Cutting 10 is a hominid site character- ised by a high proportion of formal tools but overall low density of artifacts, a low bone : stone ratio, low density of bones ...
Palaeoecology of Africa 1988 19: 213-219
Some features distinguishing various types of occurrence at Elandsfontein, Cape Province, South Africa G. AVERY South African Museum, Cape Town, RSA
ABSTRACT The distribution of animal and artifactual remains at three occurrences at Elandsfontein show clear distinctions. Cutting 10 is a hominid site characterised by a high proportion of formal tools but overall low density of artifacts, a low bone : stone ratio, low density of bones and no hyaena coprolites. Hyaena Den 2 is a hyaena nursery den, as evidenced by damage to bones, the pattern of their distribution, high bone : stone ratio and the presence of numerous coprolites and resembles the distributions in a modem brown hyaena den. Death Site 1, which combines predominance of cranial remains of very large mammals and evidence of trampling, with low artifact density, represents an area near a water source where animals died naturally or were killed by large carnivores or people. Other comparative sites are discussed.
INTRODUCTION The Elandsfontein (Hopefield) fossil site lies some 100 km northwest of Cape Town. Wind erosion has created a dunefield within which dunes shift roughly in accordance with the alternating seasonally prevailing northwesterly or southeasterly winds. There is a general scatter of fossilized bones and stone artifacts on deflated interdune surfaces throughout the complex. The historical background to the site is provided in Singer & Wymer (1968). Recentlyexposed localized concentrations of material in primary context provide examples of three types of accumulation by different agencies. Comparison of the fauna with East African material, and the occurrence of an early Homo sapiens and Acheulian artifacts indicate that the bulk of the material from Elandsfontein is Middle Pleistocene in age (Singer & Wymer 1968, Hendey 1974, Klein 1978). The accumulations reported here also fall within this period. The development of early H. sapiens is considered to have 213
taken place no earlier than some 400-200 000 BP (e.g. Rightmire 1984). Although difficult to establish, direct association of H. sapiens remains with artifacts is likely and the dating of both is consistent. The artifacts belong to the later part of the Acheulian Industrial Complex, which is now considered to have lasted till 200 000 BP at the latest (Singer & Wymer 1968, Volman 1984). An extinct kudu (Tragelaphus [Strepsiceros] sp.), which is also found in the Middle Pleistocene at Duinefontein 2 (DFT 2) with a minimum date of 125 000 BP, Volman (1984) suggests isotope stage 6 (between 195 000 and 128 000 BP). This taxon is not preserved in early Upper Pleistocene occurrences, e.g. Swartklip and Klasies River Mouth (Klein 1976). Its presence supports a Middle Pleistocene date for Hyaena Den 2 (HD 2) and other sites in which examples of this species have been found. An Upper Pleistocene date is also unlikely in view of the fact that anatomically modern man, in association with Middle Stone Age artifacts, is present in the earliest levels at Klasies River Mouth, which are thought to date to some 120 000 to 130 000 BP (Butzer 1978, Singer & Wymer 1982, Volman 1984). OCCURRENCES Although the samples from the sites are generally small except for Cutting 10 (C 10), the artifacts are essentially similar, apart from the large number of hand axes on C 10, which are absent at the other sites. It seems reasonable, therefore, to assume that these sites are roughly contemporary. The occurrences are complex, however, and the problem remains that each may represent an aggregation over time, by both people and other animals. Although the bones and artifacts at each of the surface occurrences appear to be associated with a common palaeosurface, planned excavations of sealed deposits are expected to resolve the possibility that this may be an artifact of deflation. The intention here is to recognize features that will assist with the identification of such complexities and the major contributors to each type of accumulation. Cutting 10 (C 10) Cutting 10 was excavated by Singer & Wymer (1968) and the larger part of the sample, from sealed deposits, was in primary context. The fauna was studied by Klein (1978). Both Singer & Wymer (1968) and Klein (1978) concluded that most of the bones had been left on the site by people after the animals had been killed and butchered. This interpretation was based on the low density of bones, absence of articular association, and the inclusion of cranial elements from very large and small animals, the minimal representation (7%) of carnivores (Klein & Cruz-Uribe 1984) and the low density of 214
