Indications of bow and stone-tipped arrow use 64 000 years ago in KwaZulu-Natal, South Africa.

Lombard, Marlize ; Phillipson, Laurel

[ILLUSTRATION OMITTED]

Introduction

The use of the bow and arrow has been interpreted as an important innovation associated with complex human behaviour during the African Middle-Late Pleistocene period (Sisk & Shea 2009). Hunting with a bow and arrow requires intricate multi-staged planning, material collection and tool preparation and implies a range of innovative social and communication skills. The use of bow and arrow technology has far-reaching implications for the reconstruction of the social, technological and cognitive complexity of its makers.

Thus, hypotheses about the very early use of this technology need to be underpinned by robust, contextualised arguments.

Why stone-tipped weapons?

It has been suggested that the earliest small stone points and backed lithics could indicate early projectile (mechanically projected) hunting technologies (McBrearty & Brooks 2000), but this notion remains to be tested. Stone-tipped weapons may be hand-propelled (as a spear) or mechanically projected (as an arrow). In both cases, the stone tip is attached, or halted, to a shaft or link-shaft made from organic materials such as wood, reed or bone. In East Africa, stone points may have been hafted from roughly 285 000 years ago (McBrearty & Tryon 2006). Southern Africa has an almost continuous proven record of the employment of stone-tipped hunting weapons during the past 100 000 years (Milo 1998; L. Phillipson 2007; Lombard & Clark 2008). The efficacy of stone-tipped hunting weapons has been demonstrated during replication (the latest being Pargeter 2007; Sisk & Shea 2009; Waguespack eta/. 2009; Yaroshevich eta/. 2010), and is prominent in the ethnographic record (Rudner 1979; Ellis 1997). Compared to their organic counterparts, stone-tipped weapons cut through tougher hides, penetrate more deeply and create larger bleeding wounds that kill or immobilise the quarry more rapidly or aid the tracking of injured game (Friis-Hansen 1990). When used to tip weapons, stone is notoriously brittle. This characteristic is sometimes considered desirable because fragments lodged in the prey increase blood loss by enlarging wounds and preventing their closure (Rudner 1979; Ellis 1997) (Figure 1). Where weapon tips with cutting edges were required, stone points or insets provided the answer. By about 45 000 years ago, Neanderthals and anatomically modern humans in the Levant probably used complex stone-tipped weapon systems. This suggests that their last common ancestors did likewise, and that the competency to do so developed first in Africa, earlier than 45 000 years ago (Shea 1993, 2006). It remains uncertain when stone-tipped projectile technology was first introduced into the arsenal.

Size and shape as evidence for projectile use

Increasingly, morphometric studies on African Pleistocene stone points are used to imply aspects of function by comparing them with North American ethnographic and experimental specimens. A decrease in the size and weight of stone points has been interpreted to indicate the onset of projectile technology by about 100 000 years ago (Brooks eta/. 2006). In contrast, TCSA (tip cross sectional area) values do not support the widespread use of mechanically projected weaponry earlier than 50 000 years ago (Shea 2006). A study of points from Rose Cottage Cave, South Africa (Figure 2), suggests that people there used projectile technology by around 30 000 years ago (Mohapi 2007). The dimensions of some early African stone points coincide with those of ethnographic/experimental examples of darts or arrows (Shea 2006). Morphometric analysis can assess the potential of artefact classes to function as projectile tips in quantitative terms (Sisk & Shea 2009), but there exists an interpretative shortfall in the method. It must be recognised that not all pointed stone artefacts measured during such analysis were used as, or intended for, weapon tips (L. Phillipson 1997, 2009). The method is thus unable to test the probability that particular artefacts were employed as projectiles. Morphometric analysis can be considered a useful first step to assess the likely presence of projectile weaponry, but should be followed by use-trace, contextual and other lines of evidential analysis (Sisk & Shea 2009; and see below).

[FIGURE 1 OMITTED]

Even when the identity of stone weapon tips is confirmed, they may indicate uses in weapons other than arrows. These include the production of composite spears with multiple stone insets along the shaft, creating elongated cutting edges (Lombard & Parsons 2008), and spearheads of small, fragile points designed to break in the animals and cause cutting wounds. This potential has been demonstrated for small hafted points used as thrusting and throwing spears (Lombard et al. 2004; Figure 1), and backed pieces that could have been used either as spears or arrows (Pargeter 2007). Thus, the presence of small points in the archaeological record does not automatically indicate projectile technology.

