09 0.99 !.27 1.03 0.20 1.08 1.01 1.00 0.63 0.86 1.01 !.17 0.81 0.54 0.83 1.07 0.79 0.98 1.05 0.93 0.98 1.03
99.93 99.99 99.81 100.00 99.66 100.06 100.89 100.54 100.39 99.81 100.16 99.97 100.14 100.13 100.21 100.56 100.23 99.56 100.23 100.22 99.93 100.27 100.20 99.87 100.50 100.02 99.47 99.85 99.83 99.67 99.49 100.07 99.72 99.81 99.63 99.73 99.94 99.76 99.67 99.73 99.66 100.08 99.87 99.60 100.02 99.13 99.98 99.62 100.42 100.51 99.59 100.11 100.10 100.13 99.95 99.57 100.06 100.40 99.78 99.97 100.51 98.93 99.37 100.11
4.45 1.42 1.89 2.98 1.68 1.37 0.58 3.14 2.81 1.66 4.00 3.30 t.03 1.64 1.13 2.55 1.03 1.89 1.59 1.49 5.37 1.42 0.60 1.59 1.89 1.79 3.24 3.17 2.15 2.62 2.78 2.77 1.35 1.44 1.09 0.99 3.02 1.98 4.45 2.85 0.70 0.55 1.74 0.31 0.19 1.96 0.34 0.72 3.00 1.56 0.51 2.86 0.28 1.75 1.35 2.67 2.40 0.77 1.68 2.93 1.62 1.03 1.92 1.38
1.00 11.00 5.32 2.84 6.38 4.57 2.93 2.88 3.89 4.77 2.77 4.22 4.55 5.11 4.16 4.63 4.70 5.61 4.71 3.49 2.20 5.12 4.87 5.90 4.68 5.34 2.35 5.04 4.54 5.05 4.71 5.79 7.95 5.77 12.5 4.43 6.09 6.94 3.82 2.97 4.43 4.09 4.56 4.72 3.23 18.70 6.96 4.94 1.08 3.14 6.77 5.00 4.06 4.05 6.02 6.51 5.08 4.71 7.05 5.18 4.62 14.70 9.19 5.48
1.40 5.25 6.30 3.84 5.61 5.43 3.52 3.78 5.10 4.39 4.41 5.71 4.22 6.72 3.69 7.42 4.94 8.59 6.14 4.15 3.61 5.62 4.32 9.32 7.07 7.66 3.58 7.16 6.90 7.94 7.41 6.78 10.30 7.79 4.56 5.87 7.69 8.49 3.85 3.23 3.78 4.75 5.12 5.53 3.36 1.46 8.37 3.92 0.79 1.82 6.55 4.84 3.56 3.66 6.55 6.90 5.42 4.15 7.59 5.77 3.89 2.82 5.00 4.86
Bern BE 6 BE7 BE 8 BE9 BE 10 BEll BE 12 BE13 BE14 BE15 BE16 BE 17 BE 18 BE 19 BE20 BE21 BE22 BE23 BE24 BE25 BE 26 BE27 BE 28 BE 29 BE 30 BE31 BE 32 BE33 BE 34 BE35 BE36 BE37 BE38 BE 39 BE40 BE41 BE42 BE43 BE 44 BE 45 BE 46 BE 47 BE48 BE49 BE50 BE52 BE53 BE 54 BE55 BE56 BE57 BE58 BE59 BE60 BE61 BE62 BE63 BE 64 BE65 BE 66 BE67 BE68 BE69 BE70
0.77 0.82 0.88 0.91 0.83 0.77 0.89 0.93 1.00 0.83 0.93 0.90 0.80 0.93 0.84 0.95 0.86 1.02 0.90 0.90 0.92 0.93 0.86 0.99 0.94 0.96 0.84 0.97 0.97 1.00 0.98 0.96 1.04 0.92 0.70 0.91 0.97 0.97 0.83 0.91 0.88 0.87 0.84 0.93 0.78 0.64 1.15 0.82 0.82 0.81 0.99 0.97 1.00 0.93 1.00 0.93 0.93 0.88 0.97 0.91 0.87 0.71 0.86 0.91
17.44 17.66 19.32 20.81 17.58 17.48 18.80 20.66 22.43 19.60 19.23 20.99 18.68 20.77 19.47 21.03 18.56 22.41 20.10 19.79 19.22 20.57 18.42 22.48 20.64 21.18 18.57 21.39 22.17 21.88 21.49 21.43 22.56 20.29 14.23 20.23 21.36 21.17 18.54 20.89 19.75 18.93 18.69 20.78 17.47 12.98 19.15 18.80 18.55 17.54 22.10 21.32 22.18 20.54 21.96 20.87 20.93 19.22 21.00 20.07 19.20 15.20 18.57 19.87
6.47 7.00 7.73 7.98 7.00 7.21 7.39 8.37 9.48 8.50 7.65 7.69 6.07 8.25 7.88 8.17 7.49 8.84 7.97 8.01 8.98 8.26 7.46 8.81 8.11 8.20 7.29 8.22 8.68 8.75 8.30 8.49 8.97 8.18 5.61 7.95 8.59 8.43 7.41 8.32 8.04 7.32 8.03 8.16 6.75 5.12 7.56 7.52 7.37 7.06 8.48 8.22 9.12 8.40 8.67 8.03 8.01 7.61 8.16 7.81 7.65 5.98 7.26 7.90
0.06 0.13 0.15 0.10 0.18 0.13 0.17 0.14 0.14 0.17 0.17 0.13 0.t0 0.19 0.12 0.17 0.18 0.15 0.14 0.13 0.16 0.17 0.09 0.20 0.14 0.15 0.13 0.17 0.15 0.16 0.17 0.15 0.16 0.15 0.12 0.15 0.18 0.16 0.15 0.16 0.14 0.16 0.21 0.22 0.13 0.11 0.06 0.13 0.14 0.13 0.19 0.17 0.16 0.12 0.17 0.17 0.15 0.14 0.19 0.11 0.12 0.11 0.14 0.15
2.44 2.10 2.41 3.18 2.12 1.65 2.41 3.45 3.35 3.85 2.66 3.21 2.00 1.99 2.45 2.67 2.14 2.67 1.97 2.51 3.57 2.24 1.98 2.70 2.59 2.07 2.82 2.69 2.13 2.67 2.80 2.47 2.06 2.64 2.02 3.04 2.62 2.51 2.77 3.18 2.72 1.75 2.69 2.01 1.91 3.67 0.75 2.34 2.73 2.51 2.85 2.81 3.54 2.97 2.51 2.57 2.72 2.57 2.51 2.30 2.31 2.17 2.06 2.36
3.13 2.94 2.58 3.14 2.82 2.41 2.48 2.55 3.10 2.82 2.15 2.56 2.70 3.27 3.30 2.26 3.09 2.25 3.15 3.32 1.71 3.33 2.89 2.31 2.35 3.16 2.48 2.36 2.79 2.16 2.00 2.31 2.77 2.79 2.45 2.17 2.34 2.10 3.53 2.62 3.45 2.62 2.84 3.08 2.98 2.59 2.41 3.21 3.51 3.27 2.15 2.20 2.64 3.08 2.44 1.89 2.22 3.34 2.56 2.23 3.23 2.63 3.00 3.34
1.22 5.21 6.36 3.01 4.45 6.29 3.05 5.01 6.15 3.45 5.63 5.71 2.76 6.25 1.80 7.80 4.02 10.40 6.51 4.67 3.73 4.72 3.19 ll.00 6.65 7.00 3.54 6.64 6.77 9.46 8.26 7.41 9.81 2.21 0.66 7.25 8.48 9.48 0.18 3.50 4.39 5.23 4.83 5.84 2.53 0.14 7.19 1.21 0.36 0.62 10.70 6.73 5.44 5.48 8.11 9.57 5.59 2.25 8.64 7.67 3.50 1.63 5.52 4.64
(Continued)
LATE LA TENE POTTERY
67
Table 1. Continued No.
SiO2 TiO2 A1203 Fe203* MnO MgO
BE71 BE72 BE73 BE74 BE75 BE76 BE77 BE 78 BE79 BE 80 BE81 BE82 BE83 BE84 BE85 BE86 BE87 BE88 BE89 BE90 BE91 BE92 BE93
62.84 50.35 60.97 51.35 54.35 53.55 55.99 51.89 62.24 49.25 48.80 56.27 55.34 50.19 59.06 49.81 56.02 53.47 58.64 50.65 51.25 53.08 55.08
0.75 0.96 0.82 0.97 0.95 0.94 0.94 0.88 0.84 0.99 0.98 0.96 0.95 0.98 0.87 0.93 0.95 0.98 0.93 0.99 0.98 0.97 0.95
18.09 21.10 18.41 21.48 20.78 20.43 21.21 19.45 18.57 22.30 21.97 21.55 21.42 21.77 19.71 20.88 21.39 21.64 20.25 21.74 21.69 21.39 21.38
7.60 7.89 7.35 8.42 8.21 8.10 8.48 7.64 7.14 8.42 8.70 8.48 8.34 8.40 7.61 7.99 8.27 8.55 8.20 8.60 8.56 8.22 8.22
0.16 0.19 0.15 0.17 0.18 0.16 0.23 0.15 0.10 0.18 0.21 0.17 0.15 0.17 0.15 0.14 0.14 0.18 0.14 0.19 0.21 0.19 0.15
2.59 2.44 2.76 2.22 2.23 2.27 2.49 2.71 1.68 2.46 2.42 2.96 2.82 2.92 2.25 2.79 2.04 2.56 2.46 2.82 2.81 2.81 2.01
Gen~ve GEl GE2 GE 3 GE4 GE 5 GE6 GE 7 GE 8 GE 9 GE 10 GE 11 GE 12 GE 13 GE 14 GE 16 GE 17 GE 18 GE 19 GE20 GE21 GE22 GE23 GE24 GE 25 GE26 GE27 GE28 GE29 GE 30 GE31 GE 32 GE 33 GE34 GE39 GE40 GE41 GE42 GE43 GE44 GE45 GE46
64.92 66.12 65.81 67.06 64.38 64.63 64.15 64.96 64.54 65.66 64.80 65.15 64.84 64.79 65.07 65.21 64.76 65.65 65.22 64.84 59.17 67.44 66.68 64.60 66.76 67.45 68.34 67.47 65.51 65.29 65.44 67.96 67.06 65.39 65.01 64.39 64.29 60.45 61.76 64.57 64.93
0.79 0.78 0.78 0.72 0.82 0.80 0.81 0.80 0.79 0.80 0.78 0.79 0.79 0.79 0.79 0.77 0.81 0.78 0.80 0.80 0.75 0.74 0.73 0.78 0.77 0.74 0.70 0.74 0.76 0.78 0.76 0.70 0.76 0.80 0.80 0.81 0.80 0.45 0.42 0.83 0.80
17.88 17.35 17.24 16.42 18.31 18.13 18.18 17.91 17.96 17.86 17.71 17.62 18.01 17.90 17.77 17.26 18.18 17.32 17.98 18.03 17.07 16.71 16.50 17.20 16.94 16.77 16.37 16.65 17.04 17.43 17.05 16.30 16.71 17.93 18.14 18.25 18.08 8.90 8.27 18.72 17.41
7.51 7.23 7.33 7.19 7.69 7.61 7.62 7.51 7.46 7.45 7.41 7.40 7.54 7.52 7.47 7.30 7.62 7.30 7.55 7.50 6.75 7.13 7.26 7.21 7.23 7.18 6.95 7.02 7.29 7.38 7.25 6.91 7.15 7.53 7.59 7.54 7.46 3.33 2.99 7.67 7.57
0.19 0.19 0.19 0.18 0.19 0.18 0.18 0.18 0.18 0.18 0.18 0.17 0.18 0.18 0.18 0.17 0.18 0.17 0.18 0.17 0.13 0.17 0.17 0.14 0.18 0.17 0.16 0.17 0.18 0.17 0.20 0.17 0.19 0.18 0.18 0.18 0.17 0.09 0.09 0.16 0.18
CaO
H20-
Na20 K20 P20~
Total
FeO
LOI
1.18 5.88 5.07 3.38 3.00 3.46 2.93 8.53 1.59 4.48 4.20 2.50 3.67 5.25 1.70 6.16 1.64 3.21 1.62 3.14 3.16 3.32 1.92
1.21 0.65 1.09 0.86 0.99 0.99 0.82 0.87 1.16 0.77 0.75 0.88 0.85 0.69 1.09 0.78 0.88 0.85 1.05 0.71 0.72 0.81 0.95
2.02 2.24 3.43 2.66 3.35 3.13 2.98 2.51 2.66 2.70 2.55 2.44 2.59 2.21 3.25 2.41 2.95 2.59 3.54 2.01 2.02 2.09 3.02
3.77 8.21 0.29 8.20 6.19 6.84 3.96 5.17 4.38 8.40 9.43 3.89 4.33 6.92 4.53 7.66 5.80 6.11 3.58 8.97 8.74 6.73 6.85
100.21 99.89 100.35 99.73 100.23 99.86 100.02 99.80 100.37 99.96 100.02 100.10 100.45 99.49 100.22 99.55 100.08 100.14 100.39 99.83 100.12 99.60 100.52
2.76 1.97 5.49 1.73 1.34 1.83 2.10 2.63 1.45 1.78 2.11 3.99 2.98 2.23 1.41 2.16 1.99 2.36 0.95 2.90 2.84 2.92 1.93
3.66 6.51 0.22 6.12 6.05 6.06 4.61 3.24 3.73 7.88 7.57 3.74 4.93 6.27 5.45 5.15 5.70 6.67 5.29 5.93 5.78 5.55 5.69
3.92 7.67 0.14 7.06 6.53 6.12 5.44 4.18 4.03 7.53 8.33 3.66 4.84 5.77 4.20 5.19 5.22 6.49 4.12 6.22 6.09 5.42 6.08
3.34 0.85 3.35 1.01 3.34 0.72 3.18 0.93 3.72 0.87 3.46 0.79 3.60 0.80 3.42 0.79 3.50 0.87 3.50 0.85 3.44 0.85 3.33 1.01 3.53 0.79 3.46 0.78 3 . 4 1 0.83 3.44 0.81 3.39 0.75 3.17 0.92 3.35 0.96 3.30 0.92 2.71 8.87 3.03 0.74 3.01 1.65 3 . 3 1 2.08 3.33 0.81 3.04 0.78 2.79 0.88 2.84 0.82 3.47 0.71 3.26 1.19 3.62 1.12 2.76 0.88 3.21 0.91 3.42 0.82 3.48 0.85 3.39 0.85 3.33 1.08 2.05 21.6 2.18 21.5 3 . 7 1 0.97 3.71 1.68
1.14 1.19 1.25 1.38 1.13 1.14 1.22 1.08 1.17 1.10 1.19 1.10 1.11 1.10 1.22 1.20 1.01 1.17 1.09 1.21 1.01 1.41 1.31 1.27 1.32 1.21 1.23 1.32 1.19 1.17 1.20 1.28 1.29 1.09 1.08 1.21 1.10 1.10 1.35 1.05 1.29
3.24 2.97 3.35 2.85 3.15 3.22 3.27 3.12 3.29 3.18 3.32 3.31 3.17 3.17 3.24 3.12 3.18 3.09 3.21 3.16 3.21 3.09 2.87 3.37 3.05 2.96 2.89 3.00 3.25 3.30 3.07 2.87 3.00 3.12 3.18 3.32 3.30 1.94 1.54 3.19 2.94
0.18 0.18 0.15 0.14 0.20 0.12 0.14 0.11 0.13 0.15 0.14 0.22 0.10 0.10 0.12 0.18 0.10 0.21 0.18 0.14 0.38 0.13 0.30 0.41 0.11 0.13 0.18 0.16 0.15 0.45 0.21 0.17 0.16 0.12 0.11 0.14 0.27 0.46 0.29 0.11 0.31
100.05 100.37 100.16 100.04 100.46 100.07 99.97 99.88 99.88 100.73 99.84 i00.11 100.07 99.80 99.63 99.47 99.98 99.79 100.51 100.07 100.05 100.59 100.49 100.36 100.51 100.43 100.51 100.19 99.54 100.42 99.92 99.99 100.44 100.40 100.42 100.06 99.88 100.39 100.37 100.07 100.82
5.46 5.75 5.62 5.05 4.66 5.87 5.54 4.24 2.58 5.38 4.63 1.74 3.14 3.48 3.95 4.37 4.39 1.75 1.32 3.97 1.39 0.04 0.05 0.10 1.06 3.75 3.93 3.76 3.24 3.06 2.94 3.80 1.73 4.54 3.07 4.87 3.20 0.49 0.29 4.92 4.82
0.02 0.04 0.00 0.21 0.34 0.00 0.00 0.39 0.85 0.35 0.40 1.98 0.51 0.45 0.68 0.18 0.44 1.92 2.07 1.13 2.29 1.00 1.45 2.88 0.45 0.47 1.36 0.99 0.46 2.55 0.77 1.14 0.87 0.50 0.52 0.60 2.49 14.7 14.8 0.34 1.27
0.11 0.10 0.05 0.15 0.19 0.12 0.09 0.20 0.18 0.18 0.17 0.75 0.20 0.19 0.18 0.11 0.19 0.68 0.87 0.16 0.55 0.24 0.27 1.05 0.10 0.20 0.33 0.27 0.20 1.05 0.26 0.32 0.38 0.22 0.17 0.23 0.76 2.16 1.25 0.19 1.27
(Continued)
68
M. MAGGETTI & G. GALE'Iq'I
Table 1. Continued No.
SiO2 TiO2 A1203 Fe203* MnO MgO
CaO
Na20 K20 P_,Os Total
FeO
LOI
GE47 GE 48 GE49 GE 50 GE51 GE52 GE53 GE 54 GE 55 GE56 GE57 GE58 GE59 GE60 GE61 GE62 GE63 GE 64 GE65 GE 66 GE 67
65.09 64.26 65.47 64.79 64.20 66.21 65.39 67.35 64.43 69.27 71.64 66.40 64.09 66.98 66.44 66.12 64.98 66.92 66.78 65.60 63.65
0.88 0.86 0.71 1.02 0.79 0.84 1.02 0.76 1.76 4.65 3.53 0.92 1.28 0.69 0.76 0.94 1.32 1.68 0.98 0.94 1.46
1.18 1.18 1.25 1.04 1.06 1.25 1.14 1.29 1.22 0.94 i.00 1.24 1.06 1.23 1.18 !.08 1.19 1.37 1.28 1.14 0.96
5.23 2.86 5.74 4.88 6.9l 4.25 5.49 3.69 0.20 3.04 3.03 3.69 4.17 5.08 3.93 3.02 3.35 3.88 4.16 5.29 5.65
0.13 0.78 0.00 0.19 0.00 0.28 0.10 0.57 1.14 1.04 1.38 0.51 1.00 0.63 0.33 4.10 0.50 0.51 0.59 0.28 0.36
18.06 18.56 17.51 18.11 18.67 17.35 17.76 16.98 17.49 14.40 13.46 17.13 18.22 17.01 17.19 17.89 17.28 15.35 16.56 17.46 18.25
7.63 7.77 7.39 7.58 7.81 7.36 7.56 7.13 7.37 5.92 5.36 7.41 7.53 7.48 7.50 8.05 7.40 6.65 7.15 7.46 7.52
0.18 0.21 0.19 0.19 0.18 0.17 0.18 0.20 0.18 0.09 0.14 0.18 0.19 0.17 0.17 0.19 0.20 0.17 0.18 0.18 0.17
Grotte du Four GF 1 62.6 1 . 1 5 21.59 GF2 62.98 0.82 19.01 GF 3 69.97 0.72 15.71
7.50 7.54 6.17
0.09 1.67 0.19 3.37 0 . 1 1 2.18
1.86 0.56 3.38 1.07 1 . 3 8 4.29 1.51 1 . 3 5 2.68
0.27 100.70 2 . 8 1 0.26 100.91 3.79 0.17 100.57 4.06
1 . 7 6 0.48 i.58 0.65 0.59 0.21
La T~ne LT 1 68.60 0.78 LT2 69.33 0.81 LT 3 63.74 0.83 LT4 61.07 0.76 LT5 62.58 0.98 LT6 64.68 0.85 LT7 65.06 0.82 LT8 67.15 0.79
15.72 17.58 18.10 16.77 20.99 18.60 18.34 16.78
4.85 4.73 7.79 6.21 5.60 5.32 6.94 6.89
0.03 0.03 0.25 0.09 0.06 0.06 0.10 0.16
1.99 2.31 3.41 2.32 2.73 2.59 2.90 2.60
4.72 1.71 1.80 8.50 3.51 3.53 1.45 1.52
0.99 1.18 0.83 1.12 0.90 1.09 1.45 1.30
2.84 2.5 3.22 3.I2 2.41 3.27 3.23 3.13
0.17 0.13 0.17 0. t9 0.75 0.15 0.17 0.19
100.69 100.31 100.14 100.15 100.51 100.14 100.46 100.51
2.52 3.04 4.45 2.85 2.98 2.61 3.77 3.29
2.85 3.85 0.91 5.45 4.07 1.76 0.84 0.67
0.67 1.03 0.38 1.27 3.42 0.68 0.25 0.31
Marin ME 1 ME2 ME3 ME4 ME5 ME6 ME 7 ME8 ME9 ME 10 ME l l
25.08 21.94 19.60 15.69 15.85 19.52 23.96 23.94 21.66 15.94 21.85
9.77 8.47 7.89 5.53 7.53 8.65 9.67 10.00 9.53 7.45 8.04
0.12 0.20 0.52 0.54 0.52 0.26 0.15 0.23 0.34 0.49 0.15
3.42 2.56 3.08 1.96 2.16 3.14 3.66 2.69 2.97 1.99 2.85
1.16 1.53 5.66 1.39 1.44 5.75 1.28 1.34 1.81 1.32 2.00
0.86 1.35 0.87 i.16 1.13 0.84 1.25 1.26 1.68 1.12 0.94
2.93 3.35 3.50 2.39 2.51 3.60 2.76 3.65 3.87 2.45 2.00
0.90 1.78 0.61 0.52 1.08 0.94 0.58 1.32 2.13 0.52 1.31
100.39 100.50 100.28 100.75 100.92 100.45 100.45 100.87 100.41 100.63 100.64
1.75 1.68 0.42 0.37 0.69 1.02 4.61 1.19 1.77 0.35 3.43
4.33 4.55 7.32 4.57 3.84 7.25 3.67 5.52 4.40 3.96 4.21
5.60 6.68 3.72 3.75 3.47 3.51 6.45 6.61 5.74 3.79 7.64
16.74 19.14 17.07 17.61 18.79 17.99 18.12 17.39 18.78 16.32 18.32 16.01 17.74 15.94 17.63
7.24 7.85 7.38 7.69 7.46 7.60 7.76 7.64 7.28 6.82 6.97 5.97 6.44 4.79 7.10
0.19 0.17 0.22 0.18 0.14 0.17 0.19 0.17 0.14 0.14 0.11 0.05 0.07 0.04 0.17
3.64 3.08 3.72 4.01 3.13 3.23 3.65 3.66 3.50 2.35 3.43 2.14 2.76 !.66 3.56
1.40 2.18 0.99 1.74 7.39 1.65 1.22 1.84 8.75 8.08 9.82 1.82 6.04 5.42 0.93
1.25 0.91 1.11 1.21 0.90 1.07 1.05 1.21 0.86 1.32 0.99 1.40 1.11 0.73 1.16
3.09 3.20 3.03 3.11 3.45 3.02 3.21 2.99 3.49 2.95 3.46 2.75 3.19 2.31 3.22
0.35 1.78 0.17 0.21 1.61 1.33 0.43 0.42 1.37 1.64 0.29 0.18 0.34 1.27 0.18
99.83 100.13 100.51 100.39 100.02 100.01 100.16 100.44 100.21 99.90 99.70 99.78 100.02 99.82 100.41
4.52 3.37 4.20 3.39 2.86 3.93 2.98 3.17 4.09 3.09 1.19 0.21 0.43 0.24 5.09
0.61 3.69 0.33 0.63 2.53 2.23 1.27 1.20 1.13 3.13 0.54 0.62 0.75 4.38 0.47
0.22 1.76 0.09 0.21 1.33 1.18 0.41 0.39 0.67 1.65 0.43 0.39 0.47 1.34 0.39
55.05 58.39 57.71 70.75 67.87 56.92 56.09 55.41 55.50 68.52 60.56
0.82 0.87 0.78 0.81 0.82 0.77 0.80 0.76 0.79 0.73 0.72 0.77 0.83 0.76 0.76 0.84 0.78 0.70 0.76 0.80 0.81
1.10 0.93 0.84 0.82 0.83 0.83 1.05 1.00 0.92 0.83 0.94
St. Triphon-Massongex TR 1 65.17 0.76 TR2 60.98 0.84 TR3 66.04 0.78 TR4 63.83 0.80 TR5 56.37 0.78 TR6 63.16 0.79 TR7 63.72 0 . 8 1 TR8 64.33 0.79 TR9 55.25 0.79 TR 10 59.57 0.71 TR 11 55.53 0.78 TR 12 68.72 0.74 TR 13 61.55 0.78 TR 14 66.84 0.82 TR 15 65.66 0.80
3.67 3.58 3.45 3.62 3.47 3.37 3.76 3.02 3.70 1.82 1.51 3.38 3.75 3.11 3.14 2.61 3.75 3.51 3.39 3.65 3.44
3.09 3.16 3.21 3.10 3.20 3.13 3.02 3.12 2.95 2.31 2.22 2.94 3.09 2.82 2.96 2.52 3.00 2.79 3.01 3.09 3.16
0.15 0.14 0.16 0.16 0.10 0.20 0.16 0.21 0.21 0.53 1.01 0.14 0.28 0.13 0.10 0.23 0.17 0.41 0.30 0.11 0.15
100.74 100.58 100.12 100.42 100.31 100.67 100.79 100.82 100.08 100.66 100.59 100.49 100.31 100.38 100.20 100.46 100.08 99.54 100.40 100.42 99.59
H200.13 0.78 0.00 0.19 0.00 0.28 0.10 0.57 1.14 0.67 0.77 0.21 0.41 0.38 0.18 3.28 0.15 0.23 0.28 0.18 0.21
(Continued)
LATE LA TI~NE POTTERY
69
Table 1. Continued No.
SiO2 TiO2 Al203 Fe203* MnO MgO CaO Na20 K20 P205 Total
FeO
LOI H20-
Yverdon YV 1 YV2 YV3 YV4 YV5 YV6 YV7 YV 8 YV 9 YV 10 YV 11 YV 12 YV 13 YV 14 YV 15 YV 16 YV 17 YV 18 YV 19 YV20 YV21
59.06 59.50 57.52 56.71 65.71 65.95 60.80 65.52 67.46 66.74 58.42 60.55 60.08 62.71 65.60 67.04 61.13 68.07 67.01 67.24 59.46
3.75 3.82 3.30 1.12 2.90 0.91 4.08 2.92 3.85 3.49 3.08 2.17 4.01 4.14 2.54 0.32 0.42 1.46 1.02 2.29 3.93
0.94 1.72 2.41 3.96 0.69 3.67 2.07 0.83 0.99 0.72 1.17 3.08 1.93 0.96 0.68 2.81 2.38 0.62 4.50 4.20 0.56
0.86 0.80 0.86 0.85 0.79 0.7l 0.80 0.79 0.74 0.77 0.87 0.79 0.80 0.84 0.89 0.95 0.81 0.81 0.66 0.79 0.87
*Total iron is givenas
21.59 20.13 21.66 21.93 17.45 15.83 18.98 17.88 16.55 17.55 22.08 18.92 19.02 19.15 18.04 17.12 19.40 18.22 14.80 16.76 20.72
Fe203.
