Fauna from five Later Stone Age sites in the Bushmanland region of South Africa Shaw Badenhorst 1 * , Isabelle Parsons 2 and Elizabeth A. Voigt 3 † 1 Ditsong National Museum of Natural History (formerly Transvaal Museum), P.O. Box 413, Pretoria, 0001, and Department of Anthropology and Archaeology, University of South Africa, P.O. Box 392, UNISA, 0003 South Africa 2 Department of Anthropology and Archaeology, University of South Africa, P.O. Box 392, UNISA, 0003 South Africa 3 McGregor Museum, P.O. Box 316, Kimberley, 8300 South Africa BADENHORST, S., PARSONS, I. and VOIGT, E. A., 2015. Fauna from five Later Stone Age sites in the Bushman- land region of South Africa. Annals of the Ditsong National Museum of Natural History 5: 1–10. In this paper, we present the faunal results of five Later Stone Age sites in the Bushmanland region of South Africa, drawing on analytical work conducted by the late E. A. (Liz) Voigt. The sites date from the end of the last millennium BC to recent times, the period during which evidence for herding practices among stone-tool users first occurs in the southern African archaeological record.Based on existing literature, three of the five sites, Melkboom 1, Biesje Poort 2 and Bokvasmaak 3, were expected to yield remains of domestic animals, while faunal components indicative of foraging were predicted for Jagt Pan 7 and Vlermuisgat. Instead, domestic animal remains were identified at Bokvasmaak 3 and Vlermuisgat.Closer examination suggests that stock-keepers may have occupied Bokvasmaak 3, as it is an open-air site that yielded sheep, cattle and even dog remains. Overall, the faunal results indicate groups of people that hunted and gathered a variety of animal species, drawing on a range of procurement techniques and an intimate knowledge of the natural landscape. Keywords: Swartkop, Doornfontein, Herders, Hunter-Gatherers, Later Stone Age, Northern Cape. INTRODUCTION A decade ago the late E. A. (Liz) Voigt agreed to conduct the faunal analyses for a research project into the Bushmanland Swartkop and Doornfontein archaeological signatures, originally defined by Peter Beaumont (e.g., Beaumont et al., 1995). He attributed the Swartkop and Doornfontein signa- tures to two different socio-economies, namely that of hunting and gathering and of herding, respec- tively (Beaumont and Vogel, 1984, 1989; Beaumont et al., 1995). The research project, conducted by one of us (I.P.), investigated four small assemblages previously excavated by Beaumont and housed at the McGregor Museum in Kimberley, South Africa, and a fifth newly excavated assemblage. The exist- ing assemblages were obtained from the open-air sites Melkboom 1, Bokvasmaak 3, Biesje Poort 2 and Jagt Pan 7 during the 1980s and 1990s. All but the last, Jagt Pan 7, were tentatively associated with Doornfontein herders (Beaumont et al., 1995). The fifth assemblage was obtained from a small cave, Vlermuisgat, in 2005 (Fig. 1) (Parsons, 2006, 2008). Considering continued research efforts towards a well-resolved reconstruction of the emergence of herding lifeways among southern African stone-tool users, it now seems that the time is appropriate for a more comprehensive publication of the faunal results for these assemblages. This paper draws heavily on Voigt’s (2006) original unpublished report, and aims to complement published analytical results for the cultural material in the five assemblages (Parsons, 2007, 2008). Disseminating these findings also has a wider relevance than illuminating subsis- tence practices in the Bushmanland region during the last 2000 years. As Liz herself put it, studies like hers ‘[u]ltimately contribute to our knowledge of the recent mammal distribution of our country and our knowledge of past landscapes’ (Voigt, 2006: 281). We did not re-examine any of the faunal remains to confirm or extend Voigt’s findings, and our discussions in this paper are based entirely on her reported identifications. She used the skeletal collections housed at the McGregor Museum in Kimberley and the National Museum in Bloemfontein, both in South Africa (Voigt, 2006), and analysed the entire excavated faunal sample from each site. Measurements of identified specimens were not taken, and cannot be provided here. Unfortunately, the size of screens used during the excavation of Melkboom 1, Bokvasmaak 3, Biesje Poort 2 and Jagt Pan 7 is also unknown. If a large mesh size was used, small elements of medium and small taxa may not have been retrieved. A 1.6 mm mesh was used at Vlermuisgat, ensuring the retrieval of such elements (Parsons, 2006). †, Deceased. *Author for correspondence. E-mail: shaw@ditsong.org.za