Stone points may not have been made or intended specifically for hand-delivered or projectile application (Mohapi 2007; Lombard & Clark 2008; Lombard & Pargeter 2008). Once projectile technology became an option, there is no reason to believe that it was always used. Many stone tool shapes and sizes can be successfully employed either in arrows or spears. Hunter-gatherer groups in Africa are known to use a variety of weapon systems including bows and arrows, spears, clubs, nets and snares. Their weapon of choice is influenced by season, prey type or group preference (Lombard & Clark 2008). In Venezuela, arrows are used with bows to hunt terrestrial game birds, but the same arrows are used as hand-held spears to kill lizards and rabbits (Greaves 1997). Although recent ethnography cannot be imposed on Pleistocene human behaviour, these examples show that broad generalisations, based on size, can obscure variability and change in hunting technologies.

[FIGURE 2 OMITTED]

Use-traces on stone tools

Functional analyses of more than 1000 Late Pleistocene stone artefacts from Southern Africa have established the use of stone-tipped hunting technologies as far back as 100 000 years ago (Lombard 2007a; L. Phillipson 2007; Lombard & Clark 2008). It has also revealed complexity and variation in hafting configurations, materials and adhesives (Lombard 200% 2007b, 2008), the knowledge of which contributes to our understanding of hunting behaviour and cognitive complexity (Lombard 2009; Wadley et al. 2009). However, since most attributes such as micro-residue distribution patterns and micro-wear will develop similarly on points used to tip spears, darts or arrows, it can be difficult to provide unambiguous evidence for the projectile delivery of stone points.

Such microscopic features as Wallner lines and fracture wings, which can indicate fracture velocity in brittle solids, may in some instances provide a means of determining mode of delivery (Hutchings 1999). Application of this method is, however, limited to the finest grained rocks, such as obsidian, chalcedony, flint and jasper, which were seldom used to manufacture points during the African Pleistocene. The detailed study of macro-fracture patterns seems more promising (Fischer et al. 1984; L. Phillipson 2007; Lombard & Pargeter 2008; Villa et al. 2010; Yaroshevich et al. 2010), but many more experiments and studies are needed before such methods can reliably distinguish between hand-delivered and mechanically projected weaponry. As with morphometrics, interpretations derived from micro- and macro-fracture analyses are not problem-free and should be substantiated with further evidence from micro-residue and micro-wear analyses (Lombard 2005b).

We suggest that the recording of use-traces on geometric artefacts, other than points, has good interpretative potential for distinguishing between pieces used as arrowheads and those used as spear insets. This potential is due to the possibility of identifying the direction of impact and artefact application based on patterning in macro-fracture, micro-wear and micro-residue accrual relative to the sizes and shapes of the geometric forms.

Contextual considerations

Projectors (bows or spear throwers) made from wood or bone are not often preserved, but there may be contextual information alluding to their presence. A fundamental principle implied by the use of bows is the indirect transmission of stored energy, thus the production of bows necessitates the choice of strong, flexible wood types (Bergman 1993). Spring-traps (snares) necessitate an understanding of how to use the energy stored in bent branches (Wadley 2010) (Figure 3). The production of cords with relatively high tensile strength is required for both technologies, as is the use of formal knots. Fishing with a line and hook or harpoon also requires long strings with high tensile strength and reliable knots. Another critical innovation was the concurrent use of two separate tools, each useless without the other, to achieve a single purpose. The use of a bow drill for fire production or for piercing resistant materials may be an archaeologically visible precursor or concomitant to the bow and arrow. Evidence for bow drills would be the presence of drilled holes in hard materials such as stone, shell or bone. While fire sticks and hard tipped drills can be manually twisted, the use of a bow drill frees one hand to apply vertical force to facilitate drilling resistant materials.

The effective range from which hunters can expect to make a kill is quite similar: about 8-18m for a thrown spear, and about 9-25m for a traditional bow and arrow (Friis-Hansen 1990; Churchill 1993; Hughes 1998; Yu 2006). This suggests that the initial advantage of an arrow over a spear may not have been its range, but its compact, portable nature facilitating multiple shots. The system is ideal for following prey over long distances and through heavily vegetated terrain. It is considered a niche-broadening technology as it decreases risks associated with large, dangerous game and increases returns on hunting smaller, fast moving terrestrial species, birds and fish (Sisk & Shea 2009). The inception of projectile weaponry is thus expected to be correlated with shifts in the archaeozoological record (Churchill 1993). In Europe, Japan, North America, and Southern and Central Africa, bow and arrow technology sometimes seems to have accompanied or followed climatic changes which resulted in heavier vegetation cover (D.W. Phillipson 1976; Barham 2000; Yu 2006; Mercader 2008).