8.10 7.12 8.38 8.22 6.90 6.65 6.98 6.80 7.01 6.32 8.38 8.26 7.13 7.67 7.21 6.12 7.35 5.79 7.00 6.18 8.07
0.15 0.27 0.16 0.36 0.16 0.13 0.18 0.16 0.32 0.18 0.14 0.14 0.29 0.19 0.17 0.27 0.16 0.10 0.08 0.07 0.16
3.62 1.11 1.19 3.14 3.26 1.14 3.60 1 . 5 9 1.11 3.17 1.73 1.06 2.82 1 . 1 9 1.38 2.I9 1.85 1.51 2 . 9 3 3 . 3 5 1.23 2.80 1 . 0 0 1.33 2.83 1.05 1.35 2.63 1.15 1.41 3.76 1.15 1.00 3 . 0 2 1 . 9 0 1.57 2 . 8 5 3 . 3 6 1.21 3.15 1.37 1.13 2.64 1 . 2 9 1.18 1.21 2 . 0 3 0.36 2.94 1 . 8 3 1.38 2.13 1.29 1.08 1.81 2 . 5 2 1.26 2 . 8 8 1 . 4 9 0.87 3.53 1.64 1.35
4.54 4.23 4.55 4.58 3.43 3.42 3.91 3.69 2.82 3.49 4.57 3.80 3.94 3.56 3.30 3.28 4.14 2.79 2.51 3.30 4.37
0.35 0.96 1.09 1.72 0.22 2.17 0.90 0.32 0.51 0.40 0.28 1.69 1.55 0.63 0.19 1.85 1.58 0.34 2.60 0.81 0.14
100.57 100.55 100.52 100.33 100.05 100.4l 100.06 100.29 100.64 100.64 100.65 100.64 100.23 100.40 100.51 100.23 100.72 100.62 100.25 100.39 100.31
0.44 0.87 1.18 2.37 0.35 2.28 0.86 0.27 0.57 0.45 0.27 1.80 1.05 0.51 0.23 1.48 1.47 0.27 2.93 2.30 0.26
LOI, loss on ignition.
is further substantiated by the observation of sandstone fragments as a characteristic nonplastic phase. As marl BE 40 contains such sandstones, it appears that sandstone fragments must be a specific characteristic of the Bernese clays. The non-plastic elements of fabric 2 differ quantitatively from those of fabric 1. Chemically, the samples belonging to this fabric cannot be distinguished from the ceramic population of fabric 1. The samples are not unusual with regard to their archaeological typology, either. Furthermore, the microscopic image does not indicate any temper addition during manufacturing. Consequently, it may be assumed that the clay used to produce these samples originated from a lean layer embedded in the otherwise fat clay deposit of Bern. In conclusion, it may be possible for BE 26, 28, 48, 53 and 71 to have been manufactured either from a local clay with a different composition from that of the fine ceramic main group, or to have been imported to this La T~ne settlement from a geologically similar or different region. BE 68, on the other hand, can be interpreted as being of local manufacture, because its chemical composition corresponds well to that of the marl BE 40. However, to reduce the size of the reference group, BE 26, 28, 48, 53, 68 and 71 are excluded from the Bernese La T~ne reference group, which comprises therefore 79 specimens. Cluster analysis of the reference group (Ward method, not log-transformed data,
squared Euclidean distances, z-scores) shows that the wasters are distributed over the main subgroups and that the two provenances cannot be differentiated. This argues for a single or different local workshops using chemically variable clay deposits.
Selection and processing of the raw materials. BE 40 and BE 52 are marls and therefore cannot be the starting material for the bulk of the fine ceramic. This is because microscopic examination reveals that the CaO occurs as finely distributed calcite in the matrix, and not as coarse particles in a siliceous clay, which could have been eliminated mechanically during treatment. However, the two marls may well have been used in the production of the CaO-rich samples. Raw material BE 40 contains hardly any non-plastic material and completely resembles the CaO-rich ceramics in this respect. This is probably because natural, fatty raw materials were used without much processing in the Bernese La T~ne ceramic. The wide range of CaO, SiO2 and A1203 concentrations shows that the clay deposit was heterogeneous in its chemical composition, and this heterogeneity was not eliminated by specific material preparation. However, as is to be expected, the fluctuation range is significantly lower in single objects than in the whole population. Five analyses of the same specimen (BE 36, 58, 9 0 - 9 2 ) show that the variations of the major and trace
70 (a)
M. MAGGETTI & G. GALETTI 30
,
,
(b)
,
1.2
I
o
53
1.1 0
0
o
% o
°
do.9
~ 2o~ 15
o
o ~OoC~
~8 ~"o.8
0
5
10
©
15
0.7
20
o
o o o
o
o
to 2
0
MgO (~°/o)
CaO (wt%) (c)
(d) 300
,
1
1
,
,
,
500
' 0
071
400
53
250
0 26
A
E t,~ o.. .....- 200
300 0
-
0 48
0
0 0
200 150
o~
100 ~8 0
100 10
I 20
68 Oi 30
t 40
= 50
I 60
I 70
I
80
Y (ppm)
050
1 O0
150
J
200
I
250
300
Ni (ppm)
Fig. 2. Correlation diagrams of selected oxide and element pairs for 85 fine ceramic samples from Bern.
elements, with the exception of Ba and Cr, lie within the normal range. As there appears to be no correlation between archaeological type and chemical composition, a purposeful selection of a particular raw material for specific objects seems unlikely. Genkve
Sixty-two samples were examined, originating either from the town of Genrve (four excavation sites) or from the surrounding area (five excavation sites). They include two clays (GE 43, found adjoining the kiln Four Rue du Cloffre; GE 44, cob from the cathedral excavations) and 60 fine ceramic
A new reference group.
sherds, of which c. 30% are made up of pottery waste from the Four Rue du Cloi'tre kiln, partly consisting of 13 firing waste fragments (GE 1-8, 49-53). The firing waste GE 67 was found in Annecy (France). As can be observed from the A1203-CaO correlation diagram (Fig. 4a), the fragments are mainly CaO-poor (Table 1). In contrast, both clays and the fine ceramic samples GE 22, 56 and 57 are characterized by very high or at least increased CaO concentrations. Considering the SIO2-A1203 diagram (Fig. 4b), all these samples stand out from the densely packed field of the remaining 57 specimens. This is corroborated by the C r - N i correlation diagram (Fig. 4c), where these outliers plot in isolation
LATE LA TI~NE POTTERY (a)
71
(b) 26
r
24-
O0
0
ooO °O8o
20-
0 0
0
~(~)0
o
18-
•
'
oOo o,i1
'
"oo
J
2o
-
_
O
N
18 16
Geneve
O Bern 4O
I
22
0
16
I
24
BE 41
22-
14
26
T
50
60
•
14 0.6
70
I 0.7
I 0.8
I 0.9
I 1.0
Geneve
I O Bern 1.1 1.2 1.3
T i 0 2 (wt%)
Si02 (wt%) (d)
500
I
I
I
I
I
I
(c) 400
26
T
1
l
~~T--©
24
-
o
o
"~300
e~
A
~ --
O..
22
O0
200
20 18
100 16
~
~
j
~
l
• Geneve © Bern
14 6
7
8
9
10
0 11
F e 2 0 3 tot ( w t % )
0
I
I
I
50
100
150
O Geneve © Bern
200
250
3013
Ni ( p p m )
Fig. 3. Correlation diagrams of selected oxide and element pairs for the reference groups of Bern (n = 79, O) and Gen~ve (n = 57, • ). (a), (b) and (c) The Bernese outliers are otherwise not different enough to postulate a foreign origin. The trends suggest the use of illitic-chloritic clays; (d) The Genevan pottery is Cr- and Ni-richer than the Bernese ceramic.
from the bulk. The samples of the subgroup (GE 5 8 - 6 7 ) with slightly lower Cr and higher Ni than the main Genevan body belong to finds from outside the town itself (Annecy, D a r d a g n y Brive, Meinier, Mornex and VandoeuvrePressy). The following conclusions can be drawn: (1) most of the analysed samples were manufactured from CaO-poor clays; (2) neither of the two analysed marls corresponds chemically to the fine ceramic; (3) 57 fine ceramic
samples form a homogeneous, CaO-poor group; (4) three fine ceramic specimens are CaO-rich (GE 22, 56, 57). It now needs to be determined if the CaO-poor specimens form a homogeneous group when examined by themselves, or if the various specimens from the town and its surroundings differ in terms of their chemical composition. A second question arises with regard to the relationship between the three outliers and the 57 samples.
72
M. MAGGETI'I & G. GALE'Iq'I
(a)
(b) 80
20
15
~8 ' @12 @4 @ 56 @
75 57 @
@
o~" 7O
57
64 @
v
0
(:~ 65 =~ ¢,/)
10
@44@
44 60
43
22 55
I
I
I
10
15
43
20
25
5
I
I
10
15
CaO (wt%) (c)
20
A I 2 0 3 (wt%)
500
400
E 300 O.
0
@22
200 56_057
~43
100
44 0
I
0
50
I
I
100 150 Ni (ppm)
I
200
250
Fig. 4. Correlation diagrams of selected oxide and element pairs for fine ceramic samples from Genbve (n = 60) and two clays (GE 43 and 44).
Cluster analysis shows that some samples from outside Gen~ve (GE 5 8 - 6 7 ) group together. This group contains the waster from Annecy. On one hand, this proves that we are dealing with different groups or different local production centres, but, on the other hand, that a similar type of raw material was used in the production of all analysed fragments. Factor analysis corroborates this statement, despite the fact that the sample number is statistically sufficient for the kiln population only. As the number of analysed sherds is high and the main group homogeneous, the outliers cannot be explained by a non-statistically representative sampling of the Genevan production. There is
no evidence of secondary contamination for outliers GE 22, 56 and 57, because their aberrant elements belong to the so-called immobile elements. It remains unclear, however, if the pieces were imported or produced from local clay with a different composition than that of the fine ceramic main group. Because of the presence of marls, GE 22, 56 and 57 may possibly be local, despite the fact that they do not correspond to the raw materials GE 43 and 44. This could be explained by chemical fluctuations in the clay or marl pit. Based on the previous discussion, a local to regional production of La T~ne fine ceramic by more than one workshop in the area of Gen~ve is indicated and allows for the
LATE LA TENE POTTERY definition of a new reference group of 57 CaOpoor samples.
Selection and processing of the raw materials. The reference group is characterized by a lean, silicate matrix containing fine-grained non-plastic elements, i.e. predominantly quartz and rare, but characteristic ultramafic grains (serpentinite and actinolitic or chloritic-actinolitic fels or schist). The maximum diameter of the nonplastic elements is 0.71 mm, but coarse single grains can be as large as 3.99 ram. Both clay and cottage plaster also show a lean, but carbonate matrix containing many silicate and carbonate non-plastic fragments up to 2.65 mm diameter. Given the chemical and microscopic similarities, it is possible that a very CaO-rich raw material similar to GE 43, present in the immediate proximity of the Rue du Clo~tre kiln, was used as cob (GE 44). For the production of fine pottery, however, a different raw material, poor in CaO and of local but still unknown provenance, was employed. Both analysed marls are extremely rich in CaO, and there may have been too many risks associated with their use (lime spalling). A CaO-richer clay than the one employed for the reference group may have been used very sporadically for the production of some fine ceramic objects (GE 22, 56, 57). Similar to the example from Bern, this kind of clay may be an inhomogeneity within the local clay deposit, because the fabric of these three samples is identical to that of the CaO-poor samples. As shown by microscopic analysis, the Genevan fine ceramic is characterized by a homogeneous structure. Its high Cr and Ni concentrations can be linked to the presence of ultramafic non-plastic grains. In the reduced parts of the samples, these have a light brown to beige colouring, whereas in the oxidized parts they are red to auburn. The nonplastic fragments of the Genevan fine ceramic therefore point to a hinterland with acidic, granitic to gneissic, as well as ultramafic rock types. This is compatible with the catchment area of the Rh6ne and Arve rivers and their glaciers, respectively. In selected correlation diagrams, the Genevan products are significantly more homogeneous and, because the use of a lean raw material, markedly poorer in aluminium (less clay minerals) but richer in silicon (more quartz) than the Bernese reference group (Fig. 3a-c). The Bernese raw material may have contained more chlorite (higher TiO2 and FezO3tot concentrations, Fig. 3b and c), but significantly less chromium and nickel (Fig. 3d). All this is supported by the XRD results showing, for lightly fired samples (i.e. below
73
800 °C), the association of quartz, illite, plagioclase and K-feldspar. According to Figure 3a-c, chlorite could be inferred as a further primary constituent in addition to illite. Furthermore, kaolinite cannot be ruled out, because of the proximity of the excavation site to the Jura mountain belt.
Pottery from other sites Grotte du Four The three analysed fine ceramic specimens belong to the CaO-poor ceramic (Fig. 5a). They vary very significantly in their A1203 concentrations and differ in matrix as well as appearance of their non-plastic elements. GF 1 has a fatty matrix (it has the highest A1203 value) and stands out because of the use of grog, whereas GF 3 has a lean clay similar to GF 2. It contains more and coarser non-plastic elements than the latter (consequently less aluminium). The use of at least two clays (fat and lean) therefore appears likely, whereby an admixture of temper to GF 3 is probable, considering the hiatal structure as well the presence of grog.
La TOne Of the eight analysed fine ceramic specimens, four are very poor in CaO (LT 2, 3, 7 and 8), three poor in CaO (LT 1, 5, 6), and only one, LT 4, is rich in CaO (Fig. 5a). LT 3 stands out from the rest of the CaO-poor group because of its increased Fe203tot, MgO, Cr and Ni concentrations (Tables 1 and 2). There is no evidence of a relationship between typology and chemical composition.
Marin The majority of the 11 analysed fine ceramic specimens are CaO-poor, except for ME 3 and 6 (Fig. 5b). The CaO-poor specimens show, for instance in their A1203 content, a twofold clustering. All these features can be interpreted as evidence for the use of three different clay sources.
Saint-Triphon and Massongex Of the 15 analysed fine ceramic samples, TR 1-14 were found at Saint-Triphon and TR 15 was found at Massongex. Six of them (TR 5, 9, 10, 11, 13, 14) are CaO-rich (Fig. 5c) and show, in contrast to the nine CaO-poor samples, low Cr and Ni concentrations (Table 2). Sample TR 12 differs from the other CaO-poor specimens by its low Cr and Ni contents. Microscopic analyses reveal that the samples were manufactured from two different
74
M. MAGGETTI & G. GALE'Iq'I
(a) (b)
25
+
25
+ 23
23
-h4A
1
21
0 i
O4
19 -
300 points per thin section) and also to overlap with the macroscopic evaluation of the size distribution of paste and aplastic materials (Trinkley & Hacker 1986; Trinkley 1998). In any pointcounting technique the assumption is that the component has a nearly spherical grain shape. With this pottery, the evidence for the presence and abundance of fibre temper is provided by secondary porosity (voids) that contains some carbonized remnants. As the fibre voids are of
CRESCENT SITE FIBRE-TEMPERED CERAMICS two different shapes and orientations, the influence of this orientation may account for some of the percentage differences (and ranges) that were observed. The point count categories used were paste (clay minerals and amorphous phases), quartz (separated by grain size), fibre, feldspar (noted as feldspar unless optical characteristics allowed specific designation as either plagioclase or alkali feldspar), opaque minerals, other (includes epidote/clinozoisite, biotite and amphibole) and ACF (argillaceous clots or fragments of air-dried clay; see Whitbread 1986). Several features are observed in both hand specimen and thin section that may represent compositional or mineralogical changes caused by the firing conditions as well as by use or burial. Oxidation features (commonly a red to red-orange colour) were observed on both the inner and outer sherd surfaces, extending inward for several millimetres. The region between these oxidized zones (called the core) is generally reduced and is either black or smoky grey. Lastly, some of the sherds show secondary carbonate infilling in the fibre void spaces. This mineralization may have resulted from burial and interaction with ground water or may be a result of use.
Petrographic results Table 1 summarizes the petrographic results and textural characteristics of the samples. The rivers of the study area (Fig. 1; e.g. Savannah, Santee, and Altamaha Rivers) drain the igneous and metamorphic basement rocks of the Piedmont region to the west and downcut through the Late Cretaceous to Pleistocene aged sediments of the Coastal Plain (Horton & Zullo 1991; Nystrom et al. 1991; Ward et al. 1991). These Piedmont-draining rivers have a larger capacity and carry relatively large amounts of sediment that is less mature than the reworked (and heavily weathered) sediment carried by Coastal Plain draining rivers (Soller & Mills 1991, p. 299). The Crescent site is located in Pliocene to Pleistocene sediments (and terrace deposits) that were deposited through either the reworking of Piedmont-derived quartz- and feldspar-rich sediments or during transgressive-regressive cycles caused by eustatic sea-level fluctuations (Nystrom et al. 1991; Soller & Mills 1991). In addition, although these sherds are from the southeastern Atlantic coastal region, there was no petrographic or textural evidence for any type of shell material or carbonate rock fragments.
121
A p l a s t i c (temper) c o m p o n e n t s
The dominant aplastic mineral is quartz with a very coarse to very fine grain size. The very coarse to coarse crystals are angular to subangular quartz rock fragments with sutured grain boundaries, undulatory extinction, and little rounding of the corners or edges. The medium grain size quartz is monocrystalline (single crystals) with a blocky (rectangular) to elongate slivers grain shape (Fig. 2). The shape and texture suggest that the mineral grains were broken away from the parent rock fragment. The fine to very fine quartz is also monocrystalline and subangular to blocky in shape. There was very little rounding of the edges of these crystals and no well-rounded crystals were observed. Feldspar is found in very small modal abundance ( 1 8 0 p p m and Ni > 1 0 0 p p m ; High-Cr) and low (Cr I-: :.,: t
,m
°'*
..w
"%4.
"•
t 3 • tl OJeo¢,W~V
$ cb
,lleci~ I,inlo !
FERRARA HISTORICAL BRICKS
~
'a~m twpt
•
lq~
139
'ea:mn~ t
Fig. 8. Scanning electron micrographs (and EDS spectra) of calcium-rich glassy patches within the matrix of sample MT15.
Fig. 10. Scanning electron micrographs of an olivine crystal in an experimental trial obtained by firing a local clay at 800 °C. Chemical analyses of the same crystal by EMPA are also reported.
1 2 3 4 5 6 ~e ~g61 ¢tS Cursor~ ;'o314 ~ V ( g d , )
7
8
9
10
Fig. 9. Scanning electron micrographs (and EDS spectrum) of a euhedral olivine crystal in sample MTI5.
architectural elements. This gives the restorer adequate information for choosing suitable new materials when replacement is necessary and to avoid incorrect restoration materials, such as were used in the Piazza Municipale of Ferrara, where new floor tiles were totally damaged and broken by f r e e z e - t h a w cycles during the first winter.
140
G. BIANCHINI ETAL.
L. Beccaluva and F. Siena are kindly acknowledged for their preliminary review of the manuscript. The authors are also grateful to R. Tassinari (Universit/a di Ferrara) and R. Carampin (CNR-IGG, Padova) for their analytical assistance, and to the reviewers for their constructive comments.
References AMOROSI, A., CENTINEO, M. C., DINELLI, E., LUCCmNI, F. & TATEO, F. 2002. Geochemical and mineralogical variations as indicators of provenance changes in Late Quaternary deposits of SE Po Plain. Sedimentary Geology, 151, 273-292. ARTIOLI, G., BAGNASCO GIANNI, G., BRUNI, S., CARIATI, F., FERMO, P., MORIN, S. & RUSSO, U. 2000. Studio spettroscopico della tecnologia di cottura di ceramiche etrusche dagli scavi di Tarquinia. In: Atti del I Congresso di Archeometria. Patron Editore, Bologna, 335-349. BIANCHINI,G., LAVIANO,R., LOVO, S. & VACCARO,C. 2002a. Chemical-mineralogical characterization of clay sediments around Ferrara (Italy): a tool for an environmental analysis. Applied Clay Science, 21, 165-176. BIANCHINI, G., MARTUCCI, A. &VACCARO,C. 2002b. Petro-archaeometric characterization of 'cotto ferrarese': bricks and terracotta elements from historic buildings of Ferrara. Periodico di Mineralogia, 71, 101-111. BLUM, J. D. & EREL, Y. 1997. Rb-Sr isotope systematics of a granitic soil chronosequence: the importance of biotite weathering. Geochimica et Cosmochimica Acta, 61, 3193-3204. BONDESAN, M., FERRI, R. & STEEANI, M. 1995. Rapporti fra lo sviluppo urbano di Ferrara e l'evoluzione idrografica, sedimentaria e geomorfologica del territorio in Ferrara nel Medioevo. In: Topografia storica e archeologia urbana. Casalecchio di Reno, Bologna, 27-42. BRINDLEY, G. & LEMAITRE, J. 1987. Thermal, oxidation and reduction reactions of clay minerals. In: NEWMAN, A. C. D. (ed.) Chemistry of Clays and Clay Minerals. Mineralogical Society, London, Monograph, 6, 319-370. CAPEL, J., HUERTAS, F. & LINARES, J. 1985. High temperature reactions and use of Bronze Age pottery from La Mancha, Central Spain. Mineralogica et Petrographica Acta, 29-A, 563-575. CHAPMAN, V. J. & CHAPMAN, D. S. 1980. Seaweeds and their Uses. Chapman and Hall, London. CULTRONE, G., SEBASTIAN-PARDO,E., CAZALLA, O., RODRIGUEZ-NAVARRO, C. & DE LA TORRE, M. J. 2000. Mineralogical changes during brick production in laboratory experiments. In: QuarryLaboratory-Monument International Congress, Pavia 2000, Proceedings Volume 1, 253-258. DUMINUCO, P., MESSIGA, B. & RICCARDI, M.P. 1998. Firing processes of natural clays. Some microtextures and related phase compositions. Thermochimica acta, 321, 185-190. ELERT, K., COLTRONE, G., NAVARRO,C. R. & PARDO, E. S. 2003. Durability of bricks used in the conservation of historic buildings--influence of
composition and microstructure. Journal of Cultural Heritage, 4(2) 91-99. FERRI, R. & GIOVANNINI, A. 2000. Analisi dello sviluppo urbanistico della citth di Ferrara nel quadro dell'evoluzione geomorfologica del territorio circostante. In: GALLINA, M. (ed.) Dal Suburbium al Faubourg: evoluzione di una realt~ urbana. ET, Milan, 9-24. LEAKE, B. E., WOOLLEY, A. R., ARPS, C. E. S., et al. 1997. Nomenclature of amphiboles: Report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. Mineralogical Magazine, 61, 295-321. LENTZ, D. R., WALKER, J. A. & ST1RL1NG, J. A. R. 1995. Millstream Cu-Fe skarn deposit: an example of a Cu-bearing magnetite-rich skarn system in northern New Brunswick. Exploration and Mining Geology, 4, 15- 31. LOPEZ-ARCE, P. 8£ GARCIA-GUINEA,J. 2005. Weathering traces in ancient bricks from historic buildings. Building and Environment, 40, 929-941. MAGGETTI, M. t~ WON DER CRONE, M. 2004. Mineral reactions in synthetic clay NaCl system. Abstracts from the 32nd International Geological Congress, Florence, 2004; Session Tl6.01--GeoarcheometO': geomaterials in cultural heritage. MAGGETTI, M., WESTLEY, H. & OLIN, J. S. 1984. Provenance and technical studies of Mexican majolica using elemental and phase analysis. In: LAMBERT, J. B. (ed.) Archaeological Chemistry III. American Chemical Society, Advances in Chemistry Series, 205, 151-191. MARCHESINI, L., AMOROSI, A., CIBIN, U., ZUFFA, G., SPADAFORA, E. & PRETI, D. 2000. Sand composition and sedimentary evolution of a Late Quaternary depositional sequence, Northwestern Adriatic coast, Italy. Journal of Sedimentary Research, 70, 829-838. MARTJNEZ-SERRANO, R. G. 2002. Chemical variations in hydrothermal minerals of the Los Humeros geothermal system, Mexico. Geothermics, 31, 579-612. MOOR, J. N. & GUNDERSON, R. P. 1995. Fluid inclusion and isotopic systematics of an evolving magmatic-hydrothermal system. Geochimica et Cosmochimica Acta, 59, 3887-3907. MURPHY, S. F., BRANTLEY, S. L., BLUM, A. E., WHITE, A. F. & DONe, H. 1998. Chemical weathering in a tropical watershed, Luquillo Mountains, Puerto Rico: II. Rate and mechanism of biotite weathering. Geochimica et Cosmochimica Acta, 62, 227-243. RICCARD1, M. P., MESSIGA, B. & DUMINUCO,P. 1999. An approach to the dynamics of clay firing. Applied Clay Science, 15, 393-409. SINGOYI, B. & ZAW, K. 2001. A petrological and fluid inclusion study of magnetite-scheelite skarn mineralization at Kara, Northwestern Tasmania: implications for ore genesis. Chemical Geology, 173, 239-253. STIAFFINI, D. 1999. 1l vetro nel Medioevo. Fratelli Palombi, Rome.
Golden mica cooking pottery from Giikeyiip (Manisa), Turkey M f J M T A Z (~OLAK 1, M A R I N O M A G G E T T I 2 & G I U L I O G A L E T T I 3
1Dokuz Eyliil University, Department of Geological Engineering, 35100 Bornova, Izmir, Turkey (e-mail: mumtaz, colak @deu. edu. tr) 2University of Fribourg, Department of Geosciences, Mineralogy and Petrography, Ch. du Mus~e 6, CH-1700 Fribourg, Switzerland Abstract: Grkeytip cooking pottery is a particular type of pottery produced according to ancient craft tradition in western Turkey. It is made by mixing 75 wt% of local red and green smectitic clays with 25 wt% of local gneissic temper. Both temper and tempered objects are rich in MgO, as can be seen from XRF analyses. The vessels are coated with a sheet-silicate enriched layer, corresponding to the 9 0 wt%) with other elements such as A1203, K20, CaO, Fe203, etc., each in concentrations of < 5 wt% (Molera et al. 1997; P~rezArantegui 1997). Not a single case of alkaline glaze has been found, so it appears that the production of silica-lead glazes was a technology that was widely diffused in the Iberian Peninsula from the very beginning of the Arab domination. Probably, the fact that Spain has been one of the main producers of galena (PbS) since Roman times has to be considered as a possible reason for this. However, the use of lead glazes offers some technological advantages with respect to alkaline glazes, such as lower fusion temperatures and a characteristic strong shine (Tire et al. 1998). From the structural point of view, the silicalead glasses are formed by a disordered network of SiO 4 tetrahedra with some A1 in the Si structural sites. In this case the A1203 acts as a stabilizer of the glassy structure to equilibrate the electronic charges between Si and other metallic elements (K, Ca, Pb, etc.) and oxygen. The role of the PbO in these glazes is to act as a flux (instead of the alkaline elements in the early Islamic glazes; Mason & Tire 1997). Looking at the PbO-SiO2 phase diagram (Geller et al. 1934, 1943), one can see several eutectic points around 750 ~C for PbO contents between 90 wt% and 70 wt%. However, the presence of small amounts of other elements such as K20, CaO, etc. slightly modifies the eutectic points, and melting can be achieved at such temperatures with lower amounts of PbO (around 50 wt%).
