Research paper Optical dating of quartz and feldspars: A comparative study from Wonderkrater, a Middle Stone Age site of South Africa Magali Barré a, * , Michel Lamothe a , Lucinda Backwell b, c , Terence McCarthy d a Laboratoire de luminescence Lux, Département des sciences de la Terre et de l’atmosphère, Université du Québec à Montréal, PO 8888, Succ. Centre-Ville, Montréal, Québec H3C 3P8, Canada b Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Private Bag 3, Wits, 2050, Johannesburg, South Africa c Institute for Human Evolution, University of the Witwatersrand, Private Bag 3, Wits, 2050, Johannesburg, South Africa d School of Geosciences, University of the Witwatersrand, Private Bag 3, Wits, 2050, Johannesburg, South Africa article info Article history: Received 18 October 2011 Received in revised form 15 January 2012 Accepted 19 January 2012 Available online 1 February 2012 Keywords: IR-luminescence K-feldspar Quartz Peat Archaeology abstract While quartz is the most used dosimeter, it has been shown that feldspars provide many advantages over quartz, essentially in terms of reproducibility and sensitivity. Unfortunately, they also suffer from instability in their luminescence signal, known as anomalous fading, which leads to an underestimation in age if no correction is applied in a spring and peat mound archaeological context, we explore the possibility of obtaining a single age for both quartz and feldspar fractions from the same sample. This work first highlights the importance of selecting two dosimeters in an archaeological or geological context. It also put in the foreground the time-consuming but gratifying approach of comparing large and small aliquots. Finally, we present feldspars with a barely detectable and measurable fading rate, whatever the protocol applied, suggesting that the solution to anomalous fading might be to find feldspar grains that do not fade. Ó 2012 Elsevier B.V. All rights reserved. 1. Introduction We present luminescence results from an archaeological site at a spring and peat mound of Wonderkrater, situated in Limpopo Province, South Africa (Fig. 1A). Wonderkrater is located in a mixed bushveld environment (Low and Rebelo, 1996), which forms part of the current tropical savanna biome (Rutherford, 1997). The site is a large concave peat mound above a mineral-rich artesian spring. It presently rises 3.5 m above the surrounding landscape and covers an area of approximately 25,000 m 2 (Fig. 1B and C). The area in which the mound occurs is presently a site of sedi- ment accumulation, but this was not always the case, because the bedrock in the area consists of remnants of Permo-Triassic Karoo Supergroup and Proterozoic basement rocks that have clearly experienced erosional degradation (McCarthy et al., 2010). The aim of this study is to provide luminescence ages for deposits characterized by a late Middle Stone Age lithic assemblage and Florisian fauna. With an ability to determine ages beyond 100 ka, luminescence dating (TL and OSL) is a research tool that is increas- ingly being applied to archaeological contexts. In this specific investigation, we explore the possibility of obtaining a unique depositional age for both quartz and feldspar from 2 samples. 2. Sampling site Samples for optical dating were collected from the south wall of Pit B, where the deposits are most clearly stratified (Fig. 1C). The upper section of Pit B yielded little, but from below a depth of 2.1 m, numerous fossils were recovered. A 1 m thick sand layer was encountered commencing at a 3 m depth, which contained a well preserved Middle Stone Age (MSA) lithic assemblage (cores, blades, flakes, hammerstones, grindstones etc.). Radiocarbon ages have been obtained from throughout the excavation, with 14 C younger than 16 ka for that part of the section found above the dated samples. The bottom part returned ages mixing and inversions that are useless for this investigation. Details of the archaeological investigations and fauna will be published in Backwell et al. (in preparation). Three samples located below the water table were collected in this section (Fig. 1C): WK3 from the top of MSA artefact-bearing sand layer at 3.12 m below the surface and WK4 from an under- lying green clay layer 4 m below the surface located in the water table. Measurements for WK5, which was collected by coring 35 cm below the base of the 2007 excavation, are still in progress. * Corresponding author. Tel.: þ1 514 987 3000x6626; fax: þ1 514 987 7749. E-mail address: barre.magali@uqam.ca (M. Barré). Contents lists available at SciVerse ScienceDirect Quaternary Geochronology journal homepage: www.elsevier.com/locate/quageo 1871-1014/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.quageo.2012.01.009 Quaternary Geochronology 10 (2012) 374e379