[FIGURE 3 OMITTED]

A definite prerequisite for hunting with stone-tipped arrows is hafting. We would therefore expect to find evidence for well-developed hafting technologies alongside stone tools considered for this application. Fletching is sometimes considered a requirement, and potential line of evidence, for bow hunting with stone-tipped arrows (Hughes 1998). An independent study, however, demonstrates that fletching is not essential for effective hunting with traditional African bows. Ju/'hoansi hunters immobilise lions and leopards in Namibia using unfletched arrows. In a test for reliability using their usual lightweight bows (approximately 1 m long, <200g) and traditional unfletched arrow shafts (hollow grass lengths of approximately 0.4m, around 3.3g) tipped with commercial drug dart syringes, an accuracy of less than 0.25m from the centre of target was achieved at ranges of up to 30m (Stander et al. 1996).

The above does not represent an exhaustive study of the contextual evidence and precursor and concomitant technologies to the bow and arrow. However, the discussion so far allows us to draw up a preliminary check-list for detecting the use of bows and stone-tipped arrows in the archaeological record:

* long, strong cords

* formal knots

* use of the latent energy in flexed wood

* fishing and fowling

* snares

* bow drills

* formal hafting technology

* broad-based, varied game procurement

* changes in faunal assemblages

* changes in climate and vegetation

No one attribute of those listed above can be used to indicate early bow and arrow technology, but when associated with morphometric, use-trace and other contextual evidence they can help construct increasingly resilient hypotheses for bow and arrow use during the Late Pleistocene.

Bow and arrow technology of 64 000 years ago

Many of the clues suggested in the list above converge in support of the existence of complex weapon systems during the Howiesons Poort phase in Southern Africa about 59 000 to 65 000 years ago (Lombard 2009). Most of the evidence originates from Sibudu Cave, a site located on a cliff above the Tongati River in KwaZulu-Natal, 15km from the Indian Ocean and just over 100m asl (Wadley 2008). Morphometric dimensions of all 79 stone segments from the Howiesons Poort layers fit within the hypothetical range of arrowheads (Wadley & Mohapi 2008). When length replaces breadth in the calculation, assuming transverse hafting of the lithics, quartz segments still fall within the hypothetical range of arrowheads, but those made of hornfels fall within the range of darts and dolerite segments fall within the range of experimental spears (Wadley & Mohapi 2008). Thus, even when erring on the side of caution, the small quartz segments mostly excavated from the oldest layers with an OSL age of 64 700 [+ or -] 2300 years old (Jacobs et al. 2008) (Figure 4), conform to implement classes that could hypothetically represent arrowheads (Wadley & Mohapi 2008).

Replicated weapons tipped with Howiesons Poort-like segments demonstrated their efficiency as projectiles (Pargeter 2007). Explicit tests for distinctions between thrown spears and projected arrows have not yet been conducted, and many of the segments could have been employed equally successfully as insets for spears or arrows (Lombard & Pargeter 2008). Further experimental support for the function of small segments as arrowheads, recently published by Yaroshevich et al. (2010), shows that when hafted transversely, as suggested for the quartz pieces from Sibudu (Wadley & Mohapi 2008), small segments are particularly efficient in terms of penetrating depth and durability. They also demonstrated that distinctive patterns of damage can be detected on 7.9-26.5% of archaeological tool samples used as projectile tips (Yaroshevich et al. 2010).