Table 3 presents a summary of the chemical composition of the Islamic and mfdejar glazes studied in this paper, classified by the colour of the different glazes, and Figure 3 shows the SiO2-PbO bivariant relationships for the three time periods (Islamic 10th century, m6dejar 13th century and 14th century). The results are calculated for each sample as the mean of several measured points (7-15) following a profile from the paste to the outer surface of the glaze, and the mean of each sample is used as a single datum point to calculate the statistics of the whole production. Most of the samples studied exhibited a uniform chemical composition following the profiles measured through the glaze (except for the inclusions), but some had profiles as shown in Figure 4. The shape of these profiles suggests that there is a diffusion of elements from the paste to the glaze and from the glaze to the paste, following the chemical gradients of each element. Development o f the g l a z e - c e r a m i c body interface
From several replications of silica-lead glazes made in the laboratory under controlled conditions (Molera et al. 2001) the diffusion of components from the ceramic body to the glaze, and the diffusion of lead from the glaze to the ceramic body has been experimentally demonstrated. Diffusion is more pronounced from an unfired clay body than from already fired one, which is consistent with the fact that the phases developed in the paste after firing are more stable. Figure 5 shows two typical diffusion profiles of fired and unfired bodies with SiO2-PbO glazes of 500 ~m thickness ( 3 0 - 7 0 wt%).
I
t
I
weight %
E .~,
I
55.78 (1.69) 57.00 (4.80)
32.84 (1.58) 32.83 (0.75)
?
54.62 (11.70) 57.68 (15.52) 53.11 (5.61)
32.76 (1.53) 30.28 (1.16) 33.59 (1.09)
I
52.83 (1.38)
33.39 (0.22)
..
N
~7
7o
>
> Z
t" >
172
M. VENDRELL-SAZ ET AL.
(Caiger-Smith 1973). However, glazed pottery and lead-glazed ware were not important in Spain until the expansion of Islam in the western Mediterranean. Following the technological parameters determined for the lead transparent glazes, the technology seems to undergo some kind of simplification. The early Islamic glazes (Murcia and Zaragoza) were applied on pre-fired bodies and after fritting the raw materials (this tradition has been demonstrated to be unnecessary). The Islamic workshop of Denia (13th century) did not frit the raw materials for the transparent glazes and they were applied over unfired bodies (and thus fired in a single operation). This technology was also used in the mddejar workshops of Paterna (13th- 15th centuries). In the Islamic workshops different recipes were used to formulate different colours; however, the mddejar technology once again simplified the process. The islamic colours were achieved by the addition of a colouring element (Cu for green, Fe for yellow, Mn and Fe for brown). In the mddejar workshops studied (particularly Paterna, but also other contemporary sites) the potters used the same glass recipe to produce ceramic glazes of different colours. They obtained the colours by applying the glaze in a different manner (on one side of the pot to obtain yellow, or on both sides to obtain green), or by using different pastes (already used to produce pottery for different uses). These later developments appear to represent a scaling-up of the process, a kind of 'industrialization', which involved a simplification of the recipes, handling of raw materials (no fritting where it could be avoided), simpler application methods, simpler processes of firing (single if possible), etc. The tin-opacified glazes are all lead glazes with tin oxide crystals as particles producing the scattering of the light. The crystal size of the opacifier has been shown to be smaller in the Islamic productions, which should mean some as yet undetermined technological difference. This is possibly a matter of temperature during the preparation of the frits, as there is a high degree of dependence between viscosity, temperature, and crystal nucleation and growth. The heterogeneity observed for the tin opacitier in the later mddejar productions seems to be related to the method of preparation and handling of raw materials; unfortunately, no frit has been found in these historical workshops of the 14th to 15th centuries, and thus for the moment the question remains unsolved. This paper has been partially developed within a project funded by a Ministerio de Ciencia y Tecnologfa (grant BQU2002-03162), and by the research project of the
Comunidad de Trabajo de los Pirineos-Diputaci6n general de Arag6n (CTPR4/2003). The authors wish to acknowledge the supply of the samples by several institutions such as the Museum of Ceramics of Paterna and the Archaeological Services of the cities of Zaragoza, Murcia and Denia.
References AGUAROD, M. C. & ESCUDERO, F. 1991. La industria alfarera del barrio de San Pablo siglos I-XIII). hi: Zaragoza: Prehistoria v Arqueolog{a. Ayuntamiento de Zaragoza, Zaragoza. AGUAROD, M. C., ESCUDERO, F., GALVE, M. P. & MOSTALAC, A. 1991. Nuevas perspectivas de la arqueolog/a medieval urbana del periodo andalus/: la ciudad de Zaragoza (1984-1991). In: Aragrn en la Edad Media IX. Universidad de Zaragoza, Zaragoza, 445-491. ALLAN, J. W. 1973. Abu'l-Qasim's Treatise on Ceramics. Iran, XI, 111 - 120. AMIGUES, F. & MESQU1DA,M. 1993. Les ateliers et la cdramique de Paterna (XIIe-XIVe sikcle). Mus~e Saint Jacques, Ville de Beziers. BAMFORD, C. R. 1977. Colour generation and control in glass. Elsevier, New York. CAIGER-SMITH, A. 1973. Tin-Glaze Potteo' in Europe and the Islamic World: The Tradition of 1,000 Years in Mayrlica, Faience and Delftware. Faber and Faber, London. GELLER, R. F. & BUNTING, E. N. 1943. Report on the systems lead oxide-alumina and lead oxidealumina-silica. Journal of Research of the National Bureau of Standards, 31, 255-270. GELLER, R. F., CREAMER,A. S. & BUNTING,E. N. 1934. The system PbO.SiO2. Journal of Research of the National Bureau of Standards, 13(2), 237-244. GISBERT, J. A. 1990. Los hornos del alfar iskimico de la Av. Montgr/Calle Teulada, casco urbano de Denia (Alicante). In: Fours de potiers et 'testares' mddidvaux en M~diterrande Occidentale. Publicaciones de la Casa de Vel~izquez, Srrie Archrologique, XIII, 75-91. GISBERT, J. A., AZUAR, R. & BURGUERA,V. 1991. La produccirn cer~imica en Daniya. El alfar isl~imico de la Av. Montg6/Calle Teulada (Denia-Alicante). In: Actas del IV Congreso A Cerfmica Medeival do Mediterrrneo Occidental, Portugal, 1987, Mertola, 247-262. JENKINS, F. A. & WHITE, H. E. 1987. Fundamentals of Optics. McGraw-Hill, New York. KANAYA. K. & OKAYAMA,S. J. 1972. Penetration and energy-loss theory of electrons in solid targets. Journal of Physics D: Applied Physics, 5, 43-58. KREIMEYER, R. 1987. Some notes on the firing colour of clay bricks. Applied Clay Science, 2, 175-183. MASON, R. B. & TITE, M. S. 1997. The beginnings of tin-opacification of pottery glazes. Archaeometry, 39(1), 41-58. MESQUIDA, M. 1987. Una terrisseria del s. XIII 1 XIV. Publicacions de l'Ajuntament de Paterna. MOLERA, J. 1996. Evoluci6 mineralrgica i interacci6 de les pastes cgdciques arab els vidrats de plom: interaccions arqueombtriques. PhD thesis, University of Barcelona.
ISLAMIC AND MUDEJAR LEAD GLAZES MOLERA, J., I~ADELL, T., MARTINEZ-MANENT, S. & VENDRELL-SAZ, M. 1993. The growth of sanidine crystals in the lead glazes of Hispano-Moresque pottery. Applied Clay Science, 7, 483-491. MOLERA, J., GARCIA-VALLI~S, M., PRADELL, Z. t~ VENDRELL, M. 1996. Hispano-moresque pottery productions of the fourteenth-century workshop of the Testar del Mol~ (Paterna, Spain). Archaeometry, 38(1), 67- 80. MOLERA, J., VENDRELL-SAZ,M., GARCIA-VALLES,M. & PRADELL,T. 1997. Technology and colour development of Hispano-Moresque lead glazed pottery. Archaeometry, 39, 23-39. MOLERA, J., PRADELL, T. & VENDRELL-SAZ, M., 1998. The colours of Ca-rich ceramic paste: origin and characterization. Applied Clay Science, 13, 187-202. MOLERA, J., PRADELL,T., SALVADO,N. & VENDRELLSAZ, M. 2001. Interactions between clay bodies and lead glazes. Journal of the American Ceramic Society, 84(5), 1120-1128. MOLERA, J., PI~REZ-ARANTEGU1,J. & VENDRELL-SAZ, M. 2005. Chemical and textural characterisation of tin glazes in islamic ceramics from eastern Spain. Journal of Archaeological Science (in press). MOSTALAC, A. 1995. Les fours islamiques de Saragosse. In: Le vert & le brun, de Kairouan Avignon, cdramiques du Xe au XVe sibcle. Rrunion des Musres nationaux, Marseille, 31-32. Muiqoz, P. 1993. Nuevos datos sohre urbanismo y alfarer~a medieval en Murcia. Verdolay, 4, 175-184.
173
NAVARROPALAZON,J. 1990. Los materiales isl~imicos del alfar antiguo de San Nicol~is de Murcia. In: Fours de potiers et 'testares' mddidvaux en Mdditerrande occidentale. Publicaciones de la Casa Vekizquez, Srrie Archrologique, XIII, 29-43. PI~REZ-ARANTEGUI, J. 1997. Les glaqures et les premiers 6maux sur la crramique islamique en al-Andalus (Espagne). TECHNE, 6, 21-24. Pt~REZ-ARANTEGUI, J. • CASTILLO, J. R. 2005. Chemical characterisation of clear lead glazes on Islamic ceramics, produced in Northern al-Andalus (Muslim Spain). Proceedings of the 31st International Symposium on Archaeometry, 1998, Budapest, Hungary. British Archaeological Reports (in press). RAFFMLLAC-DESFOSSE, C. 1994. Cdramiques glagurdes mddidvales. Recherche de donndes physiques sur les techniques de fabrication et altdration. These doctoral, Universit6 Michel de Montaigne, Bordeaux III. ROSSELLO, G. 1995. La crramique verte et brune en alAndalus du Xe au XIIIe si6cle. In: Le vert & le brun, de Kairouan gl Avignon, cdramiques du Xe au XVe sikcle. Rrunion des Musres nationaux, Marseille, 105-117. TITE, M. S., FREESTONE, I., MASON, R., MOLERA, J., VENDRELL-SAZ, M. & WOOD, N. 1998. Lead glazes in antiquity. Methods of production and reasons for use. Archaeometry, 40(2), 241-260. VENDRELL, M., MOLERA, J. t~z TITE, M. S. 1999. Optical behaviour of tin glazes. Archaeometry, 42(2), 325-340.
Archaeometric analyses of game counters from Pompeii R. A R L E T T I t, A. C I A R A L L O 2, S. Q U A R T I E R I 3, G. S A B A T I N O 3 & G. V E Z Z A L I N I ~
1Dipartimento di Scienze della Terra, Largo S. Eufemia, 19, 1-41100 Modena, Italy (e-mail:
[email protected]) 2Soprintendenza Archeologica di Pompei, via Villa dei Misteri, 2, 1-80045 Pompei (NA ), Italy 3Dipartimento di Scienze della Terra, Salita Sperone, 31, 1-98166 Messina, S. Agata, Italy Abstract: Among the glass finds of the Pompeii excavations, numerous objects of opaque
and transparent glassy material of different colours were recovered and classified as game counters. The main aims of this work were to characterize these samples so as to identify the materials used as colorants and opacifying agents, and subsequently to deduce the technology used for their production. The results of the chemical and mineralogical analyses obtained for game counters were also compared with those obtained for transparent and opaque glass artefacts. The chemical analyses were carried out, using only 300 mg of sample, by both wavelength-dispersive electron microprobe and X-ray fluorescence analysis. The crystalline phases present in the opaque glass were identified using both an automatic X-ray powder diffractometer and a Gandolfi camera. Secondary and backscattered electron images were obtained to study the distribution and morphology of the opacifier particles, and qualitative chemical analyses were obtained with an energy-dispersive system. All the game counters analysed can be classified as silica-soda-lime glass. Two calcium antimonates (CaSb206 and Ca2Sb2OT) were identified in the opaque white, green and blue glass, and Pb:Sb207 particles were detected in the opaque yellow glass. Particles of metallic copper were detected by both energy-dispersive system and X-ray powder diffraction. These results support the hypothesis that transparent game counters were obtained by remelting of fragments of common transparent artefacts. In contrast, opaque finds were probably produced using the glassy paste employed in the production of mosaic tesserae.
Roman glass manufacturing reached maximum output in the first to second centuries AD. In fact, Plinius, Martial, Juvenal and other Latin authors of these centuries spoke of abundant and growing glass production, as well as improvements in recycling processes. Pompeii, smothered by volcanic ash, represents a reliable example of the use and habits for this period; only in the Pompeii and Herculaneum excavations is it possible to observe in abundant detail the results of improvements in glassblowing techniques in the first century on Roman tables. Most archaeologists have focused their attention on near eastern production centres, considering Italian production to be of a lower standard. However, several reasons suggest the presence of glass manufacturing in Campania in the first century AD. The region known as Campania felix was not only the residence of renowned philosophers and emperors, but also one of the most thriving and active regions of
the Empire. Pozzuoli harbour represented the principal centre for the supply of foodstuffs and for the transit of goods shipped from Egypt and intended for Rome. Ships loaded with glass fragments and ingots also arrived, as mentioned by Cicerone in his writings. Pozzuoli seems to have been a famous glass production centre, as proven by the discovery of a glass furnace (Gialanella 1999). The presence of a glass production centre near Pozzuoli (or in general in Campania) and the great increase and spread of glass in this period and area is attested by several historical sources (Strabo, Geographia; Petronius, Satyricon). Among the glass finds of the Pompeii excavations, some hundreds of glassy paste objects were recovered and classified as 'game counters'. Plinius, in his Historia Naturalis, defined these items as the result of recycled glass remelting. Only few of these are transparent; most are opaque in a wide range of colours. Hence, they seem to represent a broad pattern of glass
From: MAGGETTI,M. & MESSIGA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 175-186. 0305-8719/06/$15.00 © The Geological Society of London 2006.
176
R. ARLETTI ETAL.
production in the Roman age and their archaeometrical study is certainly of interest, especially concerning the use of colouring and opacifying agents. Coloured opaque glass is among the earliest glass in archaeological records (Newton & Davidson 1989), but these materials did not occur in significant quantities until the middle of the second millennium BC (Mass et al. 2002). Many samples of opaque glass have been analysed recently to identify and characterize the colouring and opacifying agents used, as well as the production technology (see, e.g. Brunet al. 1991; Mass et al. 2002; Mirti et al. 2002; Shortland 2002a). However, such artefacts have never been analysed so far; hence the aim of this study is twofold: (1) to characterize these glass samples so as to define their chemical and mineralogical composition; (2) to understand the technology used for their production. Concerning the latter point, the assertion of Plinius (i.e. the use of recycled glass) is questioned by the paucity of opaque vessels and glassware in Pompeii finds. Along with the game counters, other fragments of more common translucent glass, usually employed for the production of artefacts, and the fragment of one opaque green vessel were sampled, to make a comparison with the materials possibly used to produce game counters.
Experimental methods WDS-X-ray fluorescence analysis The chemical composition of major, minor and trace elements of transparent samples was obtained by wavelength-dispersive spectrometry-X-ray fluorescence (WDS-XRF). By contrast, because of an anomalously high content of some elements such as Pb, Cu, Co and Sb, the opaque samples were studied by electron microprobe analysis (EMPA). For this study an analytical procedure was set up with the purpose of obtaining precise and accurate chemical results for major, minor and trace elements using only 300 mg of sample (Arletti 2005; Arletti et al. 2005). The data were obtained using a Philips PW1480 XRF spectrometer, at the Earth Sciences Department of the University of Modena and Reggio Emilia. The glass was carefully pulverized and mixed with one small drop of organic glue, then pellets with boric acid as the support were prepared by applying a pressure of 7 ton m -z. The major and minor element (Si, Ti, AI, Mn, Mg, Fe, Ca K, Na) concentrations were computed using a program developed by Franzini
& Leoni (1972). The trace element (Nb, Zr, Y, St, Ce, Ba, La, Ni, Co, Cr, V, Sb, Zn, Cu, As, Pb) concentrations were computed using calibration curves (103x c.p.s./element concentrations) obtained after the measurements of 11 silica glass standards (GBW 01-11) of the Institute of Geophysical and Geochemical Exploitation (Langfang, China). To correct the matrix effect of the major constituents on the trace elements, the equations of Leoni & Saitta (1976) were applied. The analytical error for major and minor elements is ~ 40
193
.................
60
..... / ' / >~ 40
.
.
.
.
.
3
.
2o
L) i
-1
0
1 2 3 Grain size (~)
4
-1
PAN
gt.-
8O
60
•
2
3
4
PAN
Monible-COte (F) , -
•
1
Grain size ((~)
Lac Vert (B)
•
0
100
_,,41 . , ~
ER131
........ : : . ~".I~; ,t" •.: .... : ........ : ~ I ~ ' ~"
ER2511
t r
/
8o
g,
~ 60 (1) ~ 40
4o
~ ~o
2 -1
0
1 2 3 Grain size (~)
4
Z
2o
-1
PAN
0
1 2 3 Grain size (~)
4
PAN
La Fuet (G)
Champoz - P. Mont Girod (C) 100 o~
-
/7"
ER140
8o
-
o~ ~. 80
-
. -
ER253
ER254
.... / I
/
I
60
~ 4o
>= 40
"5
20
8 -1
0
1 2 3 Grain size (~)
4
-1
PAN
ForSt de Berole (D)
0
1 2 3 Grain size (~)
4
PAN
Souboz-Montaigu(H)
100
100 ER248
ER255 4 ER256// / /
8o
8o
~ ~o g
~
g
60
.1= ~ 4o
~ 4o
~ ~°
~ O ....
0 Grain size (~)
/
20 0
-1
0
1 2 3 Grain size (~)
Fig. 3. Cumulative grain-size frequency curves of the Hupper sand samples (sieve analysis).
,
,
4
PAN
194
G. ERAMO
Table 4. Grain-size data for the refractor)." and crucible samples by thin-section analysis (vol% ) 05: mm:
< - 1 >2
- 1-0 1-2
0-1 0.5 - 1
1-2 0.25 -0.5
2-3 0.125 -0.25
3-4 0.063 -0.125
>4 < 0.063
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.09 0.00 0.77 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.20 0.00 0.00 0.25 0.00 0.00 0.00 0.27 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.72 1.71 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00 0.00 0.10 0.30
3.92 1.04 1.71 1.93 2.25 4.08 2.96 4.64 3.50 1.38 2.64 1.88 1.09 0.39 0.39 2.15 0.86 2.75 3.96 1.99 1.43 0.98 1.43 0.98 1.28 3.57 0.51 2.51 1.62 2.66 2.91 0.72 0.57 0.91 1.48 1.69 1.47 1.98 1.13 0.93 0.56 1.19 0.97 1.84 1.11
8.81 12.16 7.62 10.89 13.30 6.21 11.50 12.96 8.47 10.80 7.91 4.33 3.83 7.56 5.10 10.67 6.39 9.17 9.66 7.40 6.51 5.69 9.17 3.54 4.95 5.35 8.81 5.20 7.94 6.45 4.73 4.15 7.60 6.92 3.87 4.14 7.56 5.49 7.16 8.55 6.40 5.25 4.65 7.32 2.58
16.15 22.51 9.14 27.12 24.75 21.17 23.98 28.43 26.34 33.45 21.66 29.94 25.32 26.16 23.92 27.25 21.59 21.83 28.15 22.56 21.75 26.67 33.05 25.20 27.11 23.53 22.37 16.67 21.48 22.77 20.18 14.26 25.67 26.05 29.52 21.09 29.46 28.51 26.74 31.12 19.59 21.02 25.78 24.21 4.80
26.10 26.50 12.95 19.32 14.85 31.26 17.48 23.98 29.10 23.43 23.16 23.35 23.68 26.16 41.18 19.86 42.31 29.54 18.14 34.84 31.11 32.55 22.79 36.61 33.88 27.81 32.88 20.79 25.27 32.26 24.00 19.49 33.08 32.24 33.76 25.05 24.79 39.02 32.39 25.11 39.36 37.63 38.57 28.08 7.29
45.02 37.70 68.57 39.98 44.85 37.09 44.08 29.98 32.60 30.93 44.63 40.49 46.08 39.73 29.41 39.87 28.84 36.70 39.84 33.21 39.21 34.12 33.29 33.66 32.78 39.75 35.42 54.84 43.68 35.86 48.18 60.65 31.37 33.88 31.37 48.02 36.72 25.00 32.58 34.20 34.09 34.92 30.04 38.45 8.53
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
2.23 2.11 1.47 2.64 4.85 1.60 1.20 1.52 1.51 0.53
4.28 6.01 7.89 6.78 5.04 8.53 5.60 7.77 8.47 5.49
19.74 22.24 30.09 28.44 30.22 21.31 22.40 21.79 22.22 23.72
31.47 30.68 28.81 37.66 32.09 22.91 21.80 24.83 24.11 30.27
42.27 38.96 31.74 24.48 27.80 45.65 49.00 44.09 43.69 40.00
Crucible fragments ER21 ER22 ER23 ER24 ER25 ER26 ER27 ER28 ER29 ER30 ER31 ER32 ER33 ER34 ER35 ER36 ER37 ER38 ER39 ER40 ER41 ER42 ER43 ER44 ER45 ER46 ER47 ER48 ER49 ER50 ER51 ER52 ER53 ER54 ER55 ER56 ER57 ER58 ER59 ER60 ER61 ER62 ER65 mean cr
Refractory fragments ER63 ER64 ER66 ER67 ER68 ER69 ER85 ER86 ER87 ER88
(Continued)
PRE-INDUSTRIAL GLASSMAKING, SWISS JURA
195
Table 4. Continued
4': mm: ER89 ER90 ER91 ERI02 ER103 ER267 ER276 ER277 ER278 ER279 ER280 ER281 mean O"
< - 1 >2
- 1-0 1-2
0-1 0.5-1
1-2 0.25-0.5
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
1.49
0.00 0.00 0.59 1.91 0.95 0.36 0.54 0.70 3.02 0.55 3.91 1.53 1.24
100
Crucibles (n = 43) ~
80
¢-
~ 60 ~ 4o E 20 O
a
-'1
6
i
½
3
4
PAN
2-3 0.125-0.25
3-4 0.063-0.125
>4
,..] >" 0
t'rl
O 7~
>
> t" O
O
O
Z: >
>, 7Z
Barda Blanca Pasarela, sample 2 Barda Blanca 4
Site, level
4.4
4.2
4
4.3
8.1
4.3
292
118
274
286
169
272
6.0
8.0
4.3
4.4
Cs
165
288
293
Rb
b~s Alerces National Park 25 Laguna La 128 Larga, sample I
Cholila 18 Juncal de Calder6n I 19 Juncal de Calder6n 2 20 Cerro Pintado, Level 4 21 Cerro Pintado, Level 2 22 Los Guanacos 3, sample I 23 Los Guanacos 3, sample 2 24 Hallazgo Aislado
17
16
Sample no,
T a b l e 3. C o n t i n u e d
923
10
252
11
II
818
18
259
10
11
Ba
60
4
46
4
5
46
3
42
3
4
Sr
Ga
16
0.7
18
0.68 29
1.4
0.75 30
0.53 29
132
Nb
3.5
75.6
1.9
19
8.99 l l0
3.04
Zr
4
14 94
405
5.1 163
1 3 . 7 364
13.7 342
Hf
4.0 I07
13.9 364
5.2 148
13.6 372
8 5 . 1 13.5 344
15
9.04 136
9.01
1.62
127
26
9.21 137
9.3
Ta
29 II
15
0.44 18
0.9
1.1
0.57 32
0.81 31
TI
Trace elements (t-tg g i) Th
9.17
38
9.5
53 24.3
19 26.7
53 25.2
50 24.2
33
57 27
19 28
52 25.4
52 25.6
Y
86
34
88.3
83.1
La
2.1
6.47
7.63
6.7
6.47
25
83.5
32.5
82.8
85.2
2 . 3 1 25.1
6.1
6.9
6.71
6.8
U
51
158
56.2
158
160
50.7
164
57
166
159
Ce
5.64
17.6
5.71
17.5
17.9
6.11
16.8
5.46
18.4
17.9
Pr
24
60
18
58
59
22
64
19
62
60
Nd
0.09
7.5
0.09 10
5.7 (}.48
12
1.7
0.4
1.9
0,6
1.81
5.0
2.0
0.77
0.35
5.14 0.78
0.78
Tm
1.79 0.33
1.0
6.2
1.2
3.9
0.66
1.77 9.74 1.82 5,08 0.78
2.75 0,58
4.88 0.77
9.58 1.81 4.95 0.76 1 . 7 6 9.72 1.8
1.79
Er 1.88 5.2
Ho
5.56 1,15 3.53 0.58
10
2.9
1.77 l0
2.41 0.44
9.9
0.09 10.2
3.2 0.36
12
12
Dy 1 . 8 7 10.2
Tb
4.95 0.91
0.42 10
2.6
0.09 10.2
5.1 0.64
12
Gd
0.09 10.3
Eu
3.4 0.63
13
12
Sm
Rare earth elements (Ixg g J) Lu
4.3 0.80
4.6 0.63
2.3 0.36
4.4 0.61
4.4 0.61
3.7 0.59
4.4 0.54
2.7 0.58
4.5 0.62
4.7 0.62
Yb
t" t"
t-" t-"
Cafiad6n Salamanca
LagunaLa Larga, sample 2 LagunaLa Larga, sample 3
5.5
44
1.2
1.9
80
3.2
678 121
0.4
14
0.88 15
0.7
3.02
0.07
31 13
56 41
3
18
8
36.1
36.7
141
471
1 . 5 1 35.6
1.9
35
36
8.82
9.2
32
353 1.16
5.9 223
35 11
18 26.8
5
64 32
752 232 80
98
5.2 142
0.8
13
78
3.7
3.8 103
2.4
7.58
0.8
7.9
18
2.22
2.1
32
35
1.52
78
208
24
26
64
59.4
3.33
153
334
48.2
51
7.02
6.01
0.47
15.6
46.3
5.82
5.55
30
19
2.5
60
176
22
25
0.16
1.9
7.0
6.7 1.08
7.2
1.0
2.47 0.46
0.9
0.42 12
3.3 0.38
1.0
4.87 0.92
7.2
3.85 44
0.9 0.26
13
42
5.1 0.61
5.5 0.58
6.0
2.8
3.9
0.65
2.0
7.6
1.2
0.58
0.96
3.8
0.59
1.87 0.35
4.3 0,85
2.4 0,37
0.8 0.11
5.5 0.76
2.69
3.7 0.56
4.3 0.83
3.23 19
0.57 0.1
5.9
21
1 . 1 1 3.36 0.58
1.2
0.96 0.2
11
41
5.4
6.0
z.
