© Texier The International Symposium on Archaeometry, Leuven, 28 May - 1 June 2012 GEOLOGICAL ORIGIN OF OCHRE FROM THE MIDDLE STONE AGE SITE OF DIEPKLOOF ROCK SHELTER (SOUTH AFRICA): TOWARDS THE RECONSTRUCTION OF SUPPLY STRATEGIES L. DAYET1, F.-X. LEBOURDONNEC1, F. DANIEL1, P.-J. TEXIER2 1Institut de Recherche sur les Archéomatériaux, Centre de Recherche en Physique Appliquée à l’Archéologie, UMR 5060 IRAMAT-CRP2A, Université de Bordeaux 3/CNRS, 33607 – Pessac Cedex, France. 2De la Préhistoire à l’Actuel : Culture, Environnement et Anthropologie, UMR 5199 PACEA, Université de Bordeaux 1/CNRS – 33405 Talence Cedex, France. During the second part of the Middle Stone Age, hundreds of red ochre pieces were discovered on numerous southern African sites suggesting a strong tradition of ochre use. Several questions are addressed as regard to the ochre pieces: how they were selected? Did the procurement of the raw materials require long-distance supply? Before answering these questions, methodological issues were treated. Indeed, how to discriminate ochre sources? ARCHAEOLOGICAL CONTEXT SURVEY AND SAMPLING Diepkloof Rock shelter was chosen for studying ochre provenance because of its long Middle Stone Age sequence, running from about 45 to 110 ky. Iron-bearing sources in the surroundings of Diepkloof are mostly shale (laminated clayish sedimentary rocks) and ferricrete (iron enriched nodules) outcrops. One shale bed (1) is located directly in the bottom of the shelter. Ferricrete sources were not found less than 20 km from the shelter. This South African site has been excavated under the supervision of Prof. Texier. Thousands of ochre pieces (red iron-bearing materials) were found. © Dayet © Hanhndiek © Dayet Main MSA sites were ochre pieces are reported. Localization of Diepkloof. METHODOLOGY - Samples were examined : fabric, texture and mineralogical components were noticed. - 65 geological and 6 archaeological powder samples were analyzed by ICP-AES (major elements) and ICP-MS (trace elements) at the Service d’Analyse des Roches et des Minéraux, CRPG, Nancy (France). - Results of bulk analyses were treated using statistical analyses. A variant of the protocol proposed elsewhere was used (see Poplaka-Filcoff et al., 2007; Eiselt et al., 2011). CHARACTERIZATION OF THE SOURCES Shale1, 2 and 3 show similar fabrics and textures. Ferricrete nodules appear to be characterzied by various fabrics. Fe content is sensitively higher within ferricretes. Sampling was oriented towards the closest outcrops, along the valley where the site is located. Four sources of shale and two of ferricrete were sampled. ATTRIBUTION OF ARCHAEOLOGICAL PIECES Most of the archaeological shale pieces show similar range of Fe content and macroscopic features than the 4 shale sources. Several archaeological ferricrete pieces show a weathering cortex different from ferricrete sources observed during surveys. This preliminary study focused on shale pieces. A different trace elements distribution between shale and ferricrete is observed when using Pearson correlations on major and trace elements. Elements positively correlated with iron in shale were favored. A. Shale+ferricrete : As, Ba, Cr, Ga, Hf, Ni, Nb, Pb, Sb, Ta, V, Y, Zr. Principal Componenent Analysis of the log Element/Fe (C.E.: 90%). Shale and ferricrete samples are distinguished, even those of the same location (shale4 and ferr1). Elements behavior depends on raw materials. B. Shale : As, Ba, Cr, Nb, Sb, V. PCA of the ratios Element/Fe (C.E.: 90%). Shale sources are discriminated when they are studied apart. The ratios Element/Fe allow a more efficient determination. C. PCA on the shale samples (see B.) and the coordinates of archaeological samples. D. & E. Binary diagrams of selected ratios Elements/Fe, distribution of shale sources and archaelogical samples (C.E.: 90%). At least two pieces can be assigned to shale1, maybe four. Chemical weathering in the sediments may explain why these last two pieces do not fit better with the shale1 distribution. Two other pieces behave differently depending on the trace elements studied. They may come from different sources. This could suggests a non local supply.  Bibliography Conclusion Using trace elements fingerprints is useful to distinguish local ochre sources and not only regional geological formations. In this case the nature of the raw materials must be considered. The origin of some archaeological pieces was determined this way (Dayet et al., in preparation). These are preliminary results which would require a wider geological and archaeological sample. The authors wish to thank the University of Cape Town, the Iziko Museum, and the Service d’Analyse des Roches et des Minéraux team. They are indebted to Sarah Wurz and John Parkington. Eiselt, B. S., Popelka-Filcoff, R. S., Darling, J. A. & Glascock, M.D. (2011). Journal of Archaeological Science 38, 3019-3028. Popelka-Filcoff, R., Robertson, J., Glascock, M. & Descantes, C. (2007). Journal of Radioanalytical and Nuclear Chemistry 272, 17-27. Texier, P.-J., Porraz, G., Parkington, J., Rigaud, J.-P., Poggenpoel, C., Miller, C., Tribolo, C., Cartwright, C., Coudenneau, A., Klein, R., Steele, T. & Verna, C. (2010). PNAS 107, 6180-6185. Tribolo, C., Mercier, N., Valladas, H., Joron, J. L., Guibert, P., Lefrais, Y., Selo, M., Texier, P. J., Rigaud, J. P., Porraz, G., Poggenpoel, C., Parkington, J., Texier, J. P. & Lenoble, A. (2008). Journal of Archaeological Science 36, 730-739. Watts, I. (2002). The South African Archaeological Bulletin 57.