The probability that segments and backed tools of the Howiesons Poort industry were used in hunting activities is indicated by macro-fracture analyses performed on 318 artefacts from Sibudu Cave, Klasies River Cave 2 and Umhlatuzana Rockshelter. Between 21-24% of the samples have impact fractures indicative of hunting (Lombard 2007a; Wurz & Lombard 2007). These frequencies are well within the range obtained for Upper Palaeolithic archaeological samples and experimental samples known to have been used as arrowheads (Fischer et al. 1984; Yaroshevich et al. 2010). Several of the small quartz segments and backed pieces from Sibudu and Umhlatuzana have fractures consistent with their having been hafted as transverse arrowheads, as described by Lombard and Pargeter (2008) and Yaroshevich et al. (2010) (Figure 5). Direct evidence for the use of Howiesons Poort segments in meat procurement strategies is derived from micro-residue analysis conducted on 53 segments from Sibudu Cave (Lombard 2007b, 2008). Of the 971 animal residue occurrences documented, 84% of all animal tissue, 67% of all bone and/or collagen, 84% of all blood and 89% of all hair fragments occur along the implements' cutting edges. Use-wear and microresidue analyses have revealed little evidence of their employment as cutting or scraping tools, and have shown that their backed portions were hafted in a variety of positions, using compound adhesive recipes.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Figure 6 shows a quartz segment from Sibudu Cave (approximately 12 x 8mm) with several lines of use-trace evidence indicating its employment as a transversely hafted arrowhead:

* Resinous tree gum is concentrated from the middle towards the backed edge along the length of the tool (Figure 6a). Plant cell imprints on the resin (Figure 6b, rectangle) indicate that the tool was hafted in a transverse direction to the grain of the plant material, probably wood or sturdy grass, used for the shaft.

* Striations originating from the cutting edge of the tool are transversely oriented indicating a transverse use-action: these often initiate from impact scars (Figures 6d & f).

* The use-direction is reiterated by the orientation of the residues on the tool (Figures 6c, e, f & g).

* There is no evidence for the tool being used for longitudinal or diagonal motions. We can thus eliminate its primary function as an end- or diagonally-hafted spearhead, or as being an inset along the side of a multi-component spear or knife.

* The impact scars are consistent to those observed on experimental tools used as transversely hafted arrowheads (Figure 6, orange arrows).

* Animal residues are concentrated along the cutting edge of the tool, often directly associated with the impact scars (Figures 6e, f & g).

[FIGURE 6 OMITTED]

The best-fit interpretation, based on these use-traces, is that this segment was used as a transversely hafted projectile tip. At least five other small quartz backed pieces from the layer dated to 64 000 years old have similar trace sets. The Sibudu quartz backed tool sample is still too small for assemblage-based, micro-residue and use-wear analyses (Lombard 2005a, 2008), but newly excavated material may remedy this.

Contextual support for bow and arrow hunting during the Howiesons Poort phase at Sibudu is derived from organic material excavated in the same layers. For example, they contained a bone point that is comparable to unpoisoned bone arrow points from the Holocene (Backwell et al. 2008). The faunal assemblage indicates a broad range for protein procurement strategies with an emphasis on taxa that prefer closed forested niches, including fast moving, terrestrial and arboreal animals (Clark & Plug 2008). Although there is uncertainty about the modes of procurement, there is evidence that people took fish and birds to the site during the same phase (Plug 2006; Plug & Clark 2008). Archaeobotanical results provide further evidence for a backdrop of mosaic vegetation including evergreen forests (Mlott 2006; Sievers 2006).

The Sibudu faunal assemblage from the site may contain yet another intriguing clue. Cords and knots were used in South Africa before and after the Howiesons Poort phase. Use-wear facets on perforated shell beads around 72 000 years old from Blombos show they were strung (Henshilwood et al. 2004), and similar beads are possibly present during the Still Bay phase at Sibudu by about 71 000 years ago (d'Errico et al. 2008), where plant twine was also used to haft stone tips by about 58 000 years ago (Lombard 2005a). Neither the stringing of beads nor the hafting of spears requires the use of long cords with the tensile strength required for bow production. However, considering the range, size, age and behaviour of animals in the Howiesons Poort faunal assemblage from Sibudu, Wadley (2010) constructs an argument for the use of traps, perhaps including snares. If snares were used, the use of cords and knots which would also have been adequate for the production of bows is implied. Most of the taxa caught in snares were probably small antelope such as the blue duiker. Restraining a struggling antelope, even a tiny one, for any length of time requires cords and knots of considerable strength. If Wadley (2010) is correct in suggesting that bush pigs could also have been caught with snares, much greater robustness for cord and knot technology is implied. As well as providing evidence for the use of cords and knots the employment of snares demonstrates a practical understanding of the latent energy stored in bent branches, the main principle of bow construction.