> C~ 0
>
0
7~
0
©
7Z 9 3~
0
Volc~inChait6n
47
30 - 0 . 0 5
4
17
0.79 17
1.0
~
263
50
14
7 -2
1044 123
849
Chile 32
7.7
0.3
4
156
5.6
9.1
325
585
193 35
160
Portada Covunco 31 Portada Covunco
Sacanana 29 Cerro Guacho R{o Villegas 30 El Rinc6n
Telsen 28
27
26
252
C. BELLELLIETAL.
1000
1
'
u.m~o,.,,2 t
¢
r..,ou~A G ~ 8
~++
100
(a) i
(b) ,
i
....
lO
t
i
i
100
1000
p,b~)
I
,
,
,
,
,
,,1
i
i
,
i
,
,,
/-
rh 1~9)
Fig. 2. Plots of geochemical REE and trace element values.
Group C. Archaeological samples 1-3, 6 - 1 0 and 15-17 from Piedra Parada, and samples 19, 21, 22 and 24 from the Cholila region come from the source located in Sacanana (sample 29), 230 km and 160 km respectively from both archaeological areas. Group D. Isolated find recovered in the Piedra Parada valley (perlitized samples 13 and 14 from Angostura Blanca) are identical and also show the same values for some elements, (Rb, K, Nb, Ce, Eu and Ti). There is a correspondence between these samples and archaeological tool samples 11 and 12. Perlitized samples 4 and 5 from Piedra Parada valley do not have clear correspondence to any known source or archaeological sample. In addition, they each have different values. Therefore we have called them 'Unknown 1' and 'Unknown 2'; nevertheless, the former has a slight resemblance to Group D. Samples 28, 30 and 32 belonging to the sources Telsen (east of Somuncura massif), Rio Villegas and Chait6n volcano (Chile) have no geochemical relationships with any archaeological sample analysed here. Finally, samples that belong to Group C show high percentages of silica in non-perlitized samples. The Si-O2 contents exceed 75%, with
high contents of K and Na and comparatively low contents of Al and Ca. Consequently, they are vitreous rhyolites with an alkaline tendency, metaluminous, that fall in the field of high-K rhyolites (Tables 2 and 3). On the other hand, Group D (Table 1) shows a lower content of SiO2, although it is still high, and it has a slightly less alkaline tendency despite the fact that it also corresponds to high-K rhyolites (Tables 2 and 3). Group A samples have lower SiO2 contents, but looking at the LOI values it is possible to relate this fact to a higher degree of perlitization. Group B has similar features to Group C (highK rhyolites, slightly alkaline and metaluminous).
Discussion and conclusions Obsidian may be characterized as one of the best quality raw materials for the knapping of lithic tools and this is the major reason why it has been widely used in the past. In Patagonia, it is a scarce rock that appears in precise locations. Its geochemical characteristics allow a chemical fingerprint analysis. It is one of the few rocks that offers the possibility of establishing potential sources and origin of tools within statistical limits.
ARCHAEOLOGICAL OBSIDIAN SOURCES, PATAGONIA Archaeological obsidian tools from several stratified and surface sites of Piedra Parada could be identified as deriving from the Sacanana source (Group C), 160 km NE from Piedra Parada, in the Somuncura massif. Additionally, there are two samples from two sites in different environments (Campo Nassif 1 in the river valley and Bajada del Tigre in the high fields, 13 km apart) that are identical and are related to perlitized samples found in Angostura Blanca (Group D). In a previous paper, we called them 'unknown X' (Bellelli & Pereyra 2002), but after the analysis presented here, it is certain that the tool samples come from somewhere near Angostura Blanca, and therefore have a local origin. Cholila archaeological samples show the use of three sources: Groups A, B and C. The Sacanana source (Group A) is 230 km to the east and the Portada Covunco source is more than 400 km to the north (Group B). Previously (Bellelli & Pereyra 2002) the latter group was labelled 'unknown Y'. Three facts lead us to consider the existence of good knapping quality obsidian rocks in the region: the results showing similarities between the archaeological sample from Cholila and the perlitized samples from Laguna La Larga (50 km to the south) (Group A), the geological data and the references of Arrigoni (1999). At present, Laguna La Larga would be the only local source regionally used, as the archaeological obsidians in groups B and C come from very distant sources. Summarizing, in the research area, obsidian rocks from four sources were used. Sacanana is the only source represented in both archaeological areas (Piedra Parada and Cholila) and also is the most frequently used. This source supplied archaeological sites located in the centre, west and east of Chubut, such as Cerro Castillo (90 km NW of the source), Los Altares (more than 100 km south), Las Plumas (more than 180km SE) (Stern et al. 2000, p. 288) and several locations on the Atlantic coast (between 270 km and 400 km east of the source) (Gdmez Otero & Stern 2005). Other sources that were taken into account in this paper are not present in analysed archaeological assemblages. This is the case for the source called Telsen by Stern et al. (2000), in the Somuncura massif, which seems to make a restricted raw material contribution to the centre-east of Chubut (Las Plumas and Peninsula de Vald~s) (G6mez Otero & Stern 2005). Also, in Argentine northwestern Patagonia there is no archaeological evidence for the use of obsidian from Chaitrn volcano, on the western slope of the Andes.
253
Moreover, the sample from Rio Villegas shows different characteristics from the rest and therefore it could not be classified. This obsidian has very poor knapping quality and a high water content. Identified obsidians appear to be associated with two petrotectonic environments: (1) an arc environment, located in the western area of Argentina and Chile, which is younger (Chait~n volcano, Chile; Laguna La Larga, Los Alerces National Park, Chubut); (2) an intraplate environment, located in the Northpatagonic Massif, to the east of the surveyed area, which is older (Sacanana, Telsen, Portada Covunco and Angostura Blanca). The sources that belong to this latter environment, based on the frequency of stone flakes and tools taken as samples in archaeological sites, seem to have been used more intensively. Volcanism in this region began in the Eocene and continued until the Pliocene. Associated pyroclastic rocks are included in the Miocene age 'Complejo Eruptivo Quifielef' and Somuncura Formation. Regarding the chronological framework, the use of obsidian (sparsely represented in the archaeological record of both areas) has been recorded in the contexts of Piedra Parada with ages of 3200 years or younger, and was absent in the previous period (5000-3200 years ago) (Bellelli 1988; Prrez de Micou et al. 1992) (the brittle fragments recovered in these contexts were not archaeological tools: Unknown 1 and 2). One of the samples analysed here (from the Sacanana source, Group C) comes from layer 2c of Campo Moncada 2, dated 3210 + 50 years Bp (UGA7621) (Prrez de Micou 2002). This indicates for the first time that this raw material was used in the research area. Its use continued until 480 + 75 years Be (Onetto 1986-1987) in Campo Nassif 1, where the sample from Angostura Blanca was collected (Group D). Regarding the Cholila area, the oldest occupations are dated 1900 years ago. The use of raw material belonging to Groups A (local) and C (non-local) was recorded since this time at the stratified site Cerro Pintado and at surface sites. In two of the latter, obsidian from Portada Covunco (Group B) was also observed.
Concluding remarks With ages from 3200 years onwards, plant-based manufactures and remains of non-local plants (from the Andean-Patagonia mountain range, where Cholila is located) obsidians are found in the archaeological contexts of Piedra Parada (P~rez de Micou et al. 1992; Marconetto 2002).
254
C. BELLELLI ETAL.
W e believe that an exchange of ideas, information, goods a n d / o r groups and a more intensive occupation of space after that date would be taking place in northwestern Patagonia, based on long-term networks of environmental knowledge, resources and communication channels. In addition to obsidian, in this period certain goods in 'prestige technologies' (sensu H a y d e n 1998) were widely circulated in North Patagonia. These items are manufactured with special care and are highly curated. A m o n g these are engraved slate plates, necklace beads, egg shells, w o o d e n and bone tools, all of them decorated, which have been recovered in both research areas. The Andes mountain range, because of its topography and its biophysical conditions (deciduous forest and dense Valdivian forest to the west) has conditioned the circulation of goods, people, information and ideas. That the Chaitrn Volcano source is not represented in the analysed samples supports this idea, and is an important clue in determining the circulation of goods on both sides of the Andes. This research has been supported by Fundaci6n Antorchas and Consejo Nacional de Investigaciones Cientfficas y Trcnicas. Special thanks are due to the two reviewers. who offered valuable suggestions. We appreciate the editors' help and patience. We also thank G. Gur~ieb, who kindly helped us with the translation.
References ACTIVATION LABORATORIES LTD. 2001. Fee Schedule. Quality Analysis. hmovative Technologies. Activation Laboratories Ltd., Ancaster, Ont. ACTIVATION LABORATORIES LTD. 2005. Techniques: Trace Elements Analysis. http://www.actlabs. com/trace_elem_analysis.htm. ARAGON, E. & MAZZONI, M. M. 1997. Geologfa y estratigraffa del complejo pirocl~istico del rfo Chubut medio (Eoceno), Chubut, Argentina. Revista de la Asociacirn Geolrgica Argentina, 52(3), 243-256. ARAGON, E., GONZ,~LEZ, P., AGUILERA, Y., MARQUETTI, C., CAVAROZZI, C. ~:; RIBOT, A. 2004. E1 domo vitroffrico Escuela Piedra Parada del Complejo Volc~nico piroclfistico del rio Chubut Medio. Revista de la Asociaci6n Geolrgica Argentina, 59(4), 634-642. ARDOLINO, A. &; FRANCHI, M. 1993. Hoja Geolrgica 4366-1 Telsen. Direcci6n Nacional del Servicio Geol6gico, Buenos Aires, Boleffn 215. ARRIGON1, G. 1999. An~ilisis del territorio de explotaci6n de los recursos b~isicos para la subsistencia de los grupos prehistrricos que habitaron el valle del r/o Desaguadero. Parque Nacional Los Alerces. Chubut. Abstracts XIII Congreso Nacional de Arqueolog[a Argentina, C6rdoba.
BELLELLI, C. 1987. El componente de las capas 3a, 3b y 4a de Campo Moncada 2 (CM2) (Pcia. del Chubut) y sus relaciones con las industrias laminares de Patagonia Central. Comunicaciones. In: 1 Jornadas de Arqueologfa de la Patagonia, Serie Humanidades. Gobierno de la Provincia del Chubut, Rawson, 27-32. BELLELLI, C. 1988. Recursos minerales: su estrategia de aprovisionamiento en los niveles tempranos de Campo Moncada 2, Valle de Piedra Parada, Rio Chubut. In: YACOBACCIO, H. (ed.) Arqueolog{a Contemporrnea Argentina. Btisqueda, Buenos Aires, 147- 176. BELLELLI, C. 1994. Excavaciones en Campo Cerda 1 (Valle medio dei rio Chubut). Actas v Memorias del XI Congreso Nacional de Arqueolog{a Argentina (Res~hnenes). Revista del Museo de Historia Natural de San Rafael, Mendoza, 14(1-4), 285287. BELLELLI, C. 2002. Dataciones por AMS de artefactos realizados con trcnicas cesteras en Campo Cerda 1 (Valle de Piedra Parada, Chubut). Cuadernos del Instituto Nacional de Antropolog{a, 19, 660-662. BELLELLI, C. & PEREYRA, F. X. 2002. An~ilisis geoqufmicos de obsidiana: distribuciones, fuentes y artefactos arqueolrgicos en el Noroeste del Chubut (Patagonia argentina). Werken, 3, 99-118. BELLELLI, C., PEREYRA, F. X., FERN,~NDEZ, P., SCHEINSOHN, V. & CARBALLIDO,M. 2000. Aproximaci6n geoarqueol6gica del sector sur de la Comarca Andina del Paralelo 42:S (Cholila, Chubut). Cuaternario y Ciencias Ambientales. Publicaci6n Especial CADINCUA-COMINCUA, Buenos Aires, 1, 15-21. BELLELLI, C., CARBALLIDO, M., FERNANDEZ, P. & SCHEINSOHN, V. 2003. El pasado entre las hojas. Nueva informaci6n arqueol6gica del noroeste de la provincia del Chubut, Argentina. Werken, 4, 25 -42. CENTER FOR EARTH RESOURCES RESEARCH 1987-1993. NewPet (~. Memorial University of Newfoundland, St. John's. ESCOLA, P., V~.ZQUEZ,C. • MOMO, F. 2000. An~ilisis de procedencia de artefactos de obsidiana: una vfa metodolrgica de acercamiento al intercambio. Arqueolog(a Contemporrnea, 6, 11-32. GLASCOCK, M. D., BRASWELL,G. E. & COBEAN, R. H. 1998. A systematic approach to obsidian source characterization. In: SHACKLEY, M. S. (ed.) Archaeological Obsidian Studies. Plenum, New York, 15-65. GOMEZ OTERO, J. c~¢ STERN, CH. 2005. Circulaci6n, intercambio y uso de obsidianas en la costa de la provincia del Chubut (Patagonia, Argentina) durante el hoioceno tardio. Intersecciones, 6, 93-108. HAYDEN, B. 1998. Practical and prestige technologies: the evolution of material systems. Journal of Archaeological Method and Theory, 5(1), 1-55. LAGE, J. 1982. Descripci6n geolrgica de la hoja 43c, Gualjaina. provincia del Chubut. Servicio Geol6gico Nacional, Buenos Aires, Boletfn 189. LIZUAfN, A. 2005. Hoja Geoldgica Esquel. SEGEMAR-IGRM, informe inrdito.
ARCHAEOLOGICAL OBSIDIAN SOURCES, PATAGONIA MARCONETTO, M. B. 2002. Anfilisis de los vestigios de combusti6n de los sitios Alero Don Santiago y Campo Moncada. In: PEREZ DE MIcou, C. (ed.) Plantas y cazadores en Patagonia. Facultad de Filosoffa y Letras, Universidad de Buenos Aires, Buenos Aires, 33-53. ONETTO, M. 1986-1987. Nuevos resultados de las investigaciones en Campo Nassif 1. Valle de Piedra Parada, Provincia del Chubut. Relaciones de la Sociedad Argentina de Antropolog(a, 17(1), 95-121. PI~REZ,A. & LOPEZ, L. G. 2004. Obsidianas Lolog. Una cantera de obsidiana en el bosque meridional neuquino. In: TAMAGNIM, M. & MENDON~A, O. (coord.) Publicacidn de rest~menes del XV Congreso Nacional de Arqueologfa Argentina. Universidad Nacional de Rfo Cuarto, Rfo Cuarto, 415. PEREZ DE MICOU, C. 2002. Tecnologfa cestera en Patagonia. Fechando artefactos. In: PEREZ DE MICOU, C. (ed.) Plantas y cazadores en Patagonia. Facultad de Filosoffa y Letras, Universidad de Buenos Aires, Buenos Aires, 55-63. PI~REZ DE MICOU, C., BELLELLI, C. & ASCHERO, C. 1992. Vestigios minerales y vegetales en la determinacidn del temtorio de explotaci6n de un sitio. In: BORRERO, L. A. & LANATA, J. L. (eds) An6lisis
255
Espacial en la Arqueolog{a Patag6nica. Ayllu, Buenos Aires, 53-82. PODESTA, M. M., BELLELLI, C., FERN,/~NDEZ, P., CARBALLIDO, M. & PANIQUELLI, M. 2000. Arte rupestre de la Comarca Andina del Paralelo 42°: un caso de anfilisis regional para el manejo de recursos culturales. In: PODESTA, M. M. & DE HOYOS, M. (eds) Arte en Ias rocas. Arte rupestre, menhires y piedras de colores en Argentina. Sociedad Argentina de Antropologfa & Asociaci6n Amigos del Instituto Nacional de Antropologfa, Buenos Aires, 175-201. RAMOS, V. & KAY, S. 1992. Southern Patagonia basalts and deformation: back arc testimony of ridge collision. Tectonophysics, 205, 261-282. STERN, CH. & CURRY, P. 1995. Obsidiana del sitio Pose Las Conchillas, Isla Traigtien (45 ° 30'S), Archipi61ago de los Chonos, Chile. Anales del lnstituto de la Patagonia, Serie Ciencias Humanas, 23, 119-124. STERN, CH., GOMEZ OTERO, J. & BELARDI, J. B. 2000. Caracterfsticas qufmicas, fuentes potenciales y distribuci6n de diferentes tipos de obsidianas en el Norte de la Provincia del Chubut, Patagonia, Argentina. Anales del Instituto de la Patagonia, Serie Ciencias Humanas, 28, 275-290.
Prehistoric polished stone artefacts in Italy: a petrographic and archaeological assessment CLAUDIO D'AMICO 1 & ELISABETTA STARNINI 2
1Dipartimento di Scienze della Terra e Geologico-Ambientali, Universitgl di Bologna, Piazza San Donato, 1, 1-40126, Bologna, Italy (e-mail:
[email protected]) 2Soprintendenza per i Beni Archeologici della Liguria, Via Balbi 10, 1-16126, Genova, Italy (e-mail: estarnini@ hotmail.com) Abstract: The paper illustrates the results of an archaeometric project on the raw material
characterization of some collections of prehistoric polished stone tools, dated from the Early Neolithic to the Bronze Age, from sites located in Northern Italy. The petrographic analyses (surface and thin-section microscopy, X-ray powder diffraction, scanning electron microscopy-energy-dispersive spectrometry, X-ray fluorescence, atomic absorption spectrometry) revealed a raw material circulation network involving the whole of Northern Italy. Here occur the outcrops of high-pressure (HP) meta-ophiolites, which were widely utilized from the Early Neolithic onwards for the manufacture of polished cutting-edged tools, which are represented by axes, adzes and chisels. Other raw materials, such as serpentinites, seem to have been preferred for the production of other types of artefacts, including stone rings used as bracelets. The analyses revealed that the prehistoric polished stone artefacts were made from uncommon lithologies such as Alpine eclogites, jades and other HP meta-ophiolites. These rocks were exploited from primary and secondary sources, mainly located in Piedmont, the Aosta Valley and Liguria. During the Neolithic these lithologies are the dominant raw material for the polished stone tools in Northern Italy and southeastern France. In the same period, in other European countries the same lithologies occur less frequently as axe or adze blades; in NW Europe they were frequently used for manufacturing long ceremonial axes, which have a typology that does not appear to belong to the Italian tradition.
Despite the fact that the Italian peninsula is rich in stone resources that can be used for polished implements, archaeometric analyses so far conducted on some of the most important prehistoric stone tool assemblages have demonstrated that high pressure (HP) meta-ophiolites were generally preferred. However, comparisons of archaeological data show that the greenstone circulation might be correlated with the exchange network of other raw materials, such as flint and, to some extent, obsidian (Williams Thorpe et al. 1979; Ammermann & Polglase 1997; Binder 1998; Barfield 2000; Blet et al. 2000). The most important flint outcrops generally exploited since the beginning of the Neolithic are located in the Gargano Promontory (Di Lernia & Galiberti 1993), in the JurassicCretaceous formations of the Lessini Hills, the Venetian Alps and the Marchigian Apennines (Cremaschi 1981; Barfield 1987, 1999; Benedetti et al. 1994, 1994-1995; Peresani 1994; Starnini 1997; Ferrari & Mazzieri 1998). Other common lithologies exploited in prehistory are
represented by other varieties of cherts, defined as radiolarites or jaspers (Del Soldato 1990), the prehistoric quarrying of which has been discovered in the Ligurian and Emilian Apennines (Maggi et al. 1994; Negrino 1998; Ghiretti et al. 2002; Campana & Maggi 2003; Negrino & Starnini 2003). Less important lithic raw material sources are known only in areas where micro-regional surveys and detailed studies have been carried out (Negrino 1999; Del Lucchese et al. 2003). Finally, the existence of four of the most important insular obsidian sources of the Mediterranean should be mentioned, namely those of Sardinia, Pantelleria, and the Pontine and Lipari Islands. Nevertheless, as mentioned above, the petroarchaeometric analyses so far conducted demonstrate that, especially in Northern Italy, only the HP meta-ophiolites were widely used and traded for the manufacture of polished stone, cutting-edged implements, especially axe and adze blades, since the beginning of the Early Neolithic (D'Amico 1998a, 2002a,b; D'Amico
From: MAGGETTI,M. & MESSIGA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 257-272. 0305-8719/06/$15.00 © The Geological Society of London 2006.
258
C. D'AMICO & E. STARNINI
et al. 2002). In fact, it has been demonstrated by
experimental studies that these rocks, because of their mechanical resistance and hardness, are the best raw material for the production of woodworking implements. Their outcrops are located in the western Alpine Arc (Fig. 1) (Franchi 1900; D'Amico et al. 1987; Dal Piaz et al. 1993; Morten 1993; D'Amico 1998b). Secondary Oligocene deposits, containing greenstone cobbles and pebbles, are distributed over a wider area, which covers a great part of Piedmont and southwestern Lombardy. Secondary deposits containing pebbles of these rocks can be found in the alluvial cones along the Ligurian Alpine-Apennine fringes (Fig. 1).
The archaeological and archaeometric framework The time-span covered by this paper extends from the Early Neolithic, when the first polished tools were manufactured, to the Copper-Early Bronze Ages, during which the last polished
stone axes were produced before being replaced by metal tools (i.e. from the seventh to the beginning of the fourth millennium uncalibrated
BP). The state of research is uneven. Petroarchaeometric analyses and systematic studies of the assemblages have been carried out mainly in Northern Italy (Fig. 1), whereas very little information is available for the rest of the peninsula. Petrographic information was obtained by mean of different analytical approaches; these include simple stereomicroscopic observations and qualitative density determinations (100% of the samples), X-ray powder diffraction (XRPD) analyses (95%), thin-section (more than 50%), and scanning electron microscopy-energy-dispersive spectrometry (SEM-EDS) analyses (less than 5%) (D'Amico et al. 1991, 1995, 1997, 1999; Compagnoni et al. 1995; Chiari et al. 1996). Samples for thin-section, XRPD, chemical and S E M - E D S analyses were obtained from broken artefacts, using small flakelets or microcores, under the
Fig. 1. Map of Northern Italy with the location of the archaeological complexes and areas investigated during the archaeometric project. White numbers indicate more recent data. Black hatched areas indicate the HP meta-ophiolite primary outcrops; cross-hatched area represents the main distribution of secondary occurrences of these rocks among the Oligocene conglomerates.
PREHISTORIC
POLISHED
STONE ARTEFACTS
supervision of archaeologists; a number of X-ray diffraction (XRD) determinations were performed directly on the surface of the artefacts (Chiari et al. 1996). Recently, some researchers began the study of polished stone tools from the Neolithic sites of Sardinia (Bertorino et al. 2002), and Leighton published the finds from Southern Italy, in particular Sicily (Leighton 1989, 1992; Leighton & Dixon 1992a,b). Furthermore, a new small assemblage from a Neolithic site in Calabria is currently under study and slowly we hope to achieve a complete picture of the entire Italian territory. At present, the occurrence of HP metaophiolite polished tools in the various geographical areas seems to reflect the 'down-the-line' model of distribution, proposed by Renfrew (1975), slowly falling off as we move away from their primary sources. Their occurrence is documented in Tuscany and more occasionally in Latium, Campania, Calabria and Sicily. The few data so far available are summarized in Table 1. The data from Sardinia (Bertorino et al. 2002) indicate that, during the Neolithic, the imports of raw materials, among which are nephrites and glaucophane schist, were probably from the continent. However, at the present state of research, a provenance from Corsica for nephrite and glaucophane schists cannot be excluded (Bertorino et al. 2002). Local raw materials, such as phonolite, are also represented as artefacts, indicating the exploitation of outcrops probably located in the Pliocene volcanic complex of Montiferru, in central Sardinia. The problem of the provenance of nephrite artefacts and the presence of HP meta-ophiolites in museum collections of Southern Italy will be discussed with more detail below. Regarding the typology of the polished stone tools, we have to point out the absence of a codified type-list (Barfield 1996). This is also
IN ITALY
259
due to the intrinsic difficulty in assigning to a precise type instruments that often simply represent different stages of use, resharpening, recycling, and even exhausted and discarded tools. The most significant classes that occur in Italy are listed in Table 2. The assemblages of studied polished stone tools are rather numerous in Northern Italy. More than 20 sites of different age were analysed (Table 3), numbering more than 1000 implements. The location of these sites is shown in Figure l and the results concerning the sites will be discussed below. Archaeometric analyses were also conducted on a large sample (182 specimens) of artefacts collected from a workshop for the production of polished stone tools located near Rivanazzano (Pavia, NW Italy: Mannoni & Starnini 1994, 1996; Mannoni et al. 1996; D'Amico et al. 2004a). In our opinion this locality deserves a more detailed description because, at present, it is the only prehistoric polished stone tool workshop so far analysed in Northern Italy. The site lies on a terrace of the Staffora River, along the fringes of the northern piedmont of the Ligurian Apennines (Fig. 1, site 18). Here, more than 400 artefacts have been collected on the surface. They consist of hammer-stones, by-products, fragmentary wastes, flakes and rough-outs that result from the manufacture of adzes, axes and chisels. They demonstrate the existence of one important workshop for the production of polished, cutting-edged tools, which can be attributed to the Neolithic period. Petrographic investigations were carried out with the aim of understanding the pattern of exploitation of the stone resources and their possible provenance (D'Amico & Starnini 2001; D'Amico et al. 2003). The sample for the petrographic analyses was selected according to two methods. The first consists of a random sampling of 90 artefacts, the second of 92 samples collected according to the macroscopic
T a b l e 1. Distribution of polished stone tools made from HP meta-ophiolites and other lithologies in
Central-Southern Italy
Lithology
Tuscany
HP meta-ophiolites Nephrites
7 -
Serpentinites Metamorphic rocks Volcanic or Plutonic rocks Sedimentary rocks
5 5
Umbria
59 -
1 3(V) 3
Latium
South Italian sites in museum collections
4 1
18 2
7 28 (V) 14
5 26 130 (V + P) 81
Calabria
-
8 6 (P) 16
Siciiy
Sardinia
1
10 24
4 68 (V) 4
7 12 (V + P) 3
P, plutonic; V, volcanic lithologies. Data from D'Amico et al. (2004b), with addition of unpublished data from Latium and Calabria.
