80 South African Archaeological Bulletin 69 (199): 80–86, 2014 Field and Technical Report IN SITU PRESUMPTIVE TEST FOR BLOOD RESIDUES APPLIED TO 62 000-YEAR-OLD STONE TOOLS MARLIZE LOMBARD Department of Anthropology and Development Studies, University of Johannesburg, P.O. Box 524, Auckland Park, Johannesburg, 2006, South Africa E-mail: mlombard@uj.ac.za (Received December 2013. Revised January 2014) ABSTRACT The preservation of blood on ancient stone tools is often challenged. Here I provide a brief report on a presumptive test for blood residues, for the first time applied to stone tools from a Middle Stone Age context. The test is unique because it allows for in situ observation and recording of chemiluminescence during positive results for blood. The results presented here, obtained on an eleven-year-old experimental sample and on quartz points from Sibudu Cave dated to about 62 ka, further strengthen other reports of blood residues found on Stone Age artefacts. I also highlight difficulties and potential for future investigation. INTRODUCTION The potential preservation of blood residues on ancient artefacts used to be the topic of heated debate (e.g. Downs & Lowenstein 1995; Eisele et al. 1995). Yet, blood films and mammalian red blood cells (erythrocytes) or ‘skeletal’ blood cell structures (where the biochemistry changes through time, but the morphology remains), with their distinct biconcave shape and unambiguous size range (e.g. Hortolá 2004, 2013; Lombard 2008) (Fig. 1a–d), are often reported in association with other animal residues on stone tools used thousands of years ago (e.g. Gibson et al. 2004; Hardy 2004; Robertson et al. 2009). Blood, preserved well enough for destructive protein analysis, was found on stone tools from different contexts dating to between 11.2 and 3.9 ka (Gerlach et al. 1996; Seeman et al. 2008; Högberg et al. 2009). During experimentation, Craig and Collins (2002) found that proteins associated with blood resi- dues are often strongly bound to the surfaces of stone tools, probably by means of short-range bonds, which could contrib- ute to their long-term preservation. The haem component of blood is relatively stable, but the protein often undergoes some degradation making the biochemical detection of blood on artefacts difficult, but feasible if appropriately approached (Gurfinkel & Franklin 1988). Thus, even though blood might not always preserve in a biologically active form (e.g. Eisele et al. 1995), it does not follow that its presence cannot be detected on artefacts from some contexts in the deep past. Although blood proteins are soluble in aqueous solutions in their ‘fresh’ state, once dry, they are less easily degraded by hydrolysis reactions, because when blood becomes dehydrated partial denaturation and aggregation takes place (Gillespie 1989). This property also contributes to other instances, unrelated to stone tools, which signify the longevity of blood traces under favourable conditions, for example, red blood cells preserved in the Tyrolean Iceman’s tissue for more than 5 ka (Janko et al. 2012). A decrease in his red-blood-cell-specific metalloprotein haemoglobin, however, indicates some degra- dation. Haemoglobin was also successfully isolated from human bones of up to 4.5 ka (Ascenzi et al. 1985). I have often reported on the presence of blood encountered on archaeological samples, yet have steered clear of overem- phasising its presence. My approach to micro-residue studies is a non-destructive one, tightly linked to specific research questions. All use-traces are considered equally important and ideally plotted on a sample of morphologically similar tools from the same context before functional interpretations are ventured (e.g. Lombard 2005, 2008, 2011). This approach requires that the residues remain in situ on the tools. It is their layering and proximity to each other, their association with mechanical wear patterns and their distribution on a represen- tative sample that hold valuable clues to tool function, and that enables assessment of coincidental residues on a tool (e.g. Wadley & Lombard 2007; Langejans 2011). Always mindful of approaches that can provide supporting evidence without destroying the context and/or integrity of micro-residues, I was recently alerted to the potential of BlueStar ® Forensic – hereafter referred to as ‘the product’ or ‘the chemical compound’ (see Blum et al. 2006 for chemistry and function). The product was developed with DNA preser- vation in mind. It can be applied directly to the tool to reveal distribution patterns of blood remains, and therefore, could have interpretative value regarding tool functions relating to past animal processing activities such as butchery and/or hunting. If shown to be effective through deep time, such an independently developed forensic product might serve as a valuable additional strand of evidence subsequent to the iden- tification of animal residues based on their micro-morpho- logical characteristics. Here I report on the application of the product, and initial results obtained on three small quartz points dating to about 62 ka (Fig. 2a–c), excavated at Sibudu Cave, KwaZulu-Natal, South Africa (see de la Peña et al. 2013 for detailed discussion on context, technology and use-trace sets). PRESUMPTIVE TESTING FOR BLOOD AND A BACKGROUND TO BLUESTAR ® FORENSIC Since the 1980s, several methods of biochemical testing for the presence or absence of blood on stone tools have been suggested and applied with various degrees of success or failure (e.g. Loy 1983; Kooyman et al. 1992; Tuross et al. 1996; Reuther et al. 2006; Heaton et al. 2009; Metheson & Veall 2014). Most of these methods involve the destruction of the residue and/or its context so that subsequent contextual and multi-stranded interpretations become unfeasible. Several techniques were derived from forensic methods, which are based either on the detection of haemoglobin and its derivatives (catalytic tests) or on the detection of proteins and amino acids (noncovalently bound to proteins) (Blum et al. 2006). Luminol (5-amino-2,3-di- hydrophthalazine-1,4-dione) is probably the most widely used presumptive test for blood in crime-scene investigation (Blum et al. 2006). Subjected to an aqueous medium, haem-containing proteins and haemoglobin are able to catalyse the chemilumi- nescence of luminol in the presence of an oxidant. The product reported on here was developed as an improved luminol formulation (Blum et al. 2006).