Conclusions

There is as yet no direct evidence for bows during the African Pleistocene, and the hypothesis that very early, stone points were used to tip darts or arrows remains unsupported by use-trace studies and contextual evidence. Some hypothesised concomitants, such as the use of bow drills, also remain elusive. However, drawing together the results of multiple studies conducted on the stone and organic materials excavated at Sibudu, we offer a robust argument for the presence of bow and arrow technology during this phase in KwaZulu-Natal, South Africa.

Considered individually, the results of morphometric analyses, archaeological experiments and micro- and macro-fracture studies may each have interpretative shortfalls for assessing the general presence of bow and arrow technology. However, when the cumulative and contextualised results of such studies are substantiated by micro-residues and other direct use-trace evidence, there is a strong argument for the use of stone-tipped arrows at Sibudu by 64 000 years ago. We do not suggest that the small quartz segments and backed pieces from KwaZulu-Natal necessarily represent the first stone-tipped arrowheads. As the range of sites, knowledge, experimentation and interpretative repertoires is expanded, multi-stranded evidence may locate earlier usage still.

Acknowledgements

We thank colleagues, friends and referees who took the time to comment on this paper, improving the outcome. Lyn Wadley's continued support for work on material from Sibudu Cave is much appreciated. MI2s research is funded by the National Research Foundation's African Origins Platform. Opinions and mistakes are our own.

Received: 13 August 2009; Revised: 3 November 2009; Accepted: 31 December 2009

References

ALLOTT, L.F. 2006. Archaeological charcoal as a window on palaeovegetation and wood use during the Middle Stone Age at Sibudu Cave. Southern African Humanities 18(1): 173-201.

BACKWELL, L., E D'ERRICO & L. WADLEY. 2008. Middle Stone Age bone tools from the Howiesons Poort layers, Sibudu Cave, South Africa. Journal of Archaeological Science 35:1566-80.

BAHAM, L. 2000. The Middle Stone Age of Zambia, South Central Africa. Bristol: Western Academic & Specialist Press for CHERUB (Centre for Human Evolutionary Research, University of Bristol).

BERGMAN, C.A. 1993. The development of the bow in Western Europe: a technological and functional perspective, in G.L. Peterkin, H.M. Bricker & P. Mellars (ed.) Hunting and animal exploitation in the Later Palaeolithic and Mesolithic of Eurasia (Archaeological Papers of the American Anthropological Association 4): 95-105. Washington (DC): American Anthropological Association.

BROOKS, A.S., L. NEVELL, J.E. YELLEN & G. HARTMAN. 2006. Projectile technologies of the African Middle Stone Age: implications for modern human origins, in E. Hovers & S.L. Khun (ed.) Transitions before the transition: evolution and stability in the Middle Palaeolithic and Middle Stone Age: 233-55. New York: Springer.

CHURCHILL, S.E. 1993. Weapon technology, prey size selection and hunting methods in modern hunter-gatherers: implications for hunting in the Palaeolithic and Mesolithic, in G.L. Peterkin, H.M. Bricker & P.A. Mellars (ed.) Hunting and animal exploitation in the Later Palaeolithic and Mesolithic of Eurasia (Archaeological Papers of the American Anthropological Association 4): 11-24. Washington (DC): American Anthropological Association.

CLARK, J.L. & I. PLUG. 2008. Animal exploitation strategies during the South African Middle Stone Age: Howiesons Poort and post-Howiesons Poort fauna from Sibudu Cave. Journal of Human Evolution 54: 886-98.

D'ERRICO, F., M. VANHEAREN & L. WADLEY. 2008. Possible shell beads from the Middle Stone Age layers of Sibudu Cave, South Africa. Journal of Archaeological Science 35: 2675-85.

ELLIS, C.J. 1997. Factors influencing the use of stone projectile tips, in H. Knecht (ed.) Projectile technology: 37-74. New York; London: Plenum.

FISCHER, A., P.V. HANSEN & P. RASMUSSEN. 1984. Macro and micro-wear traces on lithic projectile points. Journal of Danish Archaeology 3:19-46.

FRIIS-HANSEN, J. 1990. Mesolithic cutting arrows: functional analysis of arrows used in the hunting of large game. Antiquity 64: 494-504.

GREAVES, R.D. 1997. Hunting and multifunctional use of bows and arrows: ethnoarchaeology of technological organisation among Pume hunters of Venezuela, in H. Knecht (ed.) Projectile technology: 287-320. New York; London: Plenum.