260
C. D'AMICO & E. STARNINI
Table 2. Typology of the tools from the Italian prehistoric sites
Tool type
Chronology and cultural context
Polished stone ring bracelets Small chisels with two opposite cutting edges Shoe-last adzes Cutting-edged tools such as axes and adzes Perforated hammers or axes Large ceremonial axes Hammerstones, often recycled cutting-edged tools
rock differences. The cumulative analysis shows a predominance of eclogites, followed, in order of importance, by glaucophane schists, jades and other HP meta-ophiolites. The occurrence of jades is surprisingly low, in comparison with the usual pattern observed for other Northern Italian sites (D'Amico & Starnini 2000, 2001), whereas that of glaucophane schists is, in contrast, rather high. This can be perhaps explained by the different character of this site, which is a workshop of primary production, where the finished tools are absent, in contrast to settlements,
Early Neolithic Middle Neolithic-Square Mouthed Pottery Culture Early-Middle Neolithic of NW, probably imports From Early Neolithic to Copper Age From Copper to Early Bronze Age Context often undefined; generic Neolithic period Neolithic
from which only finished and used tools have usually been analysed. All the lithotypes of this site can be considered of local provenance, collected as pebbles from the alluvial deposits, naturally enriched in these rocks, which result from the erosion of the Oligocene conglomerates (Fig. 1). It is thus possible to interpret this site as a workshop for the production of such tools, mainly because of the absence of other categories of finds, which are typical of the settlement areas. On the basis of the typology of the artefacts collected, the
Table 3. Polished stone tool assemblages from Northern Italy studied and published during the petro-archaeometric project (see also Fig. 1) Site Alba (CN-Piedmont) Brignano Frascata (AL-Piedmont) Arene Candide (SV-Liguria) Rivanazzano (PV-Lombardy) Vh6 (CR-Lombardy) Ostiano, Dugali Alti (CR-Lombardy) Ostiano, Casotte (CR-Lombardy) Volongo, La Pista (CR-Lombardy) Isorella (BS-Lombardy) Bogliaco (BS-Lombardy) Lonato, Fornasetta (BS-Lombardy) Lonato, Pr~ (BS-Lombardy) Lonato, Case Vecchie (BS-Lombardy) Monte Netto (BS-Lombardy) Remedello (BS-Lombardy) Cascina Ferramonda (BS-Lombardy) Malegno (BS-Lombardy) C~ dei Grii (BS-Lombardy) Casatico di Marcaria (MN-Lombardy) Bancole (MN-Lombardy) Porto Mantovano (MN-Lombardy) San Lazzaro di Savena (BO-Emilia) Gaione (PR-Emilia) Ponte Ghiara (PR-Emilia) Adige Valley, Trentino Valer (PN-Friuli) Sammardenchia (UD-Friuli)
Site type
Chronology
No. of artefacts
Open-air settlement Open-air settlement Cave site Open-air workshop Open-air settlement Open-air settlement Open-air settlement Surface find Open-air settlement Surface find Open-air settlement Open-air settlement Open-air settlement Open-air settlement Graveyard Open-air settlement Surface find Cave site Open-air settlement Open-air settlement Open-air settlement Open-air settlement Open-air settlement Open-air settlement Open-air settlement Open-air settlement Open-air settlement
Early and Middle Neolithic Early Neolithic Early and Middle Neolithic Neolithic Early Neolithic Early Neolithic Middle Neolithic Neolithic Early Neolithic Neolithic Neolithic Copper Age Early Bronze Age Neolithic Copper Age Early Neolithic Copper Age Middle Neolithic Middle Neolithic Middle Neolithic Middle Neolithic Copper Age Neolithic Middle Neolithic Neolithic Early Neolithic Early and Middle Neolithic
115 34 19 182 15 10 31 1 2 1 1 1 1 1 I 2 1 5 4 7 16 100 261 39 80 1 291
AL, Alessandria; BO, Bologna; BS, Brescia: CN, Cuneo; CR, Cremona; MN, Mantova; PN, Pordenone; PR, Parma; PV, Pavia; SV, Savona; UD, Udine.
PREHISTORIC POLISHED STONE ARTEFACTS IN ITALY workshop can be attributed to the Neolithic period. Most of the finds come from settlements, although from the Middle Neolithic onwards the practice of sometimes placing polished stone tools as grave goods in burials began. This ritual is known for the Square Mouthed Pottery Culture, in Northern Italy. It indicates not only the economic value of these objects, but also their symbolic significance. The use of polished stone axes or adzes as grave goods continued until the Copper Age, as documented from the finds of the Remedello cemetery (Fig. 1, no. 8). A general diachronic trend was observed, comparing the various assemblages: the number of polished stone tools is usually low during the Early Neolithic; it increases dramatically during the Middle Neolithic. This discussion has shown that at present we do not have a complete and clear picture for the Late Neolithic, as we lack a sufficient number of studied assemblages. Nevertheless, we are able to deduce that a dramatic change in the raw material exchange system took place at the onset of the Copper Age: the exploitation and procurement of raw materials from long distances dropped in favour of the use of local stone resources, even though they are not of optimal quality. This pattern has been observed not only for the polished stone tool production, but also for the chipped stone industry (D'Amico et al. 1998, 2001; Negrino & Starnini 2003). This phenomenon, however, is particularly evident in NE Italy, where eastern serpentinites replaced western HP meta-ophiolites for the production of the characteristic shaft-hole hammer axes of the Copper Age (D'Amico et at. 1996b; Montagnari Kokelj 2001). A similar pattern also occurs in the Appennine area near Bologna, where local basalts became very common as a raw material during the Copper Age (D'Amico et al. 2000b).
HP meta-ophiolites as a dominant raw material for cutting-edged tool production in Northern Italy The HP meta-ophiolites, employed for the manufacture of axes and other tools, consist of high-pressure-low-temperature (HP-LT) metabasites (eclogites with minor omphacite schists, N a - P x metabasalts, glaucophane schists, greenschists that have undergone retrograde metamorphism) and their differentiates (jades or Na-pyroxenites, including jadeitites,
261
omphacitites, Px-mixed jades), as well as ultramafic rocks (serpentinites), typical of the Alpine geology. As mentioned above, in Northern Italy, fine-grained H P - L T alpine eclogites, jades (Na-pyroxenites) and a few geologically related HP meta-ophiolitic lithologies (omphacite schists, glaucophanic rocks, retrograde-metamorphosed greenschists, serpentinites and other unusual lithologies), represent the main raw material exploited for the manufacture of Neolithic axe blades (usually 90% or more of all lithologies, at least 70% at each site). They are known as 'greenstones' or 'pietre verdi' in the archaeological terminology. More than 1000 axes, adzes and chisels, some ornaments and other less common greenstone tools, and many fragments, sampled from both important sites and museum collections, have been analysed. Optical surface observation, density, thin-section, XRPD, microprobe analyses and bulk chemistry have been employed, alone or in combination, for their petrographic identification, aimed at an archaeometric interpretation. A review of the results so far obtained has been published by D'Amico et al. (2004b), together with a discussion and reinterpretation based on the extensive petro-archaeometric literature. That study is a preliminary report of the data currently available for the HP metaophiolites of Europe. From a typological point of view, most of the Italian polished stone tools consist of axe or adze blades, used for cutting or working wood (Table 4): they are often damaged (fragmented) and/or worn-out. Unworn, complete, ritual or ceremonial axes are much more rare. These tools were mainly employed for woodworking and forest clearance from the Neolithic onwards all over Europe. Chisels and ring bracelets are less frequent, and other types of implements, among which are burnishers, polishers, rough-outs, reused instruments and pebbles, occur in varying percentages. Table 5 gives an updated, summary picture of the number and percentage of blade axe or adze blades, chisels, ornaments and other tools of Northern Italy, according to their lithology. Most of them are Neolithic, although a few belong to the Copper and Bronze Ages. The HP meta-ophiolites predominate, with percentages up to 90%. They are mainly represented by alpine eclogites and jades, although less frequent lithologies also occur. The relationship between lithology and typology (Table 4) shows the predominance of the HP meta-ophiolites for the production of axe blades and chisels (eclogites >jades>> other lithologies), whereas different lithologies
262
C. D'AMICO & E. STARNINI
Table 4. Typology v. lithology of the anah'sed Italian polished stone tools from the Neolithic to the Bronze Age Tool types
Axes Adzes Large, ceremonial axes Rough-outs Shafthole hammeraxes Chisels Hammers, burnishers, polishers Pebbles and unidentifiable Shoe-last chisels Small, miniature axes or adzes Ring bracelets
Eclogites Jades Omphacite Glaucophane Other HP Serpen- Chlorite- Paragonite Other (jadeite) schists metatinites schists schists lithologies schists ophiolites XXXX XXXX XX
XXX XXX XXX
X X X
(X) (X)
X
X
X
XX
(X)
(X) (X) (X)
X X X
X
XX X
X X
X X
X (X)
(X)
(X) XX
(X) X
X
X
X
X
X
X
X
X
XXX
XXX
X
(X)
X
X
XX
XX
XXX
X
The number of x s expresses the frequency, from higher to minimal and sporadic (X).
prevail for the o r n a m e n t s and o t h e r tools. T h i s is d u e to the p r e f e r e n t i a l use o f hard, tough, h e a v y m a t e r i a l s (eclogites, jades, o m p h a c i t e schists, etc.) for the m a n u f a c t u r e o f f u n c t i o n a l tools, and softer r o c k s (serpentinites, p a r a g o n i t e schists, etc.) for the p r o d u c t i o n o f rings and other ornaments.
The predominance of eclogites and jades and o t h e r H P m e t a - o p h i o l i t e s e x t e n d s f r o m northw e s t e r n to all o f N o r t h e r n Italy. T h i s p r e d o m i n a n c e has cultural, l i t h o - t e c h n o l o g i c a l r e a s o n s r e l a t e d to the h a r d n e s s , t o u g h n e s s and d e n s i t y o f the r a w material, and aesthetic r e a s o n s (green colour and translucency of a number of
Table 5. Distribution of the different lithologies according the tool t3"pes Lithogical groups
HP meta-ophiolites
All samples (n = 1018)
Axe or adze blades (n = 668)
No.
%
No.
%
No.
%
No.
%
No.
%
847
83.2
602
90. l
27
90.0
13
32.5
205
73.2
including Alpine eclogites Jades (Na-pyroxenites) Omphacite (jadeite) schists Glaucophane rocks Green schists Serpentinites Other lithologies t
415 240 40 26 23 103 171
40.8 23.6 3.9 2.5 2.3 10.1 16.8
including 327 183 28 15 18 31 66
48.9 27.4 4.2 2.3 2.7 4.6 9.9
Chisels (n = 30)
including 14 7 1 4 1
46.7 23.3 3.3 13.3 3.3 -
3
10.0
Ornamental objects (n = 40)
including 2
5.0 -
11 27
27.5 67.5
Other artefacts* (n = 280)
including 74 48 11 7 4 61 75
26.4 16.8 3.9 2.5 1.4 21.8 26.8
"Undeterminableobjects, hammers, pestles, burnishers, polishers, pebbles, reused fragments and flakes. Paragonite schists, chlorite schists, spotted slates, limestones for manufacturingbracelets. Nephrites, cinerites-tuffites, porphyritic volcanites or sub-volcanites, sandstones, siltites, silexites, gabbros, basalts, granites, etc. for axe or adze blades, chisels, pendants, burnishers and polishers, or as pebbles. n, total number.
35.6 29.4 52.6 54.4 35.9 48.7 63.9 36.7 45.4 53.2 60.4
58.3 20.0 36.1
34 19 182 39 261 36 30 44 47
96
24 80 291
Eclogites
115
No. o f a n a l y s e d samples
2.5 1.3
-
5.9 6.0 20.5 5.7 5.6 2.3 6.4
2.6
Omphacite schists
25.0 16.2 22.3
19.8
29.4 36.8 11.5 15.4 21.8 13.9 40.0 22.7 19.1
36.5
Jades
8.3 1.3 0.7
-
23.1 2.6 5.0 2.8 2.3 2.1
3.5
Glaucophane schist
HP meta-ophiolites
4.2 -
-
2.9 0.5 2.6 5.0 4.6 2.1
6.1
Other H P metaophiolites
4.2 32.5 :I: 7.0 §
6.3
20.6 . 5.3 1.1 20.5* 6.9* 5.6* i 3.4 11.4 8.5
7.0
Serpentinites
.
. 2.5 0.7
-
2.1
0.9 .
Nephrites
.
.
.
.
-
-
2.7
1.4
-
--
--
-
3.3
0.9
Paragonite shists
0.4 3.3
-
Chloriteschist
Other lithologies
25.0 27.8
13.5 t
11.8 5.3 3.3 2.6 6.5 8,3 3.3 1 1.4 6,3
7.0
Other/ local
Data sources: Alba, D'Amico et al. (2000a); Brignano Frascata, D'Amico & Starnini (1996); D'Amico et al. (2000c); Rivanazzano, D'Amico et al. (2003); Ponte Ghiara, Bernab6 Brea et al. (2000); Gaione, Bernab6 Brea et al. (1996), S. Lazzaro di Savena, D'Amico et al. (2000b); Vhr, Starnini et al. (2005); Ostiano, D'Amico (1995); Starnini et aI. (2005); Mantua and Brescia Provinces; Starnini et al. (2005); Province of Verona, Lunardi (2003); Fimon D'Amico & Lunardi (unpubl. data); Trentino, Sammardenchia, D'Amico et al. (1997). *Probably total or partial, Apennine provenance. ? Only preliminary examinations. Others are undefined green stones, probably some type of HP meta-ophiolite. ~.Many pebbles from Adige river, local/regional contribution. § At least partial provenance from easternmost Alps.
Alba (CN), Pigorini M u s e u m collection B r i g n a n o Frascata (AL) A r e n e C a n d i d e (SV) R i v a n a z z a n o (PV) Ponte Ghiara (PR) G a i o n e (PR) S. L a z z a r o di S a v e n a (BO) V h 6 (CR) Ostiano (CR) M a n t u a and Brescia P r o v i n c e s sites V e r o n a P r o v i n c e sites (Natural History. M u s e u m collections) F i m o n (VI) Trentino sites S a m m a r d e n c h i a (UD)
Sites/areas
T a b l e 6. D i s t r i b u t i o n o f l i t h o l o g i e s at the N o r t h I t a l i a n sites ( v a l u e s a r e p e r c e n t a g e s )
~Z
O
t~
© ~
('3
~ ©
264
C. D'AMICO & E. STARNINI
lithologies). These both most probably justify the almost complete absence of other lithologies, which are nevertheless common in other European countries (D'Amico et al. 2004b). Table 6 shows the basic data for the lithologies from each North Italian site or region so far analysed. The Alpine eclogites+jades predominate in general, and in each single assemblage, with varying percentages (eclogites from some 30 to >60% and jades from some 12 to 40%). Differences can be noted between a few assemblages in the relative numbers of eclogite and jade samples. For example, the Alba assemblage (D'Amico & Ghedini 1996; D'Amico et al. 2000a; for the location see Fig. 1) shows roughly the same number of jades and eclogites, whereas those of Rivanazzano and, less markedly, Sammardenchia (D'Amico et al. 1992, 1996a; D'Amico & Felice 1994; Pessina & D'Amico 1999; for the location see Fig. 1) and Gaione (Bernab6 Brea et al. 1996) show a higher predominance of eclogites over jades. Other differences concern the presence or absence or different percentages of single minor HP meta-ophiolite and non-meta-ophiolite lithologies. It is possible that they all have a meaning, which at present is unknown.
Nature and provenance of the HP meta-ophiolites employed for making tools As shown in Tables 5 and 6, eclogites and jades are the most representative rocks. It is well known that HP-meta-ophiolites have undergone subduction of the oceanic crust, HP metamorphism and metasomatism, and subsequent tectonic exhumation from depth. Similar H P - L T metaophiolites and related rocks crop out along the Western Italian Alpine watershed (and in part of Valais, Switzerland) as primary geological bodies within the Penninic Tectonic Units of the Western Alps (Compagnoni 1977; Messiga et al. 1993; Mottana 1993). This is a unique case in Europe and one of the few in the world, as far as the association of eclogites and jades is concerned (Compagnoni et al. 1995; D'Amico et al. 2002). Apart from the above-mentioned primary outcrops, the same rocks are also present in the Oligocene conglomerate secondary deposits of the NW Apennines, which consist of an old basin rich in blocks and pebbles of these and other lithologies, as well as boulders, cobbles and pebbles in the Quaternary moraine and alluvial deposits of Piedmont, Liguria and SW Lombardy, which are derived from both the primary outcrops and the Oligocene deposits. In all
these regions, the HP meta-ophiolites employed for the manufacture of the Neolithic stone tools, and in particular axes and chisels, represent at least 90% of the total assemblage. This demonstrates a widespread utilization of these rocks during the Neolithic. An almost identical situation is known from both Northern Italy and Southern France (Compagnoni & Ricq-deBouard 1993). The exceptions are represented by the northeasternmost Italian regions and the Alpine valleys (e.g. Friuli, Trentino), where the percentages drops to 60-70%. Here other rocks were exploited or imported for tool manufacture. West and north of Provence, south of the Apennines, and in the Danubian and Dalmatian regions, the use of HP meta-ophiolitic rocks drops gradually or abruptly (D'Amico et al. 2004b; D'Amico 2005) following a 'down-theline' model (Renfrew 1975; Barfield 1996). The HP meta-ophiolitic rocks were probably gathered mainly from secondary sources, i.e. the fiver valley fluvial and glacial pebblecobble-boulder deposits along the fringes of the Po Plain (Ricq-de-Bouard et al. 1990; Ricq-de-Bouard & Fedele 1993; D'Amico et al. 1995, 2004b; D'Amico & Starnini 2000, 2005). Recently the exploitation of highland zone primary outcrops of the Voltri Group, Monviso and Valais, has been proposed by Errera (2004) and P6trrquin et al. (2005), for the manufacture of beautiful ceremonial jade axes in France and other Western European countries. This idea is reasonable and convincing, as the production of this variety of long, thin axes from cobbles is undoubtedly difficult and improbable, but it remains to be verified. However, this source is considered improbable so far for most of the Italian tool typologies.
Petrographic characterization of the main iithologies Axe or adze blades
Axe or adze blades and chisels are woodworking implements (Fig. 2). They are mainly obtained from HP meta-ophiolites and rarely from other rocks. The lithologies are briefly described below (readers are referred to D'Amico et al. (2004b) for further details), with relative percentages shown in Tables 5 and 6. The chemical characterization of the HP meta-ophiolites is shown in Table 7. Eclogites (Figs 3 and 4) are commonly finegrained, heterogranular and variously sheared to mylonitic texture. Their components are Na-pyroxenes (omphacite _+ jadeite) + garnets (or derived chlorite) + some compatible or
PREHISTORIC POLISHED STONE ARTEFACTS IN ITALY
Fig. 2. Smallaxe blade from the Early Neolithicsite of Vh6 (Lombardy).
secondary minerals: rutile or ilmenite and/or titanite, always present and often relatively abundant; zoisite-epidote and paragonite, mostly within whitish pseudomorphs of plagioclase (or lawsonite: Compagnoni et al. 1995); common apatite, zircon and other accessory minerals; very sporadic quartz or chloritoid; rare secondary glaucophane, actinolite and albite. Fetich (dark green), Mg-rich (bright or medium green) and intermediate eclogites can be chemically distinguished (Table 7). They are typical Alpine eclogites, very different from the coarser-grained H P - H T eclogites from the Variscan basements of Central and Western Europe (or some areas in the Alps), which have rarely been used for the manufacture of stone axe blades in Europe (Hovorka et al. 2001). Omphacite (jadeite) schists are similar to the eclogites, without garnets or chlorite pseudomorphs after garnet. Zoisite-epidote, chlorite, paragonite, and sometimes albite may be important mineral phases. Jades (Na-pyroxenites) are lacking in garnet and usually contain 90% or more of Napyroxenes of varying composition, plus accessory and secondary mineral phases, among which paragonite is relatively common, as well as rutile or ilmenite or titanite; zircon is often abundant as an accessory mineral. Jades have been conventionally classified as jadeitites and Fe-jadeitites (jadeite or Fe-jadeite 90% or more), omphacitites (omphacite 90% or more) and Px-mixed jades (both jadeite and omphacite in various percentages). Both Fe-rich and Mgrich compositions, dark and light respectively, although of various shades, are chemically recognized (Table 7). Blastic ('saccharoidal') and variously sheared to mylonitic textures are
265
present (Figs 5 and 6). Blastic textures are more common in jadeitites than in other jades. Glaucophane schists (or felses) contain glaucophane (or crossite) as a dominant or very important phase, in various combinations with zoisite-epidote, chlorite, garnet, lawsonite, albite, ore minerals and often residual Napyroxenes. The greenschists, from which a few tools are made, are all eclogites that have undergone retrograde metamorphism; they are very fine grained and composed of albite, chlorite, epidote, actinolite, ore minerals, and relict omphacite and garnets. Serpentinites are usually characterized by the association antigorite + magnetite, + chlorites + residues of Ca-pyroxenes, amphiboles, and rarely olivine. The raw material provenance of the serpentinite polished tools may be from not only the NW Italian sources, but also the Central and Eastern Alps (D'Amico et al. 2004b), and, possibly, the Apennines (lizardite instead of antigorite). The definition of the different provenances is rather ambiguous and therefore attributions are often uncertain. Many other lithologies, which occur in small quantities, differ from the HP meta-ophiolites. They include nephrites, actinolite-hornblende schists, cinerites, andesite-dacites, porphyries, basalts, gabbros, granites, limestones, sandstones, silexites, vitric tufts, cherts and spotted slates. Many of them have been used to produce non-cutting-edged tools, such as burnishers, polishers and pestles, or occur as pebbles or unidentifiable tool fragments. Only a few have been used for axe blades or axe or adze rough-outs, or ornaments or ring-bracelet rough-outs. Among these latter lithologies only the following two merit mention (D'Amico et al. 2004b). Actinolite-hornblende schists, with Amph >>Ab, Ep, ore minerals, etc., in the shape of shoelast adzes of transalpine typology ('Hinkelstein' type) occur sporadically: in the Adige Valley (one specimen from the Middle Neolithic site of La Vela near Trento, dated to c. 5500 uncal, m,) and Friuli (from Sammardenchia, near Udine, probably imports from the Dalmatian coast: P. Biagi, pers. comm.) and testify, by their morphology and petrography, to occasional exchange with transalpine cultures (Barfield 1970). Nephrite (nearly 100% tremolite) has been employed in the production of only a few axes. It is characterized by a fine-grained to felty or schistose texture, occasionally diablastic. It is clearly different in mineralogy and texture from the nephrites described by Kalkowsky
266
C. D'AMICO & E. STARNINI
Table 7. Mean chemical composition (values are percentages) of selected HP meta-ophiolites Rock: n:
Fe-eclogites 32
Mg-eclogites 21
SiO2 TiO2 A1203
48.88 ± 4.29 2.45 ± 0.92 12.19 _ 1.55 15.05 ± 2.91 0.29 ± 0.11 4.68 ± 2.00 7.06 ± 2.29 7.41 ± 2.22 0.07 ± 0.08 0.82 ± 0.69 1.32 ± 1.10 14 ± 16 21 ± 20 356±469
49.81 _ 2.08 1.28 +_ 0.33 13.77 ± 1.90 8.70 _ 1.31 0.18 ± 0.08 10.20 ± 2.06 8.59 ± 1.10 5.65 ± 1.00 0.09 ± 0.08 0.14 ± 0.15 1.78 ± 0.85 237 + 108 113 ± 48 1 4 6 ± 121
Fe203tot MnO MgO CaO Na20 K20 P2Os LOI Cr Ni Zr Rock: n:
Jadeitites 16
Fe-jadeitites 15
Mixed jades 12
SiO2 TiO2 A1203
58.37___1.04 0.61 __+0.34 19.34 ± 1.05 2.61 ± 0.87 0.06 ± 0.02 2.74 ___ 1.09 2.79 ± 0.86 11.90 ___ 1.48 0.24 ± 0.50 0.22 ± 0.26 1.19 ± 0.67 15 ± 13 41 ± 32 735 ± 467
55.99 ___ 1.12 1.07 ± 0.26 15.69 ± 1.37 9.13 ± 1.88 0.16 ± 0.03 2.16 _ 1.16 3.19 ± 0.94 11.45 ± 0.70 0.07 ± 0.07 0.32 ± 0.21 0.83 ± 0.57 18 ± 14 37 ± 48 t045 _ 779
56.76 0.69 15.19 5.03 0.11 5.35 5.88 9.79 0.08 0.29 0.89 60 69 326
Fe203tot MnO MgO CaO Na20 K20 P205 LOI Cr Ni Zr
Intermediate eclogites 6 48.76 2.37 12.07 12.94 0.28 8.18 9.23 5.42 0.05 0.08 1.16 115 53 291
4- 1.37 + 1.09 + 1.41 ± 0.98 + 0.09 ± 0.44 ± 1.09 ± 0.76 ± 0.03 ± 0.10 +_ 1.02 ± 108 ± 23 ± 149
Fe-mixedjades 7
+ 1.12 + 0.39 + 1.63 ± 1.76 ± 0.03 ± 1.79 _ t.71 ± 1.23 ± 0.09 ± 0.85 ___ 0.62 _ 74 ± 64 ± 390
54.82 1.50 13.88 9.38 019 3.42 5.58 9.54 0.05 0.77 1.03 13 25 902
Omphacite schists 9
+ 2.06 + 0.66 ± 1.51 + 2.30 ± 0.04 ± 0.78 __+ 1.91 _ 0.74 ± 0.03 _ 0.69 _ 0.53 ± 8 ± 18 _ 548
48.38 1.24 15.27 8.23 0.15 10.08 8.37 5.27 0.12 0.12 2.82 266 143 115
Omphacitites 5 54.63 0.93 10.10 7.84 0.13 8.29 9.96 6.89 0.13 0.09 1.21 203 147 546
+ + ± ± + + ± ± ± _ ± ± + ±
0.53 0.58 1.83 0.96 0.03 1.62 1.40 0.95 0.14 0.18 0.60 267 174 703
± 1.44 ± 0.40 4- 0.50 ± 1.12 ± 0.02 + 2.53 ± 0.94 ± 1.08 ± 0.12 ± 0.08 ___0.87 ± 58 + 46 ±51
Fe-omphacitites 2 54.86 1.18 14.30 13.09 0.15 1.96 5.92 6.47 0.07 0.85 1.17 17 15 1706
X-ray fluorescencedeterminationsfor all chemicalvaluesexceptNa20 and MgO (atomicabsorption spectrometrydeterminations).Mean values (%), ppm, standard deviation (___)and numberof analysed samples. Samples selected from Tables 2 and 3 (see D'Amico et al. (2004b) for furtherdetails), n, total numberof samples; LOI, loss on ignition. (1906) for Liguria and cannot c o m e f r o m these outcrops. A p r o v e n a n c e f r o m Switzerland (Grisons) can be proposed. Here both nephrite outcrops and their local use for the production
o f Neolithic tools have been well d o c u m e n t e d ( M u t s c h l e c h n e r 1948; Giess 1994). Nephrite axes are m e n t i o n e d here because they are mistakenly considered, by some
Fig. 3. Petrographic thin section of very fine eclogite (sample from Sammardenchia, cross-polarized light (NX), 51 x).
Fig. 4. Petrographic thin section of sheared eclogite (sample from Sammardenchia, NX, 51 ×).
PREHISTORIC POLISHED STONE ARTEFACTS IN ITALY
267
Fig. 5. Petrographic thin section of jade with blastic texture (sample from Sammardenchia, NX, 51 × ).
Fig. 7. Ring bracelet from the Early Neolithic site of Vhb (Lombardy).
Fig. 6. Petrographic thin section of jade with mylonitic texture (sample from Sammardenchia, NX, 51 x ).
workers, to have the same origin as the eclogite and jade axes (Bertorino et al. 2002; P~tr~quin et al. 2002, 2005; Errera 2004). This suggestion does not seem to have any basis. From a geological point of view, nephrites are not high-pressure rocks as the jades and eclogites are. Therefore their provenance must be from different formations. Archaeologically speaking, nephrite axes or adzes are very few or absent in Northern Italy (Table 6).
potentially derived from many areas (Table 7) (D'Amico et al. 1995, 2004b; D'Amico & Starnini 2000), and were exploited for making ornaments or working implements other than cutting-edged tools (Table 5). Felty and pure chlorite-schists and felses are present in a reasonable quantity exclusively at Sammardenchia, in Friuli (D'Amico et al. 1997). Occasionally. unusual lithologies, among which are spotted slates and limestones, may occur. Hard-stone ring bracelets, such as jades, are very rarely represented (Alba in Piedmont). 'Others' are represented by shapeless finds, flakes from manufacture and maintenance of implements, hammers, pestles (Fig. 8),
O r n a m e n t s a n d o t h e r tools
Ornaments are mainly represented by ring bracelets (Fig. 7) and a few pendants, made from soft rocks, which may be classified as follows. Paragonite schists (about 100% paragonite) occur at many sites (Venturino Gambari 1996; D'Amico et al. 1997, 2000a) and most probably are from Piedmont HP metamorphic deposits, probably the 'Sesia-Lanzo' zone (Traversone 1996). Serpentinites are present in all the collections, usually in small quantities (Table 6). They are
Fig. 8. Recycled cutting edged-tools reused as pestles (from the Neolithic sites of Vhb and Ostiano, Lombardy).
268
C. D'AMICO & E. STARNINI
burnishers, polishers, pebbles and reused fragments, of all lithologies, as reported in Table 6. Rather frequent are HP meta-ophiolite, reused, worn-out axe blades; splintered and shapeless fragments of the same and/or different material; and occasional, locally collected lithologies, as mentioned by D'Amico et al. (2004b).
Discussion and conclusions To conclude, the data described here show a complicated and articulated picture. It is interesting to mention briefly the situation outside Italy. Thicker and shorter axe blades, obtained from the same lithologies, similar to those of the Italian tradition, occur in other areas of Europe together with ceremonial axes. These latter are often polished all over the surface and are not worn. This indicates that they were not used for working wood. The noticeable occurrence of jade and alpine eclogites, especially amongst the Western, and more rarely, Central European axe blades (mainly in form of symbolic or ceremonial axes), demonstrates the distribution of the HP meta-ophiolites from NW Italy to France, Germany, Benelux, Britain, etc. (Campbell Smith 1965; Wolley et al. 1979; Compagnoni & Ricq-de-Bouard 1993; D'Amico 1993, 2000, 2005; Ricq-deBouard 1996; Ricq-de-Bouard et al. 1996; D'Amico et al. 2004b), up to a distance of 1000-1500 km from their source. The presentday unsystematic and still incomplete petrographic knowledge of the raw materials, including, possibly, HP meta-ophiolitic rocks, of the polished stone tools in Europe, including the Italian peninsula, allows only a preliminary comparison of these areas with the situation described here for Northern Italy. The currently available picture of the widespread Neolithic greenstones diffusion and circulation, all over Europe, will need further research and more petrographic identifications, if we are to achieve a correct interpretation of the manufacture and variability of use of the same lithological material, from a generally functional use in Northern Italy and its closely related regions, to its very important exploitation for the production of 'symbols of social prestige' in N W Europe.
References AMMERMANN,A. & POLGLASE,C. 1997. Analyses and descriptions of the obsidian collections from Arene Candide. In: MAGGL R., STARNINI, E. & VOYTEK, B. (eds) Arene Candide: a functional and environmental assessment of the Holocene sequence
excavated
by
L.
Bernabb
Brea
(1940-50).
Memorie deil'Istituto Italiano di Paleontologia Umana, Nuova Serie, 5, 573-592. BARHELD, L.-H. 1970. La stazione neolitica de 'La Vela" presso Trento. Considerazioni sulle tombe a cista nel Trentino Alto Adige. Studi Trentini di Scienze Naturali, sezione B, 47(1), 35-55. BARF[ELD, L.-H. 1987. Recent work on sources of Italian flint, h~: SIEVEKING, G. DE G. & NEWCOMER, M.-H. (eds) The Human Uses of Flint and Chert. Proceedings of the 4th International Flint Symposium, Cambridge, 231-239.
BARFIELD, L.-H. 1996. Le asce in pietra levigata nel Neolitico d'Europa e d'Italia. In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 57-66.
BARFIELD. L.-H. 1999. Neolithic and Copper Age flint exploitation in Northern Italy. Universitdtsforschungen zur priihistorishen Archiiologie, 55,
245-252. BARFIELD, L.-H. 2000. Commercio e scambio nel Neolitico dell'Italia settentrionale. In: PESSINA, A. & Musclo. G. (eds) La Neoliti~azione tra Oriente e Occidente, Atti del Convegno, Udine, 2 3 - 2 4 aprile 1999, Museo Friulano di Storia Naturale, Udine, 55-66. BENEDETTI, R., CAMPANA, R., D'AMtco, C. & NANNETTI, M. C. 1994. Petroarchaeometry of Epigravettian and Mesolithic flints in the Val Cismbn-Lagorai area (NE Italy). The flint supply question. Preistoria Alpina, 28(1), 33-49. BENEDETTI, M. R., D'AMICO, C. & NANNETTI, M. C. 1994-1995. Studio petroarcheometrico preliminare di selci neolitiche alpino-padane a scopo di confronto tra siti e ricerca di provenienza. Ani della Societh per la Preistoria e Protostoria della regione Friuli- Venezia Giulia, 9, 171 - 179. BERNABO BREA, M., D'AMICO, C., GHEDINI, M., GmRETTI, A. & OCCHL S. 1996. Gaione, loc. Case Catena. h~: VENTURINO GAMBAR1, M. (ed.) Le vie della pietra verde. L 'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 122-136. BERNABO BREA, M., D'AMICO, C., GHED1NI, M. & MAZZ1ERI, P. 2000. La pietra verde di Ponte Ghiara, Fidenza (PR). Mineralogica et Petrographica Acta, 43, 233-243. BERTORINO, G., FRANCESCHELLI, M., LUGLII~, C., MARCHI, M. & COLUMBU, S. 2002. Petrographic characterization of polished stone axes from Neolithic Sardinia. Archaeological implications. Periodico di Mineralogia, 71(special issue), 87-100. BINDER, D. 1998. Silex blond et complexit6 des assemblages lithiques dans le Nrolithique liguroprovenqal. Rencontres mdridionales de Prghistoire r~cente, Deuxibme session, Arles, 1996, APDCA, Antibes, 11 i - 128. BLET, M., BINDER, D. & GRATUZE, B. 2000. Essais de caractrrisation des silex Brdouliens provenqaux par analyse chimique 616mentaire. Revue d'Archgomgtrie, 24, 149-167.
PREHISTORIC POLISHED STONE ARTEFACTS IN ITALY CAMPANA, N. & MAGGI, R. (eds) 2003. Archeologia in Valle Lagorara. Diecimila anni di storia intorno a una cava di diaspro. Origines, Studi e Materiali pubblicati a cura dell'Istituto Italiano di Preistoria e Protostoria, Firenze. CAMPBELL SMITH, W. 1965. The distribution of jade axes in Europe with a supplement of the catalogue of those from the British Isles. Proceedings of the Prehistoric Society, New Series, 31, 25-33. CHIARI, G., COMPAGNONI, R., GIUSTETTO, R. & RICQ-DE-BOUARD, M. 1996. Metodi archeometrici per lo studio dei manufatti in pietra levigata. In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 35-53. COMPAGNONI, R. 1977. The Sesia-Lanzo Zone: high pressure-low temperature metamorphism in the Austro-alpine continental margin. Rendiconti della Societd Italiana di Mineralogia e Petrologia, 33(1), 335-374. COMPAGNONI, R. & RICQ-de-BOUARD, M. 1993. French neolithic stone axes in 'jade': petrological variability and origin. Terra Nova Abstracts, 5(Supplement 4), 5-6. COMPAGNONI,R., R1CQ-de-BOUARD,M., GIUSTETTO,R. & COLOMBO, F. 1995. Eclogite and Na-pyroxenite stone axes of southwestern Europe: a preliminary petrologic survey. Bollettino del Museo Regionale di Scienze Naturali, 13(Supplement 2), 329-359. CREMASCHI, M. 1981. The source of the flint artefacts for the Central Po Plain and Apennine sites, between the 7th and the 2nd millennium he. Staringia, 6, 139-142. DAL PIAZ, G. V., Gosso, G., PENNACCHIONI, G. & SPALLA, M. I. 1993. Geology of eclogites and related rocks in the Alps. In: MORTEN, L. (ed.) Italian Eclogites and Related Rocks. Accademia Nazionale delle Scienze, Scritti e documenti, 13, 17-58. D'AMIcO, C. 1993. Eclogites and jades as prehistoric implements in Europe. Terra Nova Abstracts, 5(Supplement 4), 7-8. D'AMIcO, C. 1995. Petroarcheometria dei reperti in pietra levigata. In: BIAGI, P. (ed.) L'insediamento neolitico di Ostiano-Dugali Alti (Cremona) nel suo contesto ambientale ed economico. Monografie di Natura Bresciana, 22, 92-96. D'AMIco, C. 1998a. La circolazione delle materie prime nel Neolitico: le pietre verdi. In: PESSINA A. & MuscIo, G. (eds) Setternila anni fa. I1 primo pane, Comune di Udine and Museo Friulano di Storia Naturale, 177-183. D'AMIcO, C. 1998b. Introduzione alia provenienza geologica dei materiali culturali. In: Atti della 4 a Giornata Le Scienze della Terra e l'Archeometria, February 1998. CUEN, Napoli, 133-137. D'AMIcO, C. 2000. La pietra levigata neolitica in Italia settentrionale e in Europa. Litologia, produzione e circolazione. In: PESSINA, A. & MuscIo, G. (eds) La Neolitizzazione tra Oriente e Occidente, Atti del Convegno, Udine, 23-24 aprile 1999. Museo Friulano di Storia Naturale, Udine, 67-80.
269
D'AM~cO, C. 2002a. La pietra levigata tra Neolitico e Bronzo Hell'area alpino-padana. Atti della XXXIII Riunione Scientifica dell'Istituto ltaliano di Preistoria e Protostoria (Trento, 1997), 417-431. D'AMtCO, C. 2002b. Polished stone between the Neolithic and Bronze Ages in Northern Italy. Preistoria Alpina, 34, 297-305. D'AMICO, C. 2005. Neolithic 'greenstone' axe blades from Northwestern Italy across Europe: a first petrographic comparison. Archaeomet~, 47(2), 235-252. D'AMICO, C. & FELICE, G. 1994. Petroarcheometria di reperti neolitici di Sammardenchia (Udine). Notizie preliminari. Atti della XXIX Riunione Scientifica dell'Istituto Italiano di Preistoria e Protostoria. Firenze, 159-165. D'AMICO, C. & GHEDINI, M. 1996. La pietra levigata della collezione 'Traverso' di Alba, nel Museo etnografico 'L. Pigorini' di Roma. Museologia Scientifica, 13(Supplement 1-2), 293-312. D'AMICO, C. & STARNINI, E. 1996. Brignano Frascata (Scavi Tin6 1984). In: VENTURtNO GAMBARI, M. (ed.) Le vie della pietra verde, L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 74-79. D'AMIco, C. & STARNINI, E. 2000. Eclogites, jades and other HP metaophiolites of the Neolithic polished stone tools from northern Italy. Krystalinikum, 26, 9-20. D'AMICO, C. & STARNINI,E. 2001. The polished stone axe workshop of Rivanazzano PV-Northern Italy): analyses of the lithological diversities in comparison with other Italian Neolithic sites. Slovak Geological Magazine, 7(4), 420. D'AMIco, C. & STARNINI, E. 2006. Approvvigionamento e circolazione di manufatti neolifici in eclogite, giada e altre litologie connesse in Italia settentrionale: lo stato della ricerca. In: Archeologie, Studi in Onore di Tiziano Mannoni, Edipuglia (in press). D'AM~cO, C., INNOCENTI, F. & SASSI, F. P. 1987. Magmatismo e metamorfismo. (Appendice 6). UTET, Torino. D'AMICO, C., BARGOSSI, G. M., FELICE, G. & MAZZEO, M. 1991. Giade ed eclogiti in pietra levigata. Studio petroarcheometrico. Mineralogica et Petrographica Acta, 34, 254-283. D'AMICO, C., FEL1CE, G. & GHEDINI, M., 1992. Lithic supplies in the Early Neolithic to Sammardenchia (Friuli), Northern Italy. Science and Technology for Cultural Heritage, l, 159-176. D'AMICO, C., CAMPANA,R., FELICE, G. • GHEDINI,M. 1995. Eclogites and Jades as prehistoric implements in Europe. A case of petrology applied to cultural heritage. European Journal of Mineralogy, 7, 29-41. D'AMICO, C., FERRARI, A., GHEDINI, M. & PESSINA, A. 1996a. Sammardenchia (Pozzuolo del Friuli). In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionate, Omega, Torino, 84-91. D'AMtcO, C., GHEDINI, M., MICHELI, R. & MONTAGNAR1 KOKELJ, E. 1996b. Le asce rotate del Friuli-Venezia Giulia. In: VENTURINO
270
C. D'AMICO & E. STARNINI
GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell 'Italia senentrionale. Omega, Torino, 229-238. D'AMIcO, C., FELICE, G., GASPAROTTO,G., GHEDINI, M., NANNETTI, M. C. 8£ TRENTINI, P. 1997. La pietra levigata neolitica di Sammardenchia (Friuli). Catalogo petrografico. Mineralogica et Petrographica Acta, 40, 385-426. D'AMIco, C., FELICE, G. & GHEDINI, M. 1998. Neolithic-to-Bronze polished stone in Northern Italy. Proceedings of the XIII Congress UISSP, Forli, 1996, Volume 3, Section 9, 389-399. D'AMIcO, C., NANNETTI, M. C., TRENTINI, P., De SALVO, F., FELICE, G. & GHEDIN1, M. 1999. Informazioni petrologiche dall'archeometria. Eclogiti e giade del Neolitico Italiano. Atti 6" Giornata 'Le Scienze della Terra e l'Archeometria ', Este, 151 - 158. D'AMICO, C., GHEDIN1, M., NANNETTI, M. C. & TRENTINI, P. 2000a. La pietra levigata neolitica di Alba (CN). Catalogo petrografico e interpretazione archeometrica. Mineralogica et Petrographica Acta, 43, 176-206. D'AMICO, C., LENZI,F., MARGUTTI,S. & NENZIONI,G. 2000b. Temoignage de la pierre polie du chalcolithique a l'est de Bologna: analyses quantitatives, qualitatives et typologiques. 2nd International Congress 'Science and Technology for the Safeguard of Cultural Heritage in the Mediterranean BasbF. 5 - 9 J u l y 1999, CNRS-CNR, Paris, 271- 278. D'AMICO, C., STARNINI, E. & VOYTEK, B. 2000c. L'industria litica di Brignano Frascata (AL): dati paleoeconomici di un insediamento del Neolitico antico padano attraverso l'analisi tipologica funzionale e lo studio della provenienza delle materie prime. Preistoria Alpina, 31, 91-124. D'AMICO, C., GASPAROTTO, G., GHEDINI, M. & SABETTA, T. 2001. Serpentiniti e metaultramafiti ad anfiboli e cloriti in asce-martello eneolitiche del NE Italiano. Geohalia 3 ~ Forum, sessione 21, Chieti, 5 - 8 settembre 2001, Riassunti, 631-632. D'AMIcO, C., BERNABO BREA, M., BIAGI, P., PEDROTTI, A. L., PESSINA, A. & STARNINI, E. 2002. Archaeometrical analyses of polished stone tools from the Neolithic to the Bronze Age in Northern Italy. In: JEREM, E. & BIR0, T. (eds) Proceedings of the 31st International Symposium of Archaeometry 98, Budapest, 1998. British Archaeological Report International Series, 104301), 691- 696. D'AMIco, C., MINALE, M., STARNINI,E. ~ TRENTINI, P. 2003. L'officina di produzione di asce in pietra levigata di Rivanazzano (PV). Dati archeometrici e catena operativa. Atti della XXXV Riunione Scientifica IIPP, Le comunitb della Preistoria italiana. Studi e ricerche sul Neolitico e le Eth dei metalli, Firenze, 981-986. D'AMICO, C., SIMONE ZOPFI, L. & STARNINI, E. 2004a. Rivanazzano, loc. La Cascinetta. hz: VENTURINO GAMBARI, M. (ed.)Alia conquista dell'Appennino. Le prime comunith delle valli Curone, Grue e Ossola. Omega Edizioni, Torino, 203 -205.
D'AMICO, C., STARNINI, E., GASPAROTTO, G. & GHEDINI, M. 2004b. Eclogites, jades and other HP-metaophiolites employed for prehistoric polished stone implements in Italy and Europe. Periodico di Mineralogia, 73(Special Issue 3), 17-42. DEL LUCCHESE, A., MARTINI, S., NEGRINO, F. & OrTOMANO, C. 2003. 'I Ciotti' (Mortola Superiore, Ventimiglia, Imperia). Una localit~ di approvvigionamento della materia prima per la scheggiatura durante il Paleolitico. Bullettino di Paletnologia ltaliana, 91-92, 1-26. DEL SOLOATO, M. 1990. Analisi petrografiche deli'industria litica scheggiata, bz: MAGGI, R. (ed.) Archeologia dell'Appennino Ligure. Collez. Di Monografie preistoriche ed Archeologiche, VIII, 209-218. D! LERNIA, S. & GAL1BERTI, A. 1993. Archeologia mineraria della selce nella Preistoria. Definizioni, potenzialit~ e prospettive di ricerca. Quaderni del Dipartimento di Archeologia e Storia delle Arti, sez. Archeologica, Universith di Siena, 36, 1-83. ERRERA, M. 2004. D6couverte du premier gisement du jade-jad6ite dans les Alpes (6t6 2004). Implications concernant plusieurs lames de hache n6olitique trouv6es en Belgique et dans les r6gions limitrophes. Notae Prehistoricae, 24, 191-202. FERRARI, S. • MAZZIERI, P. 1998. Fonti e processi di scambio di rocce silicee scheggiabili. In: PESS1NA, A. & MuscIo, G. (eds) Settemila annifa, llprimo pane, Comune di Udine and Museo Friulano di Storia Naturale, 165-169. FRANCHI, S. 1900. Sopra alcuni giacimenti di rocce giadeitiche helle Alpi occidentali e Appennino ligure. Bollettino della Reale Commissione Geologica Italiana, 31, 119-158. GHIRETTI, A., NEGRINO, F. & TOZZI, C. 2002. Estrazione del diaspro e produzione di strumenti a ritocco bifacciale in localit/t Ronco del Gatto (Bardi, Parma): modificazioni economiche e tecnologiche tra la fine del Neolitico e l'eth del Rame nell' Appennino ligure-emiliano. In: FERRARI, A. & VISENTINI, P. (eds) 1l declino del mondo neolitico. Ricerche in Italia centrosettentrionale fra aspetti peninsulari, occidentali e nord alpini. Atti del Convegno, Pordenone, 403 -408. GIESS, H. 1994. Jade in Switzerland. The Bulletin of Friends of Jade, 8, 1-46. HOVORKA, D., SPISIAK, J., MERES, S., llla~ova, L. & Dc Bl KOVA, K. 200 I. Petrological and geochemical evidences of the eclogite hammer-axe from Nitriansky Hradok site (Neolithic, Lengyel culture, Slovakia). Slovak Geological Magazine, 7, 355-363. KALKOWSKY, E. 1906. Geologie de Nephrites im siidlichen Ligurien. Zeitschrift der Deutshen. Geologischen Gesellschaft, 58, 307-378. LEIGHTON, R. 1989. Ground stone tools from Serra Orlando (Morgantina) and stone axes studies in Sicily and Southern Italy. Proceedings of the Prehistotqc Society', 55, 135 - 159. LEIGHTON, R. 1992. Stone axes and exchange in south Italian prehistory: new evidence from old collections. Accordia Research Papers, 3, 7-36.
PREHISTORIC POLISHED STONE ARTEFACTS IN ITALY LEIGHTON, R. & DIXON, J. E. 1992a. Jade and greenstone in the prehistory of Sicily and Southern Italy. Oxford Journal of Archaeology, 11(2), 179-200. LEIGHTON, R. & DIXON, J. E. 1992b. Alcune considerazioni sulle asce levigate in Italia Meridionale ed in Sicilia. In: Papers of the Fourth Conference of Italian Archaeology, Volume 3, New Developments, Part 1, London, 9-28. LUNARDI, A. 2003. Le lame d' ascia in pietra verde del territorio veronese dal Neolitico all'et~ del Bronzo: petrografia, tipologia e funzione. Atti della Societb per la Preistoria e Protostostoria della regione Friuli- Venezia Giulia, 13, 57-110. MAGGI, R., CAMPANA, N., NEGRINO, F. & OTTOMANO, C. 1994. The quarrying and workshop site of Valle Lagorara (Liguria, Italy). The Accordia Research Papers, 5, 73-96. MANNONI, T. & STARNINI,E. 1994. I1 contributo delle analisi petrografiche nello studio dell'officina litica di Rivanazzano (PV). In: D'AMICO, C. & CAMPANA, R. (eds) Le scienze della Terra e l'Archeometria. Giornata di studio, Bologna, 21 Aprile 1994, 21. MANNONI, T. & STARNINI, E. 1996. Rivanazzano-Caratteristiche tecniche della lavorazione--La petrografia. In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 119-120. MANNONI, T., STARNINI,E. & SIMONE ZOPFI, L. 1996. Rivanazzano. In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde, L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 119-122. MESSIGA, B., SCAMBELLURI,M. & TR1BUZIO, R. 1993. Petrology of eclogitized mafic ophiolites from Western Alps. In: MORTEN, L. (ed.) Italian Eclogites and Related Rocks. Accademia Nazionale delle Scienze, Roma, Scritti e Documenti, XIII, 79-96. MONTAGNARI KOKELJ, E., 2001. Pietra verde, Neolitico e Post-Neolitico, Carso e Friuli (Italia Nord-orientale): lo stato della questione. A t t i e Memorie della Commissione Grotte 'E. Boegan', 38, 71-86. MORTEN, L. (ed.) 1993. Italian Eclogites and Related Rocks. Accademia Nazionale delle Scienze, Roma, Scritti e Documenti, XIII. MOTTANA, A. 1993. Italian eclogite-facies minerals: phase relationships, chemical compositions and crystal structures. In: MORTEN, L. (ed.) Italian Eclogites and Related Rocks. Accademia Nazionale delle Scienze, Roma, Scritti e documenti, XIII, 121 - 141. MUTSCHLECHNER, G. 1948. Vorkommen der Steinzeitlichen Werkstoffe Nephrit und Jadeit in unsere Alpen. Der Schlern, 107-109. NEGRINO, F. 1998. The quarrying and workshop site of Valle Lagorara (Maissana, La Spezia). In: Lithic technology: from raw material procurement to tool production. Workshop 12, Workshops-Volume II, Proceedings of XIII International Congress of Prehistoric Sciences (Forli, 1996), Number 6, 831-836.
271
NEGRINO, F. 1999. Palaeolithic evidences of quarrying activity at 'I Ciotti' (Mortola Superiore, Ventimiglia, Imperia, Italy). In: Abstracts, VIII International Flint Symposium, Bochum, 55-56. NEGRINO, F. & STARN1NI, E. 2003. Patterns of lithic raw material exploitation in Liguria from the Palaeolithic to the Copper Age. In: Actes de la Table ronde internationale: Les Matibres Premikres Lithiques en Prdhistoire. Inventaire, caractdrisation et circulation des matikres premikres lithiques durant la pr~histoire (paldolithique, m~solithique et n~olithique). Aspects rdgionaux et inter-r~gionaux. Aurillac (France), 20-22 Juin 2002. Prehistoire du Sud-ouest, 2003, Supplrment, 5, 235-243. PERESANI, M. 1994. Flint exploitation at epigravettian and mesolithic sites on the Asiago Plateau (Veneto Prealps). In: BIAGI, P. (ed.) Highland Zone Exploitation in Southern Europe. Monografie di Natura Bresciana, 20, 221-243. PESSINA, A. & D'AMICO, C. 1999. L'industria in pietra levigata del sito neolitico di Sammardenchia (Pozzuolo del Friuli, Udine). Aspetti archeologici e petroarcheometrici. In: FERRARI, A. & PESSINA, A. (eds) Sammardenchia--Cueis. Museo Friulano di Storia Naturale pubblicazione, 41, 23-92. PI~TRI~QUIN,P., CASSEN, S., CROUTSCH, C. & ERRERA, M. 2002. La valorisation sociale des longues haches dans l'Europe Nrolithique. In: GIULAINE, J. (ed.) Matdriaux, production, circulations du Ngolithique ~ l'Age du Bronze. Errance, Paris, 67 -98. PI~TRI~QUIN,P., PI~TRI~QU1N,A. M., ERRERA, M., et al. 2006. Voltri, Viso et Valais. A l'origine des grandes haches plies alpines au V ° millrnaire en Europe occidentale. In: Materie prime e scambi nella Preistoria Italiana. Atti della XXXIX Riunione Scientifica dell'Istituto Italiano di Preistoria e Protostoria, Firenze, 25-27 november 2004 (in press). RENFREW, C. 1975. Trade as action at a distance. In: SABLOFF, J. & LAMBERG-KARLOVSKY, C. C. (eds) Ancient Civilization and Trade. Omega Edizioni, Albuquerque, NM, 1-59. RICQ-DE-BOUARD, M. 1996. Petrographie et socidtds neolithiques en France mdditerran~enne. L'outillage en pierre polie. Centre de Recherches Archrologiques, Monographie, 16. RICQ-DE-BOUARD, M. & FEDELE,G. F. 1993. Neolithic rock resources across the Western Alps: circulation data and models. Geoarcheology, 8(1), 1-22. RICQ-DE-BOUARD, M., COMPAGNONI,R., DESMONS,J. & FEDELE, F. 1990. Les roches alpines dans l'outillage poli nrolithique de la France Mrditerranrenne. Classification, origine, circulation. Gallia Prdhistoire, 32, 125-149. RICQ-de-BOUARD, M., COMPAGNON1, R., COLOMBO, F. & DEISS, W. 1996. Le materiel des Musres de Provence et du Dauphin~. Contribue 7t la connaissance des courants de circulation transalpins, au Neolithique. Museologia Scientifica, 13(Supplement 1- 2), 313- 328. STARNINI, E. 1997. Raw material procurement and use strategies. In: BARONI, C. & BIAGI, P. (eds)
272
C. D'AMICO & E. STARNINI
Excavations at the high altitude mesolithic site of Laghetti del Crestoso (Bovegno, Brescia--Northern Italy). Ateneo di Brescia, Brescia, 21-24. STARNINi, E., D'AMICO, C., BIAGI, P., GHEDIN1,M. & PITTI, G. 2005. Strumenti in pietra levigata dalla Lombardia orientale: aspetti archeometrici e culturali. Bullettino di Paletnologia Italiana, 95, 1-62. TRAVERSONE, B. 1996. Oggetti ornamentali. In: VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino, 197-202.