HENSHILWOOD, C.S., F. D'ERRICO, M. VANHEAREN. 2004. Middle Stone Age shell beads from South Africa. Science 304: 404.

HUGHES, S.S. 1998. Getting to the point: evolutionary change in prehistoric weaponry. Journal of Archaeological Method and Theory 5(4): 345-403.

HUTCHINGS, W.K. 1999. Quantification of fracture propagation velocity employing a sample of Clovis cannel flakes. Journal of Archaeological Science 26: 1437-47.

JACOBS, Z., R.G. ROBERTS, R.E GALBRAITH, H.J. DEACON, R. GRUN, A. MACKAY, P.J. MITCHELL, R. VOGELSANG & L. WADLEY. 2008. Ages for the Middle Stone Age of South Africa: implications for human behaviour and dispersal. Science 322: 733-5.

LOMBARD, M. 2005a. Evidence for hunting and halting during the Middle Stone Age at Sibudu Cave, KwaZulu-Natal, South Africa. Journal of Human Evolution 48: 279-300.

--2005b. A method for identifying Stone Age hunting tools. South African Archaeological Bulletin 60: 115-20.

--2007a. Evidence for change in Middle Stone Age hunting behaviour at Blombos Cave: results of a macro-fracture analysis. South African Archaeological Bulletin 62: 62-7.

--2007b. The gripping nature of ochre: the association of ochre with Howiesons Poort adhesives and Later Stone Age mastics from South Africa. Journal of Human Evolution 53: 406-19.

--2008. Finding resolution for the Howiesons Poort through the microscope: micro-residue analysis of segments from Sibudu Cave, South Africa. Journal of Archaeological Science 35: 26-41.

--2009. The Howiesons Poort of South Africa amplified. South African Archaeological Bulletin 64: 4-12.

LOMBARD, M. & J.L. CLARK. 2008. Variability in Middle Stone Age hunting behaviour: aspects from the lithic and faunal records, in S. Badenhorst, E Mitchell & J.C. Driver (ed.) People, places and animals of Africa: essays in honour of Ina Plug (British Archaeological Reports International series 1849): 46-56. Oxford: Archaeopress.

LOMBARD, M. & J. PARGETER. 2008. Hunting with Howiesons Poort segments: pilot experimental study and the functional interpretation of archaeological tools. Journal of Archaeological Science 35: 26-41.

LOMBARD, M. & I. PARSONS. 2008. Blade and bladelet function and variability in risk management during the last 2000 years in the Northern Cape. South African Archaeological Bulletin 63: 18-27.

LOMBARD, M., I. PARSONS & M.M. VAN DER RYST. 2004. Middle Stone Age lithic point experimentation for macro-fracture and residue analysis: the process and preliminary results with reference to Sibudu Cave points. South African Journal of Science 100:159-66.

McBREARTY, S. & A.S. BROOKS. 2000. The revolution that wasn't: a new interpretation of the origin of modern behaviour. Journal of Human Evolution 39: 453-563.

McBREARTY, S. & C. TRYON. 2006. From Aucheulean to Middle Stone Age in the Kapthurin Formation, Kenya, in E. Hovers & S.L. Khun (ed.) Transitions before the transition: evolution and stability in the Middle Palaeolithic and Middle Stone Age: 257-77. New York: Springer.

MERCADER, J. 2008. Under the canopy: the archaeology of tropical rain forests. New Brunswick; Piscataway (NJ): Rutgers University Press.

MILO, R.G. 1998. Evidence for hominid predation at Klasies River Mouth, South Africa, and its implications for the behaviour of early modern humans. Journal of Archaeological Science 25: 99-133.

MOHAPI, M. 2007. Rose Cottage Cave Middle Stone Age lithic points: does technological change imply change in hunting techniques? South African Archaeological Bulletin 62: 9-18.

PARGETER, J. 2007. Howiesons Poort segments as hunting weapons: experiments with replicated projectiles. South African Archaeological Bulletin 62: 147-53.

PHILLIPSON, D.W. 1976. The prehistory of eastern Zambia (Memoir, British Institute in Eastern Africa 6) Nairobi: British Institute in Eastern Africa.

PHILLIPSON, L. 1997. Edge modification as an indicator of function and handedness of Aucheulean handaxes from Kariandusi, Kenya. Lithic Technology 22(2): 171-83.