VENTURINO GAMBARI, M. (ed.) 1996. Le vie della pietra verde. L'industria litica levigata nella preistoria dell'Italia settentrionale. Omega, Torino. WtLLIAMS THORPE, O., WARREN, S. E. & BARFIELD, L.-H. 1979. The sources and distribution of archaeological obsidian in Northern Italy. Preistoria Alpina, 15, 73-92. WOLLEY, A. R., BISHOP, A. C., HARRISON, R. J. KINNES, I. A. 1979. European Neolithic jade implements: a preliminary mineralogical and typological study. Stone Axes Studies, 23, 90-96.
The stone of the inscribed Etruscan stelae from the Valdelsa area (Siena, Italy) A. G A N D I N l, E. C A P E Z Z U O L I 1 & A. C I A C C I 2
lDipartimento di Scienze della Terra, Universitg~ di Siena, Via Laterina 8, 53100 Siena, Italy (e-mail: gandin @unisi, it) 2Dipartimento di Archeologia e Storia delle Arti, Universitgt di Siena, Via Roma 56, 53100 Siena, Italy Abstract: The results of an archaeometric study on five inscribed Etruscan funerary stelae,
produced during the second half of the sixth century ac, in the Valdelsa area (NW of Siena, Italy), are reported. The style of the inscriptions suggests the presence of a local culture and a common language. The stone used to make the stelae is a laminated limestone consisting of the alternation of more or less compact bands whose petrographic features can be compared to those of travertines (continental carbonates deposited from hot springs). The fabric of the stone of these stelae implies provenance from a thermal deposit, in the Valdelsa area, where calcareous tufa dominates; however, hot water deposits are probably to be found in the Gracciano Val d'Elsa area, not far from the localities where the stelae were found.
Five inscribed funerary stelae of Etruscan age have been found in the Valdelsa valley, NW of Siena in southern Tuscany. They were made from a concretionary limestone referred to as 'travertine', a term that, in Italian, encompasses a wide variety of continental carbonates deposited by lacustrine, palustrine or fluvial waters, or by hot or cold spring waters (i.e. a very diverse range of physico-chemical, biological and climatic conditions). Historically, 'travertine' has been used in central Italy for the production of stone objects and as a building stone because of its good stratification, availability on the surface and ease of quarrying. Recent studies have shown that these concretionary carbonates (Pentecost 1995; Ford & Pedley 1996) can be classified on the basis of their depositional environment. Thus, laminated accretionary deposits related to hydrothermal waters are called travertine. Travertines are relatively unaffected by climatic factors. They are characterized mainly by macrocrystalline facies, consisting of large ray-crystals, associated with coated bubble travertine, paperthin rafts and shrub travertine (Guo & Riding 1998). Travertine exhibits regular plane parallel, centimetre-scale lamination, high growth rates and low organic content, and is slightly enriched in 13C (813C from - 2 to 10%0 and total dissolved inorganic carbon (TDIC; 1 5 - 6 0 mM 1-1), magnesium, strontium and sulphur (Pentecost 1995; Ford & Pedley 1996).
The term calcareous tufa has been coined in the recent English language literature (Viles & Goudie 1990; Pedley 1990) to indicate nonthermal continental carbonates (i.e. those derived from fluvial or spring waters in balance with subaerial surface temperatures). They form as encrustations of microcrystalline calcite on supports consisting mainly of higher plants (bryophytes, reeds, etc.), usually at a low deposition rate. They normally have irregular or massive stratification, are affected by climatic conditions, and have low contents of TDIC ( 1- 7 -(10) mM 1-1), strontium and magnesium, with strong depletion in 13C (~13C from - 2 to - 12%0 (Pentecost 1995; Ford & Pedley 1996). Although the distinctive features of the various types of carbonates are still under discussion, some workers believe that a reliable classification can be made on the basis of the petrographical and geochemical characters (Pedley 1990; Pentecost 1995; Ford & Pedley 1996; Capezzuoli & Gandin 2004). There are extensive outcrops of travertine and calcareous tufa in southern Tuscany. This paper reports a study of the lithological and genetic characteristics of the stone from which the five stelae were cut, aimed at establishing the stone's genesis and possible provenance. The latter is important information for the archaeology of the production of that period, as it permits the reconstruction of historical human-environment relationships and of the
From: MAGGETTI,M. & MESSIGA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 273-282. 0305-8719/06/$15.00 © The Geological Society of London 2006.
274
A. GAND[N ET AL.
commercial links between diverse areas in Etruria (Mangani 1992). However, the identification of the origin of stone used in antiquity is often a complex archaeometric problem, particularly for continental limestones, because of the great variability in the vertical and lateral distribution of the limestone facies that have been utilized. Here we discuss the possible location of the quarry on the basis of our study of the microfacies of the stone and of the characteristics of travertine and calcareous tufa present in the area surrounding the Valdelsa valley (Fig. 1).
The Etruscan stelae The studied stelae are housed in various museums in their discovery area, as follows. Stele 184531: Museum of Santa Maria della Scala (SMS), Siena, deriving from Campassini (Monteriggioni); 120cm high, 60cm wide, about 15-22 cm thick, 412 kg weight (Becatti 1933) (Fig. 2a). Stele 13918: Museum of SMS, Siena, deriving from Toiano (Sovicille); 131 cm high, 70 cm wide, about 15 cm thick, 327 kg weight (Pauli & Danielsson 1893) (Fig. 2b).
A stele in the Museum of San Gimignano, recently discovered near Ulignano (San Gimignano); about 70cm high, 55 cm wide, about 15 cm thick (Fig. 2c). Stele 139435: Museum of Colle Val d'Elsa, deriving from Morticce di Mensanello (Colle Val d'Elsa); 93.5 cm high, 51 cm wide, 12 cm thick (Fig. 2d). A stele in the Museum of Casole d'Elsa, deriving from Le Poggiola (Casole d'Elsa); 69cm high, 101 cm wide, 17.5-20cm thick (Chigi 1877) (Fig. 2e). These Etruscan stelae are an interesting example of funerary artefacts used in northern Etruria in the second half of the sixth century Be, particularly in the areas of Volterra, Rusellae, Vetulonia and in the Valdarno (Fig. 1). In most cases, the stele is in the form of an arch (thus the name 'horseshoe stele') and follows a model common in the Padanian Etruria region at the end of the fifth century BC and particularly typical of the Volterra area, where it first appeared during the second half of the sixth century BC (Bruni 1995). In such stelae, the deceased is celebrated mainly by representations of warriors or characters holding the lituum (sceptre), but banquets (with a few people), dances and games are also
,./_
LIVORNO
Elba IStar~d /% ~
(}
0
/
I1(
[ ] Ulignanc ~, ~. [] Morticce di' : Mensanello~ [ ] Le Poggiola 0
[ ] Campassini
10 20
30Kin
[ ] Toiano ,~
Outcrop of travertine
~
Outcropof
calcareous tufa area of Figure 6
~ "'"''
....
~'
'~
""
"
' '
~
I
I
Fig. 1. Distribution of the Valdelsa stelae and location of the surrounding concretionary carbonate deposits.
STONE OF INSCRIBED E T R U S C A N STELAE
275
a - Campassini b - Toiano c- Ulignano d - Morticce Mensanello e - Le Poggiola
Fig. 2. The five inscribed Etruscan stelae studied.
di
276
A. GANDIN ET AL.
illustrated. Unlike other funerary styles (e.g. the so-called 'Chiusi cippi', characterized by more complex scenes and produced by an urbanized society; see Cristofani 1978), stelae addressed a mainly rural world and emphasized with a few essential elements (including inscriptions) the prestige and power of Etruscan princes (Zifferero 1991). The inscribed Etruscan stelae of the Valdelsa valley are part of this theme of death-celebration, although their specific function is not always clear (e.g. funerary symbol, commemorative monument or tomb door). Moreover, it has been suggested (Zifferero 1991) that stelae were used to mark and possibly organize the funerary space on the land owned by noble Etruscan families. The localities where the Valdelsa stelae were found (Fig. 1) are in the valleys of the Elsa River and Rosia Creek (a tributary of the Merse River), which are separated by the Montagnola Senese metamorphic ridge: the former valley includes the sites of Ulignano (San Gimignano), Le Poggiola, Morticce di Mensanello (Casole d'Elsa) and Campassini (Monteriggioni), whereas the latter includes Toiano (Sovicille). The Valdelsa stelae lack figurative representations and the deceased is celebrated by inscriptions (Fig. 2). The incised cornices, following the curved profile of the stele, also have a 'decorative' function in addition to their informative role. This tradition is represented by an even older example in the Siena area, the so-called 'Tomb of the Alphabet' (Bartoloni 1997) discovered near Monteriggioni. Although the tomb has been lost, archival documents testify to a wall 'decoration' consisting exclusively of inscriptions and an alphabet. Nevertheless, the primary function of the use of writing appears to be related to the social affirmation and rank of the deceased, as well as to awareness of the distinctive meaning it assumed: knowledge of writing was something to boast about. Therefore, the owners of the inscribed stelae belonged to a wealthy class, well aware of themselves and of the degree of social affirmation that writing assured (Bruni 1997, 2002). This class of stele used their inscriptions to document the inheritance of their land and the holding of power, including control of the transit routes to and from the Valdelsa valley (Ciacci 1999). Funerary writing appears to have been used in the Valdelsa area since the second half of the seventh century BE, but was most widespread around the end of the following century, with the production of specific objects such as stelae, 'cippi' and urns. The production of stelae consisted of: cutting the stone along the
horizontal stratification of the deposit; extraction, probably using wooden wedges; cutting the 'horseshoe'; smoothing the surface to receive the inscription; plus plastering and decoration (Ciacci 2004). The last two operations were revealed by analyses of superficial films, which showed the existence of a lime-finish pigmented with carbon-black overlying a grey covering plaster probably used for decoration (Guasparri et al. 2004). The operations necessary to cut and finish the stelae required unspecialized labourers, probably part of the servitus of a socially and culturally emancipated clientele. Writing was the duty of scribes, connected to the emergent families rather than to specific workshops, who were also able to create new styles of graphic writing. In fact, some recently discovered inscriptions on a stele (Martelli 1975) and a cinerary urn (Cianferoni 2002) found in the area of Colle Val d'Elsa, belonging to the Etruscan family Shekuntena (in Latin, Secundii), highlight a particular treatment of the initial sibilant of the noble name, which used a digram created ex-novo (sh) to indicate a phoneme not as yet recorded in Etruscan (Maggiani 2003; Ciacci 2004). Material and methods Material
All the studied stelae consist of a 10-15 cm thick slab of limestone formed by more or less regular alternations of centimetre-thick bands composed of flat or slightly wavy compact (Fig. 3a) and vacuolar facies, the latter characterized by the presence of calcium carbonate tubules (Fig. 3b). All the laminae tend to be lenticular and thus of variable thickness, although the more compact bands generally predominate in number or thickness. The slabs were extracted from the mother rock along detachment surfaces parallel to the stratification or lamination. Their state of preservation is generally good, although some of them show a slight alteration of pedogenetic origin, apparently attributed to late diagenetic exposure (see 'Interpretation'), or wear signs probably caused by use different from the original function (Guasparri et al. 2004). Methods o f sampling and analyses
Five cylindrical samples (2 cm diameter and 2 cm deep), one per slab, were taken with a microcorer from an unexposed area of the stelae (to preserve their aesthetic quality) to characterize the microfabric.
STONE OF INSCRIBED ETRUSCAN STELAE
277
Fig. 3. (a) Alternationof fine porous and slightly vacuolarbands. Stele 139435. (b) Regular successionof relatively compact bands. Stele from Le Poggiola. To make the sampling representative and comparable among stelae, each sample was taken from the compact facies. This allowed us to obtain a more complete sample from which was possible to acquire more information to characterize the stone. Two thin sections, parallel and orthogonal to the stratification plane of the slab, were cut from each cylinder for a detailed petrographic analysis.
Results
Fabric of the stone of the stelae The bands with a fine porous structure and compact appearance as well as those with more or less accentuated vacuolarity show a common spongy, thrombolitic fabric, consisting of a thin network of dense microcrystalline calcite (of micrite and microsparite size) containing relicts of organic matter (filaments and coccoids), and sometimes peloids with a clotted inner structure (peloidal microfacies; Pedley 1994). Internally, these peloids show a lumpy (grumous) fabric interpreted as the result of low-Mg micrite precipitation on coccoid bacterial agglomerations (Chafetz & Folk 1984). This network, referred to as microbial bindstone (bacterial mats; Folk et al. 1985), outlines small irregular pores (primary porosity) lined by cement of clear rhomboidal-scalenohedral (Fig. 4a) and/or acicular crystalline calcite (Fig. 4b), partially (Fig. 4c) or completely occluded by equant spar. Occasionally, small tubules composed of dense micrite (Fig. 4d), either unfilled or filled with equant sparry calcite, are trapped in the microbial bindstone. They represent small clusters or fragments of encrusted macrophyte stems and/or roots, usually well represented in the porous facies.
Some samples have rosettes of zoned crystals of dark calcite (Fig. 4e) caused by the presence of carbonaceous material (bacterial oncoids of Chafetz & Meredith 1983; Chafetz & Guidry 2003). They appear to have formed after cementation occurred, at the expense of the microbial fabric, probably following the circulation of meteoric or pedogenetic waters. Most samples also have relatively welldeveloped secondary porosity, associated with dissolution and/or bioturbation cavities produced by root systems, which are usually empty or partially filled with a yellowish clay (Fig. 4f). The macrofabric characteristics of the vacuolar facies can be related to the PhytoclasticPhytohermal facies (Ferreri 1985) or Reed Travertine (Guo & Riding 1998), formed by encrustation of the stems of marsh plants such as reeds. The tubules are usually massed together and oriented in the plane of the stratum (Phytoclastic facies), but are sometimes perpendicular to it (Fig. 5), arranged in small clusters of stems in a life-like position (Phytohermal facies). The stone of the Le Poggiola stele has the best preserved depositional and diagenetic fabric and the lowest total porosity (about 5%), whereas the stelae from Toiano, Campassini, Ulignano and Morticce di Mensanello have relatively high secondary porosity (up to 22%), indicating the occurrence of dissolution processes related to late-diagenetic or epidiagenetic recirculation of meteoric waters.
Interpretation The depositional and diagenetic evolution of these carbonates can be inferred from the fabric of the dominant spongy microfacies. The first phase, i.e. deposition, corresponds to formation
278
A. GANDIN ET AL.
Fig. 4. (a) Cement of clear rhomboidal-scalenohedral crystalline calcite. Stele 184531. (b) Cement of clear acicular crystalline calcite. Stele from Ulignano. (c) Thin network of dense microcrystalline calcite (spongy texture) with relicts of organic matter, which outlines irregular primary pores partially occluded by crystalline calcite. Stele from Le Poggiola. (d) Stems filled with cement and surrounded by a thin layer of micritic clots. Stele 184531. (e) Rosettes of dark calcite crystals probably related to recirculation of meteoric waters after deposition. Stele 139435. (f) Very fine clayey material deriving from the vegetated zone of the ground (soil) and deposited within small secondary cavities. Stele 13918.
of the microbial bindstone, whose genesis is believed to be related to biomediated precipitation of microcrystalline calcite on a substratum formed by mucilaginous masses of extracellular polymeric substances produced by unicellular
micro-organisms (cyanobacteria and cyanophytes) trapped inside the microcrystalline precipitate (Pedley 1994; Riding 2000). This phase alternated with ephemeral episodes of tentative colonization by reeds, fragments of
STONE OF INSCRIBED ETRUSCAN STELAE
Fig. 5. Remainsof small tufts of marsh reeds encrusted with calcium carbonate. Stele 13918. The pencil is 5 cm long.
which in the form of stems transported by water are mainly found entrapped in the spongy ground-mass. The first generation of cement, consisting of isopachous fringes of rhomboidal or acicular crystals, records the persistence during early diagenesis of active circulation of calcium carbonate-rich water within the porous network of the bindstone. The precipitation of equant spar, i.e. the second generation of cement, indicates meteoric phreatic circulation during later diagenetic stages, and the presence of incompletely occluded pores implies an interruption of water circulation. The presence of dissolution vugs and internal clayey sediments is evidence of active circulation of undersaturated or chemically aggressive waters of probable meteoric or pedogenetic origin. These observations allow a palaeoenvironmental reconstruction of these concretionary deposits. The texture characterized by mats of micro-organisms could have been preserved only in the absence of their natural enemies, i.e. grazers such as gastropods, which feed on their colonies. The concomitant scarcity or absence of macrophytes indicates environmental conditions unfavourable to higher organisms (Riding 2000). These conditions were determined by one or more physico-chemical characteristics of the water, i.e. high temperature, presence of H2SO 4 and/or HzS and low levels or lack of oxygen. Such conditions occur in palustrine systems located either near thermal sources (Folk et al. 1985) or in fluvio-lacustrine systems (Pedley 1994). However, episodes of dilution by rain or mixing with cold, more oxygenated waters could have created favourable, albeit ephemeral, conditions for the development
279
of pioneer plants such as reeds. The presence of peloids, occurring also in deposits formed in pools with stagnant non-thermal water (Pedley 1994), confirms this reconstruction. Moreover, the evidence of pedogenetic processes indicates that these deposits, formed in very shallow pools or marshes, were periodically subject, at least in the proximal areas, to wet and dry conditions. Consequently, the primary microbial deposit was affected by dissolution and/or bioturbation by roots, transport of clayey material toward the bottom and reprecipitation of calcite (recorded by rosettes of dark calcite; Fig. 4e), probably induced by oscillations of the water table. Such water-level oscillations also affected the depth of the marsh and are reflected in the carbonate sediments (Freytet & Plaziat 1982; Alonso-Zarza 2003). The presence of phytohermal structures indicates the marginal, shallower part of the marsh, with more diluted water and less energy (Toiano stele, Fig. 5; Campassini stele, Fig. 4d). The associations of phytoclastic facies indicates an input of plant remains by occasional water transport (Morticce di Mensanello stele, Fig. 3a), whereas the scarcity or absence of macrophyte remains indicates less oxygenated areas of the water basin (Le Poggiola, Figs 3b and 4c). The good preservation of the depositional and early diagenetic fabric can be related to an early interruption of water circulation, which resulted in incomplete occlusion of the residual porosity, and/or the formation of a poorly developed dissolution-vug system with internal clayey sediments. Evenly laminated carbonates characterized by similar depositional and diagenetic fabrics have also been observed in recent carbonates forming in distal pools of active hydrothermal systems at Tivoli (Folk et al. 1985), in the Mammoth Hot Springs (Yellowstone National Park, Chafetz & Guidry 2003) and in distal deposits of a well-preserved Messinian hot spring complex occurring near Volterra (Pignano) in southern Tuscany (Bossio et al. 1992), as well as in the lacustrine-palustrine environments of ambient-temperature water systems (Pedley 1994).
Discussion The stone used to make the five Valdelsa stelae is very different from the two local building stones deriving from the Mesozoic Tuscan Unit (Fig. 6) and available in the Valdelsa area, both very commonly used during the Middle Ages (Rodolico 1953): marble (a hard, not easily quarried rock, used as an ornamental stone in
280
A. GANDIN ET AL. -\
i ".~Sa ~G,mii na-~" ,~
cl
CDF
a
C • , - : 4 " . " : COF. ~ d e , F:,~ Synthem • -.-o., Ld - tn'~l~,'~: Og - co,'~lome~ ..... s - sand, a - day} • ..~.;-;...; • . • . • • . ° . • • . • . • ! CAI_ - Cal~na.a Synthem . . . . . . . . . . . . . FOC Torrente Foo Synthern • ""
•
.
•
* *"
" ""
• °"
*.'~
l ~
(:
~
o l l e Val d'Elsa •
° ""
c/'r'..." . •
[ ~
:oci r a c c l a n o j• ~
~ii~
........
\-
POG PoggibollfaSynt~em
Poggibonsi S y n t h e m (POG) Abbadia Synthem (ABB) Calcinaia Synthem (CAL) T.Foc~ S y n t h e m ( F O C ) Bellavista Synthem (BEL) tl - calcareous lithofacies t2 - detritical l i t h o f a c i e s C a m p i g l i a d e i Foci Synthern (CDF)
~ Pliocenemanne
.--..v
sediments Miocene continental sediments
i : ii i Monteriggioni
~
re-NeogeneLigurian
a n d T u s c a n Units
Le Caldane Spring
~;,
I I 1Km
Fig. 6. Geological scheme of the Valdelsa valley (modified from Capezzuoli & Sandrelli 2004). Inset: stratigraphic relationships of the Quaternary synthems recognized in the Valdelsa Basin (not to scale).
Siena Cathedral (Giamello et al. 2005)) and 'Cavernoso Limestone' (a brecciated rock, relatively soft but very vacuolar, mainly used as ashlar stonework (Rodolico 1953; Giamello et al. 2003))• The stele stone is a well-laminated, well-stratified concretionary limestone, generally rather compact, resistant to atmospheric agents, easily worked, easily extracted along the stratification surfaces and readily found on the surface without the need to open large quarries. Nevertheless, the petrographical data from the stele samples are not sufficient to determine without doubt whether the initial palustrine environment with extreme conditions was part of a hot-spring complex, and thus fed by hot water rich in poisonous gases (HzS, SO2-SO4). It may equally have been part of a fluviopalustrine system fed by cold poorly oxygenated water of karst origin. In fact, the absence of the macro-crystalline facies, exclusively developed in travertines (Guo & Riding 1998) and the scarcity of macrophyte remains in a living position and of faunas typical of calcareous tufas (e.g. ostracodes and gastropods) tend to exclude both a thermal origin s e n s u stricto and a clearly fluvio-palustrine environment• Therefore, in view of current knowledge and the high lateral variability of these deposits (Guo & Riding 1999), it is not possible to precisely define
either the general depositional context or the place of origin of the single slabs used to make the stelae. We can, however, be confident that the palaeoenvironment was marshy, and contained reed travertine. Water was possibly derived from thermal waters partially cooled after outflow from the resurgence and diluted with meteoric and/or karst-derived waters• Nevertheless, an analysis of the distribution of the fossil and currently forming laminated freshwater carbonates in central-southern Tuscany provides clues to the provenance of the material used by the Etrnscans. Travertines, and occasionally hydrothermal springs related to them, are very widespread outside southern Valdelsa, forming the deposits of Rapolano Terme (Guo & Riding 1998) and Acqua Borra (Castelnuovo Berardenga) to the SE, Pignano (Volterra, Bossio et al. 1992; Capezzuoli et al. 2004) to the west, Frosini (Chiusdino) to the SW and Iano (Gambassi Terme) to the north (Fig. 1). In these localities, proximal to the spring, well-laminated crystalline facies, associated with coated bubbles, shrub and paper-thin raft travertine, are dominant whereas distal facies, transitional to calcareous tufas, are less abundant• In contrast, both active and fossil calcareous tufas are predominant in the Valdelsa area. In fact, repeated phases of continental carbonate deposition occurred in
STONE OF INSCRIBED ETRUSCAN STELAE this area during the Quaternary, represented by five orders of terraces (Fig. 6). These are formed by thick micritic carbonates deposited in a lacustrine-palustrine environment (Campiglia dei Foci Synthem) and by generally crudely bedded, poorly cemented calcareous tufas indicative of a fluvio-palustrine environment (Abbadia Synthem, Calcinaia Synthem, T. Foci Synthem, Bellavista Synthem; Capezzuoli & Sandrelli 2004). The origin of the calcareous tufa deposits, present only in part of the valleys of some local watercourses (Elsa River, Staggia Creek and Foci Creek), is attributed to the activity of local karst and thermal springs during the Quaternary (Capezzuoli & Sandrelli 2004). However, as active springs are now rare in the area and the thermal springs are of low volume and low temperature, it can be assumed that most of them ceased to be active when local tectonic activity ceased (Capezzuoli & Sandrelli 2004). In conclusion, a potential extraction site may be sought in the Pignano area (Messinian travertines of the Volterra basin) or in southern Valdelsa, in an area with signs of thermal springs related to the deposition of Quaternary carbonates. With our current knowledge, precise location of the site is unlikely. However, the results of continuing research suggest the zone of Gracciano Val d'Elsa (in the southern sector of the area, see Fig. 6) as a possible source of the stone material used by the Etruscans. In fact, an extensive, poorly exposed tabular outcrop of concretionary limestone has recently been detected adjacent to one of the rare thermal springs still active in the area (Le Caldane Spring). In these well-bedded deposits at present under study, which show many of the lithological features of intermediate travertines or reed travertine, small-scale quarrying of limestone slabs took place until recent times. This activity occurred randomly where the small plateau morphologies allowed access to the rock stratification. For this reason, largescale quarries never existed at Gracciano Val d'Elsa and quarrying activity today is represented only by a series of holes, which are now part-filled with soil and vegetation. The possible extraction site occupies an approximately central position with respect to the localities where the stelae were found (Fig. 1). This is compatible with the view that Etruscan artisans, rather than carry materials from the farther Volterra, may have preferred to use deposits near their workshops (Petrelli et al. 2004), with the possibility of relatively short transportation of the material.
281
This paper has been financially supported by the Siena PAR (University Research Plan) grant (to A.G.). The writers thank G. C. Cianferoni (Superintendent of the Archaeology of Tuscany), M. Manganelli (President of Gruppo Archeologico Colligiano), M. Bezzini (President of Societ~t Archeologica Valdelsana of Casole d'Elsa) for kindly allowing the sampling of all the stelae, and Gruppo Archeologico of San Gimignano and A. Mennucci for facilitating access to the stele of Ulignano. The authors would like to thank P. Christie and M. Pedley for improving the English version of the text. We are grateful also to two anonymous referees for helpful discussion and comments.
References ALONSO-ZARZA, A. M. 2003. Palaeoenvironmental significance of palustrine carbonates and calcretes in the geological record. Earth-Science Reviews, 60(3-4), 261-298. BARTOLON~, G. 1997. La Tomba dell'Alfabeto di Monteriggioni. In: Etrusca et ltalica. Scritti in ricordo di Massimo Pallottino. Universith degli Studi 'La Sapienza'-Consiglio Nazionale delle Ricerche e Istituto per l'Archeologia EtruscoItalica, Istituti Editoriali e Poligrafici Internazionali, Pisa-Roma, 25-49. BECATTI, G. 1933. Monteriggioni (Siena). Scavo di una tomba a camera ed alcuni trovamenti archeologici. Notizie degli Scavi di Antichita, Accademia
Nazionale dei Lincei, Roma, 150-159. BOSSIO, A., CERRI, R., COSTANTINI,A., et al. 1992. I1 bacino di Volterra. In: Guida alle Escursioni postcongresso. 76 ° Riunione estiva SGI: l'Appennino settentrionale. Escursione B4: I bacini distensivi neogenici e quaternari della Toscana, 244-270.
BRUNI, S. 1997. La Valdera e le colline pisane inferiori: appunti per la storia del popolamento. In: Aspetti della cultura di Volterra etrusca fra l'et& del Ferro e l'etb Ellenistica e contributi della ricerca archeologica alla conoscenza del popolo etrusco, Atti del XIX Convegno di Studi Etruschi ed Italici, Volterra, 129-171.
BRUNI, S. 2002. La Valle dell'Arno: i casi di Fiesole e Pisa. In: MANGANELLI,M. ~ PACCHIANI,E. (eds) Cittb e territorio in Etruria. Per una definizione di cittb nell'Etruria settentrionale. Grupp Archaeolo-
gical Colligiane, Colle Val d'Elsa, 272-344. CAPEZZUOLI, E. 8~ GANDIN, A. 2004. I 'travertini' in Italia: proposta di una nuova nomenclatura basata sui caratteri genetici. II Quaternario, Italian Journal of Quaternaty Sciences, 17, 273 -284. CAPEZZUOLI, E. • SANDRELLI, F. 2004. I sedimenti quaternari del settore meridionale della Valdelsa (Provincia di Siena). II Quaternario, Italian Journal of Quaternary Sciences, 17(1), 33-40. CAPEZZUOLI, E., GANDIN, A. & SANDRELLI, F. 2004. Neogene-Quaternary continental carbonates: the Quaternary deposits of Valdelsa and the Miocene travertines of Pignano (Volterra). In: MOR~NI, D. & BRUM, P. (eds) The 'Regione Toscana' project of geological mapping; case histories and data acquisition. Regione Toscana, Firenze, 89-96.
282
A. GANDIN ET AL.
CHAFETZ, H. S. & FOLK, R. L. 1984. Travertines: depositional morphology and the bacterially constructed constituents. Journal of Sedimentary Petrology, 54, 289-316. CHAFETZ, H. S. t~; GUIDRY, S. A. 2003. Deposition and diagenesis of Mammoth Hot Springs travertine, Yellowstone National Park, Wyoming, USA. Canadian Journal of Earth Sciences, 40(1 l), I515-1529. CHAFETZ, H. S. & MEREDITH, J. C. 1983. Recent travertine pisoliths (pisoids) from Southeastern Idaho, USA. In: PERYT, T. (ed.) Coated Grains. Springer, Berlin, 456-455. CHIGI, B. 1877. Siena. Notizie degli Scavi di AntichitL Accademia Nazionale dei Lincei, Roma, 301-304. CIACC1, A. 1999. I1 periodo etrusco. Note sul popolamento e l'economia. In: VALENTI, M. (ed.) Carta archeologica della Provincia di Siena. lll--La Val d'Elsa (Colle Val d'Elsa e Poggibonsi), Nuova Immagine, Siena. 300-312. CIACCI, A. 2004. Le stele etrusche dell'Alta Valdelsa. Gli aspetti archeologici ed epigrafici, hi: CIACCI, A. (ed.) Monteriggioni-Campassini. Un sito etrusco nell'Alta Valdelsa. AIl'Insegna del Giglio, Firenze, 183-210. CIANFERONI, G. C. 2002. L'Alta Valdelsa in et/~ orientalizzante e arcaica, h~: MANGANELLI, M. t~¢ PACCHIANI, E. (eds) Cittgt e territorio in Etruria. Per una definizione di cittb nell 'Etruria Settentrionale. Colle di Val d'Elsa, 83-126. CRISTOFANI, M. 1978. L'arte degli Etruschi. Produzione e consumo. Giulio Einaudi, Torino, 139-147. FERRERI, V. 1985. Criteri di analisi di facies e classificazione dei travertini pleistocenici dell'Italia Meridionale. Rendiconti Accademia Scienze Fisiche e Matematiche, Napoli, 52, 1-47. FOLK, R. L., CHAFETZ, H. S. 8z TIEZZI, P. A. 1985. Bizarre forms of depositional and diagenetic calcite in hot-spring travertines, Central Italy. In: SCHNEIDERMANN,N. 8,1; HARRIS, P. (eds) The Biology of Blue- Green Algae. Blackwell Scientific, Oxford, 349-369. FORD, T. D. & PEDLEY, H. M. 1996. A review of tufa and travertine deposits of the world. Earth-Science Reviews, 41, 117-175. FREYTET, P. & PLAZIAT, J. C. 1982. Continental carbonate sedimentation and pedogenesis--Late Cretaceous and Early Tertiary of southern France. Contributions to Sedimentology, 12, 1-213. GIAMELLO, M., GUASPARRI, G., MUGNAINI, S., SABATINI, G. & SCALA, A. 2003. Lo studio dei materiali lapidei del centro storico di Siena. Arkos, 2, 22-29. GUASPARRI, G., NARDELLI,M. G. & SCALA, A. 2004. Studio delle superfici lapidee di sue stele. In: CIACCI, A. (ed.) Monteriggioni-Campassini; un sito etrusco in Alta Valdelsa. All'Insegna del Giglio, Firenze, 222-227. GIAMELLO, M., DROGHINI, F., MUGNAINI, S., GUASPARRI, G., SABATINI, G., SCALA, A. &
MORANDINI, M. 2005. I1 Pavimento marmoreo del Duomo di Siena. Caratterizzazione dei materiali e dello stato di conservazione. In: CACIORGNA, M., GUERRINI, R. & LORENZONI, M. (eds) Studi interdisciplinari sul Pavimento del Duomo di Siena. lconografia, stile, indagini scientifiche. Opera della Metropolitana di Siena, Collana di studi e ricerche. Cantagalli, Siena, 173-197. Guo, L. & RIDING, R. 1998. Hot-spring travertine facies and sequences, Late Pleistocene, Rapolano Terme, Italy. Sedimentology, 45, 163-180. Guo, L. & RIDING, R. 1999. Rapid facies changes in Holocene fissure ridge hot spring travertines, Rapolano Terme, Italy. Sedimentology, 46, 1145-1158. MAGGIANI, A. 2003. Rivista di Epigrafia Etrusca, Studi Etruschi, Istituto Nazionale di Studi Etruschi e Italici, Firenze, LXIX, 362-364. MANGANI, E. 1992. Castelnuovo Berardenga (Siena). L'Orientalizzante recente in Etruria settentrionale: tomba A della necropoli principesca del Poggione (1980). Notizie degli Scavi di Antichitb., Roma, Accademia Nazionale dei Lincei, 42/43, 5-82. MARTELLI, M. 1975. Rivista di Epigrafia Etrusca, Studi Etruschi, Istituto Nazionale di Studi Etruschi e ltalici, Firenze. XLIII, 200-201. PAOLI, C. & DANIELSSON O.A. 1893. Corpus lnscriptionum Etruscarum, I(4620). PEDLEY, H. M. 1990. Classification and environmental model of cool freshwater tufas. Sedimentary Geology, 68, 143-154. PEDLEV, H. M. 1994. Prokaryote-microphyte biofilms and tufas: a sedimentologicai perspective. Kaupia, 4, 45-60. PENTECOST, A. 1995. The Quaternary travertine deposits of Europe and Asia Minor. Quaternary Science Reviews, 14(10), 1005-1028. PETRELLI, M., PERUGINI, D., MORONI, B. & POLl, G. 2004. Travertine, a building stone extensively employed in Umbria from Etruscan to Renaissance age: provenance determination using artificial intelligence technique. Periodico di Mineralogia, 73(3), 151-169. RIDING, R. 2000. Microbial carbonates: the geological record of calcified bacterial-algal mats and biofilms. Sedimentology, 47, 179-214. RODOLICO, F. 1953. Le pietre delle cittb d'ltalia. Le Monnier, Firenze. VILES, H. A. & Goudie, A. S. 1990. Tufas, travertines and allied carbonate deposits. Progress in Physical Geography, 14, 19-41. ZIFFERERO, A. 1991. Forme di possesso della terra e tumuli orientalizzanti nell'Italia centrale tirrenica. In: HERRING, E., WHITEHOUSE, R. WILK1NS, J. (eds) The Archaeology of Power. Papers of the Fourth Conference of Italian Archaeology, 1. Accordia Research Centre, London, 107-134.
Geological tools to interpret Scottish medieval carved sculpture: combined petrological and magnetic susceptibility analysis S U Z A N N E M I L L E R 1, F I O N A M. M c G I B B O N 2, D A V I D H. C A L D W E L L 1 & N I G E L A. R U C K L E Y 3
1National Museums of Scotland, Chambers Street, Edinburgh EH1 2PB, UK (e-mail: s. mille r @nms. ac. uk) 20ffice of Lifelong Learning, The Universi~ of Edinburgh, 11 Buccleuch Place, Edinburgh EH8 9LW, UK 3The Old School House, Kirkbuddo DD8 2NQ, UK Abstract: Geological surveys of 172 early and late medieval sculptured stones from central
and western Scotland have been undertaken to determine the provenance of the materials used. Non-destructive petrological studies (including grain size, mineralogy distribution, and textural and structural characteristics) and magnetic susceptibility measurements are used to characterize the sculptured stones and potential source material. The results indicate that for the early medieval sculpture: (1) all the sculptured stones are sandstone with the exception of one siltstone and one granite; (2) the sedimentary rocks are consistent with sources in the Lower Old Red Sandstone of the area; (3) from within the Lower Old Red Sandstone, a number of different geological units have been used. For the late medieval sculpture, the results indicate that: (1) various rock types have been used including schist, slate and sandstone; (2) non-locally derived material is used extensively in some areas, suggesting a more developed network for procurement of raw materials. The analytical techniques used also provide additional information in the art historical interpretation of a number of carved stones by identifying carved fragments from the same monument.
Medieval sculpture in East Central Scotland (Angus, Tayside and Perthshire) is exemplified by some 120 eighth to 10th century (Pictish) sculptures and sculptural fragments (e.g. Fig. 1). These were published and catalogued by Cruden (1964). The Royal Commission for Ancient and Historical Monuments Scotland (RCAHMS) is in the process of drawing and photographing the collections. Medieval West Highland sculpture (e.g. Fig. 2) was defined and surveyed in an important monograph by Steer & Bannerman (1977). There are some 600 examples ranging in date from the 14th to the 16th centuries, spread over 86 sites throughout the Western Highlands of Scotland. Typically, the Pictish sculptures are standing stones with elegant designs and symbols, which often include abstract designs and pictures of objects and animals (both real animals and fantasy creatures). The West Highland sculpture is typified by grave monuments and commemorative crosses. Decoration includes effigies (generally military in nature), ships, hunting scenes, scrollwork and inscriptions. For the
purposes of this work, the early medieval sculpture (from East Central Scotland) will be referred to as 'Pictish sculpture' and the late medieval sculpture (from Iona and Oronsay) will be referred to as 'West Highland sculpture'. Many of the sculptures are in the care of Historic Scotland (e.g. the St. Vigeans collection) whereas others are part of the collections of the National Museums of Scotland, are in local authority care, or are held privately. The present project was initiated with the aim of improving previous interpretation of the sculpture by bringing geological knowledge and skills to bear on the subject, leading to a reassessment of where the sculpture was carved and a new understanding of power and patronage in medieval Scotland. The project has involved close collaboration with RCAHMS and Historic Scotland. Using the non-destructive, in situ techniques of macro-petrology and magnetic susceptibility, a survey of the most representative sculptures has been undertaken. Pictish sculpture collections surveyed include: St Vigeans, Aberlemno, Meigle, Pictavia and
From: MAGGETT1,M. & MESSIGA,B. (eds) 2006. Geomaterials in Cultural Heritage. Geological Society, London, Special Publications, 257, 283-305. 0305-8719/06/$15.00 © The Geological Society of London 2006.
284
S. MILLER ETAL.
(a)
Although previous studies have used such techniques to characterize artefacts, they have been largely restricted to objects composed of igneous rocks (e.g. Williams-Thorpe & Thorpe 1993; Peacock 1995, 1997; WilliamsThorpe et al. 1996, 1997; Markham 1997; Williams-Thorpe & Henry 2000) or low-grade metamorphic rocks (Floyd & Trench 1988). Here, the techniques are applied to sedimentary, high-grade metasedimentary and high-grade meta-igneous rock artefacts.
Geological setting Pictish sculpture
(b)
Fig. 1. Examples of carved early medieval (Pictish) symbol stones from East Central Scotland. (a) Aberlemno No. 2; (b) Meigle No. 3.
Meffan, together with individual sculptures at other sites (Fig. 3). West Highland sculpture collections surveyed are two of the largest and most representative collections, located at Iona and Oronsay (Fig. 3).
The oldest rocks underlying the sites of early medieval sculpture belong to the Dalradian Supergroup and crop out to the NW of the Highland Boundary Fault zone (Fig. 4) in the Grampian Highland terrane. They were deposited as sandstones and mudstones (principally marine), which were subsequently metamorphosed to cleaved grits, shales, phyllites and schists. They are considered to be of Cambrian age (540-490 Ma) and were subjected to polyphase deformation during the Caledonian Orogeny. During the Early Devonian ( 4 0 9 - 3 8 6 M a ) vast quantities of sediment were eroded from the Caledonian mountains and carried by large rivers to a subsiding area of low ground in the area of the present Scottish Midland Valley. Coarse detritus was deposited in large conglomerate fans whereas sandstones and siltstones were deposited on extensive alluvial plains by braided rivers, in a semi-arid environment. This dominantly arenaceous and commonly red-coloured, non-marine succession is known as the Lower Old Red Sandstone and is c. 9000 m thick. The term Old Red Sandstone has been used in the UK since 1822 (Conybeare & Phillips 1822) to denote the terrestrial sediments that are roughly equivalent in age to the Devonian marine deposits in SW England and continental Europe. These are generally recognized as being c. 4 0 0 - 3 6 0 M a old, although there is still considerable debate as to the exact age of the Scottish Old Red Sandstone (McKerrow et al. 1985). The six major lithostratigraphical divisions of the Lower Old Red Sandstone are the Stonehaven, Dunnottar, Crawton, Arbuthnott, Garvock and Strathmore groups, with various subdivisions at formation level (Armstrong & Paterson 1970). Associated with the sedimentary units are a number of contemporaneous volcanic rocks (particularly lavas) of similar age. No sculptures in this group have been carved from volcanic
GEOLOGY OF SCOTTISH MEDIEVAL SCULPTURE
(a)
285
(b)
Fig. 2. Examplesof late medieval sculpture from the West Highlandsof Scotland. (a) Slab of Angus MacDonald, Iona. (b) Slab of Murehad MacDuffieof Colonsay, 1539; Oronsay Priory.
rock, therefore the Old Red Sandstone lavas will not be considered further here. Early Devonian sedimentation and volcanism in the area was terminated by the onset of earth movements (from around 386 Ma) producing two asymmetric folds with a N E - S W trend: the Strathmore Syncline and the Sidlaw Anticline. Younger sequences were subsequently deposited and eroded, leading to the Lower Old Red Sandstone forming the principal country rock in the area of study. The sculptures that form the early medieval (Pictish) dataset are located within the
northeastern part of the Midland Valley of Scotland. This area lies almost entirely within the Midland Valley terrane and crops out as Lower Old Red Sandstone strata. One sculpture (the Dunfallandy Stone) is sited on Dalradian metasediments (Fig. 4). Although the Dunfallandy Stone is located to the north of the Highland Boundary Fault, within Dalradian metasediments, the lithology of the sculpture is of Lower Old Red Sandstone. Potential sources of rock have been identified from various outcrops of Lower Old Red Sandstone, indicating that only rocks from the
S. MILLER ETAL.
286
j I"
~c9
Oro
Fig. 3. Map of Scotland showing the distribution of Pictish sculpture (East Central Scotland (circles)) and West Highland sculpture (circles with crosses). Arbuthnott, Garvock and Strathmore Groups are likely to have been utilized for the stone sculptures under investigation.
extensively between the Moine Thrust and the Great Glen Fault (Fig. 5). The Moine is subdivided into the Morar, Glenfinnan and Loch Ell Groups. The Morar and Loch Eil Groups are shallow marine sediments whereas the Glenfinnan Group is indicative of transgressive deposition. The Dalradian Supergroup is a succession of clastic sediments and limestones (both principally marine) with minor volcanic rocks, which were subsequently metamorphosed to cleaved grits, shales, slates, phyllites, schists, metalimestones and meta-igneous rocks. Dalradian sedimentation occurred during a prolonged extensional phase, lasting c. 330 Ma (c. 800470 Ma). The rocks were subjected to polyphase deformation during the Caledonian Orogeny. The succession is at least 25 km thick and is divided into the Grampian, Appin, Argyll and Southern Highland Groups. The youngest rocks underlying the West Highland sculptures are igneous rocks of Tertiary age, associated with volcanic centres that were active between c. 60.5 and 55 Ma. The extrusive sequences range from alkali olivine basalt to trachyte, erupted from fissures. Intrusive rocks, representing the now exposed roots of major volcanic centres, display a wide range of rock types. No sculptures in this group have been carved from igneous rock, therefore the Tertiary lavas will not be considered further here. The sculptures that form the late medieval dataset reported here are located on the islands of Iona and Oronsay in the West Highlands. The geology of the two islands is dominated by Lewisian and Torridonian rocks.
Methods
West Highland sculpture The oldest rocks underlying sites of West Highland sculpture belong to the Lewisian Complex of the Hebridean terrane. The complex comprises a variety of gneissose rocks, metagreywackes, quartzite, meta-limestones and meta-igneous rocks with polyphase metamorphic and deformational histories. Locally, this ArchaeanProterozoic basement is overlain by Proterozoic sediments, the Torridonian Supergroup. Some sculpture also rests on Moine and Dalradian Supergroup suites cropping out within the Grampian Highland and Northern Highland terranes (Fig. 5). The Moine Supergroup is a thick succession of strongly deformed and metamorphosed siltstones, mudstones and sandstones that crop out
All petrological analysis was undertaken independently of the magnetic susceptibility measurements to ensure that no bias was recorded in the grouping of the rock types.
Petrology All available sculptured stones have been examined using non-destructive petrological techniques to provide a 'field identification' of the rock type. This type of petrological analysis has provided a basic identification of rock type and has been used to distinguish between general rock types. All examinations included the following measurements: (1) colour (with reference to Munsell soil colour charts); (2) grain size (with reference to standard grainsize measurements on the micrometre scale);
GEOLOGY OF SCOTTISH MEDIEVAL SCULPTURE 5
I
287
0
Upper Old Red Sandstone & younger strata Strathmore Group - sedimentary Garvock Group - sedimentary
r
] Arbuthno4t Group - sedimentary Arbuthnott Group - igneous
I
'/ "~
Lower Old Red Sandstone
Crawton Group- sedimentary & igneous
gunoC~r Group- sedimentary Sto neh=ven Group - sedimentary
(~
J
Dalradian & associated strata OUNKELE
O
Locations of early medieval
\
"~
# ..
0
5
0
15
10 10
20 3(1
20
25Mdes
40 K,IometreS
0
Fig. 4. Geological map of East Central Scotland showing the distribution of Pictish sculpture in the area under investigation. (3) macroscopic mineralogy (i.e. mineralogical content that can be ascertained by examination with 10× magnification hand lens); (4) textural and structural characteristics such as parallel bedding or lamination, cross-bedding, jointing, other planar fabric and grain-size variation;
(5) clast distribution and composition; (6) weathering characteristics. Division of the specimens into 'rock types' has been primarily based on the textural and mineralogical characteristics. Colour has been used only as a general guide to overall appearance because,
Tertiary basalts
Tertiary intrusive rocks
t
I
Dalradian metamorphic rocks
Moine metamorphic rocks
Lewisian gneiss Locations of late medieval sculpture
Fig. 5. Geological map of Western Scotland showing the location of the main late medieval sculpture centres.
288
S. MILLER ET AL.
in many cases, the sculptures have undergone varying degrees of weathering and/or cleaning, both activities that could significantly alter the colour of the surface of the specimen. All outcrop specimens (i.e. potential source rocks) have been examined using petrological techniques to classify rock type. In addition to the measurements taken for the sculptures, all examinations of outcrop specimens also include microscopic mineralogy, i.e. mineralogical content from thin-section examination. Samples of potential source (raw) materials include both historical and modern quarry outcrop and other outcrops. Obviously, some of these actual exposures are modern but by sampling at all possible localities we have aimed to assemble a comprehensive dataset, covering as many rock types as possible. All sandstones are classified according to their mineralogy and using the sandstone classification scheme of Folk, where all rocks containing less than 15% fine-grained matrix are classified in terms of the three principal components; quartz, feldspar (plus granite and gneiss clasts) and other rock fragments (Folk 1974).
Magnetic susceptibility Magnetic susceptibility was measured with an Exploranium KT-9 Kappameter, giving a measurement of the true susceptibility. The KT-9 has several operating modes but to achieve a consistency of readings the 'pin mode' of working was used throughout. The accuracy on a flat rock is estimated to be + 3% in the 'pin mode'. Care was taken to avoid magnetic contamination. For the sculptures, this can be caused by three principal sources, namely, methods of mounting (e.g. metal pins or rods), repairs to the sculptures and floor or substrate material (e.g. concrete). At all locations, stones with a thickness less than 6 c m were treated as suspect, as no tables for adjusting the readings from thin rock units are available (WilliamsThorpe et al. 2000). Whenever feasible a series of 12 readings were taken from the front and rear faces of the carved stones away from all possible sources of magnetic contamination. The average of the 12 readings represents one dataset per sculpture. Over 180 datasets for sculpture were taken. Seventy datasets taken from quarry or outcrop locations have been utilized to provide a magnetic susceptibility database of potential sources (raw materials). The complete datasets
equate to 2208 measurements for the sculptures and 840 measurements for the source rocks.
Results Lower Old Red Sandstone outcrop From the petrological examinations of the outcrop rocks, significant differences between sandstones from the Arbuthnott, Garvock and Strathmore Groups can be discerned. Within groups and formations, and even within outcrop localities, there can be significant differences in grain size and gross texture (e.g, cross-bedding, lamination, ripples, etc.). However, the overall mineralogical characteristics are generally consistent within outcrop. From the examination of thin sections, it is clear that there are very recognizable differences between outcrops on a microscopic scale (Table 1). The magnetic susceptibility data suggest that the units studied display a wide range of readings (Fig. 6). Arbuthnott group. All sandstones belong to the Dundee Fm. They are generally arkosic sandstones and litharenites. They are characteristically drab coloured (grey to pinkish grey) although the outcrop at Kingoodie is a dull red colour, as a result of very late stage weathering and development of hematite cement. Generally, the Arbuthnott sandstones contain a much more significant proportion of chlorite than either Garvock or Strathmore sandstones. The range of magnetic susceptibility measurements displayed by the rocks of the Dundee Fm is very varied (Fig. 6a). This suggests that magnetic susceptibility cannot be used as a tool for distinguishing these rocks independently. However, the readings in this group varied from sandstones to much finer silty sandstones, and micaceous flagstones. The finer material, often with an increase in mica, generally results in higher readings. It is therefore possible to characterize different lithologies within the formation. Garvock group. Sandstones cropping out to the NW of the Sidlaw Anticline lie within the Scone Fro. Those cropping out to the SE of the Sidlaw Anticline belong to one of the Red Head Fm, Arbroath Sandstone unit or Auchmithie Conglomerate unit. The Scone Fm sandstones are predominantly litharenites (containing very little or no calcite). The sandstones of the Read Head Fm and Auchmithie Conglomerate are sub-arkosic whereas the Arbroath Sandstone units are calcareous
Litharenite
Litharenite
Litharenite
Calcareous arkose
NO 5074 5312
NO 5074 5312
NO 4605 4948
Arbuthnott Gp Dundee Fm (Baldardo Quarry, Turin Hill)
Arbuthnott Gp Dundee Fm (Baldardo Quarry, Turin Hill)
Arbuthnott Gp Dundee Fm (Balmashanner E Quarry)
Arbuthnott Gp Dundee Fm (Aberlmeno Churchyard outcrop) Arbuthnott Gp Dundee Fm (Baldardo Quarry, Turin Hill)
NO 5074 5312
Petrological classification
Arkose
Grid reference
NO 5222 5557
Stratigraphic group and formation (location) Very angular, well sorted; poly- & monocrystalline qtz, weathered fsp, minor aligned bt & ms, extensive clay matrix, opq grains, rim & cement Very angular, poorly sorted; poly& monocrystalline qtz, weathered fsp, aligned, kinked ms, bt & chl, opq grains, rims & interstitial cement, clay cement, lithic (volcanic) clasts Very angular-subrounded, poorly sorted, immature; poly- & monocrystalline qtz, weathered fsp, kinked bt, ms & chl, extensive hem & clay matrix, dominant lithic (volcanic) clasts Angular- subrounded, poorly sorted, immature; poly- & monocrystalline qtz, weathered fsp, extensive aligned, kinked ms, bt & chl, extensive opq rim & cement, clay & qtz cement Very angular, mod well sorted, grain supported; poly- & monocrystalline qtz, weathered fsp, significant chl, bt & ms, matrix of clay & cc, opq grains and some cement
Petrological characteristics
Weak red
Red claystone blebs Massive and crossbedded units
Massive and silty sandstone interbedded
vcg-cg
f-mg
(Continued)
Reddish grey
Weak red Volcanic
Massive and crossbedded units
vcg-cg
Reddish brown
Colour
Weak red
Massive and crossbedded units
cg
Clast composition
Volcanic
Massive sandstone
Fabric
f-mg
Grain size
Table 1. Petrology of the potential source rocks from the Lower Old Red Sandstone (LORS) of East Central Scotland
Litharenite
Subarkosic siltstone
NO 5547 4485
NO 5547 4485
NO 4516 3955
NO 4516 3955
NO 4405 4846
Arbuthnott Gp Dundee Fm (Carmyllie NE Quarry)
Arbuthnott Gp Dundee Fm (Carmyllie NE Quarry)
Arbuthnott Gp Dundee Fm (Dodd Quarry)
Arbuthnott Gp Dundee Fm (Dodd Quarry)
Arbuthnott Gp Dundee Fm (Forfar Bypass by South Leckaway)
Calcareous arkose
Calcareous arkose
Arkose
Litharenite
Petrological classification
NO 5547 4485
Grid reference
Arbuthnott Gp Dundee Fm (Carmyllie NE Quarry)
Stratigraphic group and formation (location)
Table 1. Continued
Very angular-subrounded, poorly sorted, immature; poly- & monocrystalline qtz, weathered fsp, aligned, kinked bt, ms & chl, some opq grains & cement, clay & qtz cement, extensive lithic (volcanic) clasts Very angular-subrounded, poorly sorted, immature; poly- & monocrystalline qtz, weathered fsp, minor opq grains & cement, qtz and clay matrix, extensive lithic (volcanic & granitic) clasts Very angular, well sorted; monocrystalline qtz, weathered fsp, aligned, kinked chl, bt & ms, minor opq grains & cement, clay cement Very angular, mod well sorted; poly- & monocrystalline qtz, weathered fsp, major kinked aligned chl, ms & bt, opq grains, clay cement Very angular, mod well sorted; poly- & monocrystalline qtz, weathered fsp, major kinked aligned chl, ms & bt, opq grains, calc & clay cement Very angular, mod well sorted; poly- & monocrystalline qtz, weathered fsp, major kinked aligned chl, ms & bt, opq grains, minor opq cement, calc & clay cement
Petrological characteristics
f-rag
vfg
Cross-bedded and massive sandstone