--2007. Reassessment of selected Middle Stone Age artefacts from Rhino Cave and White Paintings Rock Shelter, Tsodilo Hills, Botswana. South African Archaeological Bulletin 62: 19-30.

--2009. Using stone tools: the archaeological evidence from Aksum, Ethiopia (British Archaeological Reports International series 1926). Oxford: Archaeopress.

PLUG, I. 2006. Aquatic animals and their associates from the Middle Stone Age levels at Sibudu. Southern African Humanities 18(1): 289-99.

PLUG, I. &J.L. CLARK. 2008. In the air: a preliminary report on the birds from Sibudu Cave, KwaZulu-Natal, South Africa (Goodwin series 10): 133-42. Vlaeberg: South African Archaeological Society.

RUDNER, J. 1979. The use of stone artefacts and pottery among the Khoisan peoples in historic and protohistoric times. South African Archaeological Bulletin 34: 3-17.

SHEA J.J. 1993. Lithic use-wear evidence for hunting by Neanderthals and early modern humans from the Levantine Mousterian, in G.M. Peterkin, H.M. Bricker & E Mellars (ed.) Hunting and animal exploitation in the Later Palaeolithic and Mesolithic of Eurasia (Archaeological Papers of the American Anthropological Association 4): 89-197. Washington (DC): American Anthropological Association.

--2006. The origins of lithic projectile point technology: evidence from Africa, the Levant, and Europe. Journal of Archaeological Science 33: 823-46.

SIEVERS, C. 2006. Seeds from the Middle Stone Age layers at Sibudu Cave. Southern African Humanities 18(1): 203-22.

SISK, M.L. & J.J. SHEA. 2009. Experimental use and quantitative performance analysis of triangular flakes (Levallois points) used as arrowheads. Journal of Archaeological Science 36: 2039-47.

STANDER, P., X. GHAU, D. TSISABA & X. TXOMA. 1996. A new method of darting: stepping back in time. African Journal of Ecology 34: 48-53.

VILLA, P., S. SORIANO, N. TEYSSANDIER & S. WURZ. 2010. The Howiesons Poort and Middle Stone Age III at Klasies River Mouth, Cave 1A. Journal of Archaeological Science 37:630-55.

WADLEY, L. 2008. The Howiesons Poort industry of Sibudu Cave (Goodwin series 10): 122-32. Vlaeberg: South African Archaeological Society.

--2010. Were snares used in the Middle Stone Age and does it matter? A review and case study from Sibudu, South Africa. Journal of Human Evolution 58: 179-92.

WADLEY, L. & M. MOHAPI. 2008. A segment is not a monolith: evidence from the Howiesons Poort of Sibudu, South Africa. Journal of Archaeological Science 35: 2594-605.

WADLEY, L., T. HODGSKISS & M. GRANT. 2009. Implications for complex cognition from the hafting of tools with compound adhesives in the Middle Stone Age, South Africa. Proceedings of the National Academy of Sciences of the United States of America 106: 9590-4.

WAGUESPACK, N.M., T.A. SUROVELL, A. DENOYER, A. DALLOW, A. SAVAGE, J. HAYNEMAN & D. TAPSTER. 2009. Making a point: wood- versus stone-tipped projectiles. Antiquity 83: 786-800.

WURZ, S. & M. LOMBARD. 2007. 70 000-year-old geometric backed tools from the Howiesons Poort at Klasies River, South Africa: were they used for hunting? Southern African Humanities 19: 1-16.

YAROSHEVICH, A., D. KAUFMAN, D. NUZHNYY, O. BAR-YOSEF & M. WEINSTEIN-EVRON. 2010. Design and performance of microlith implemented projectiles during the Middle and Late Epipalaeolithic of the Levant: experimental and archaeological evidence. Journal of Archaeological Science 37: 368-88.

YU, PEI-LI. 2006. From atlatl to bow and arrow, in F. Seller, R. Greaves & P.-L.Yu. (ed.) Archaeology and ethnoarchaeology of mobility: 201-20. Gainsville (FL): University Press of Florida.

Marlize Lombard (1) & Laurel Phillipson (2)

(1) Department of Anthropology and Development Studies, University of Johannesburg, P.O. Box 524, Auckland Park, Johannesburg 2006, South Africa (Email: mlombard@uj.ac.za)

(2) 11 Brooklyn, Threshfield, North Yorkshire, BD23 5ER, UK (Email: dw.l.phillipson@btinternet.com)