artifacts, most of which were formal cutting tools suitable for butchery. Debitage or any other clues suggestive of a campsite, were absent. Klein (1978), following Isaac's (1971) suggestion that low densities of artifactual debris and bone characterize kill or butchery sites, showed that the C 10 sample fell well within the range of low bone : stone ratios and densities at other sites, such as DFT 2 and Torralba, which had been interpreted as butchery or kill sites (Table 1). He recorded no coprolites in C 10 and only limited evidence for the presence of carnivores. Cutting 10 is the only well-preserved occurrence of this nature that has so far been found at Elandsfontein. The composition of this occurrence is the same as that at Torralba B 5 (Table 1). It also shows similarity to DFT 2 horizon 2 in its distribution of bones (Klein 1976: Fig. 1), although densities and the bone : stone ratio on the latter are higher and the contribution of carnivores (11%) perhaps more significant. While the distribution of bones and artifacts appears to be equally random, no large formal tools occurred in the DFT 2 assemblage (Deacon 1976). This trend towards higher bone density and bone : stone ratios is also apparent in the other Torralba samples (Table 1). Torralba B 1 and B 4, in particular, have higher proportions of bones and higher bone : stone ratios, suggesting greater similarity with DS 1, and even HD 2. It is relevant that, in a subsequent analysis, Klein (pers. comm.) found a few hyaena coprolites in the Torralba samples. There is clearly considerable variation between the proportions and densities of the Table 1. Total excavated area (&) and number of specimens (N), with number per square metre (/m2) and proportions (%) of bones, stones and coprolites, and bone : stone ratios Site Area N Bones Stones Coprols Bone : /m2 % m2 /m2 % /m2 % stone 240 716 Cutting 10 1.9 63.7 1.1 36.3 0 0 1.8 : 1 Duinefontein 2(2) 50 437 6.2 71.2 2.4 27.2 0.1 1.6 2.6 : 1 Torralba B 1 616 86.5 13.5 6.4 : 1 Torralba B 2 202 74.3 25.7 x 2.9:1 Torralba B 4 16.9 x 455 83.1 4.9 : 1 Torralba B 2B 184 76.1 23.9 x 3.2 :1 Torralba B4AA 100 75.0 25.0 x 3.0:1 Torralba B 4A 354 79.4 20.6 x 3.8 : 1 Torralba B 5 37.7 x 62.3 313 1.7 : 1 33 437 11.7 88.3 0.5 Hyaena Den 2 3.7 1.1 8.0 24.1 : 1 Bone 'circle' 20 529 26.5 98.7 0.4 1.3 Skeleton Coast Park 21 249 11.9 100.0 +++ Death Site 1 119 413 3.2 91.0 0.3 8.7 0M1 0.2 10.4: 1 x = present in small but unknown numbers +++= many, not counted, preservation factor not known -= unavailable Data on stones at C 10 from Volman (1984); bone circle sample coprolites counted by G.Avery
215
SUMMARY AND CONCLUSIONS It is possible to identify with certainty occurrences where carnivores are the major contributors (HD), those in which human activity is principally reflected (C 10) and those that show little evidence specifically attributable to either carnivore or human activity (DS). The pattern of bone distribution and damage at hyaena dens, with numerous coprolites, are typical of brown hyaenas. The sparse scatter of artifacts suggests that people were not major contributors, but emphasises the complexity of accumulation on any palaeosurface. In this context it should be noted that there is a rhinoceros death site on the same surface within 10 m of HD 2. It appears that kill/butchery sites (C 10) can be recognized by the low density of artifacts, with or without a high number of formal tools, and a generally sparse scatter of bones with limited evidence for carnivore activity. It should be noted that such sites may be used by a succession of animals and that the proportion of diagnostic elements will vary. Localities (DS) at which the cranial remains of very large mammals are common, where there are few smaller species and evidence of trampling, indicate limited carnivore and human activity as such. They appear instead to represent natural deaths or kills from which moveable elements have been scavenged and removed to a hyaena den or camp site. Artifact densities are low and suggest that the human element is not as great as that of natural death or carnivores. The features discussed provide some measure of the important similarities and distinctions between accumulations formed as a result of different agencies and conditions at Elandsfontein and other sites. At the same time, however, it must be emphasised that, at some levels, this is an oversimplification of the complexity of the succession of activities to be expected in these contexts.
ACKNOWLEDGEMENTS Financial assistance from the Council for Scientific and Industrial Research, Foundation for Research Development and field and laboratory assistance by members of the South African Archaeological Society, students at the University of Cape Town, D.M. Avery, M. Rowan, V.M. van Zyl and others are gratefully acknowledged. REFERENCES Brain, C.K. 1967. Bone weathering and the problem of bone pseudo-tools. S. Afr. J. Sci. 63: 97-99.
218
Butzer, K.W. 1978. Sediment stratigraphy of the Middle Stone Age sequence at Klasies River Mouth, Tsitsikama coast, South Africa. S. Afr. archaeol. Bull. 33: 141-151. Conybeare, A. & G. Haynes 1984. Observations on elephant mortality and bones in water holes. Quat. Res. 22: 189-200. Deacon, J. 1976. Report on stone artfacts from Duinefontein 2, Melkbosstrand. S. Afr. archaeol. Bull. 31: 21-25. Hendey, Q.B. 1974. The Late Cenozoic Camivora of the south-western Cape Province. Ann. S. Afr. Mus. 63: 1-369. Hill, A. 1980. A modern hyaena den in Amboseli National Park, Kenya. In R.E. Leakey & B.A. Ogot (eds.), Proc. of the 8th Pan African Congr. on Prehistory and Quaternary Science: 137-138. Nairobi: International Louis Leakey Memorial Institute for African Prehistory. Inskeep, R.R. & Q.B. Hendey 1966. An interesting association of bones from the Elandsfontein fossil site. In Actes du V. Congres Panafricain de Prehistoire et de l'Etude du Quaternaire: 109-124. Tenerife: Museo Arqueologico. Isaac, G.L.L. 1971. The diet of early man: Aspects of archaeological evidence from Lower and Middle Pleistocene sites in Africa. World Archaeology 2: 278-298. Klein, R.G. 1976. A preliminary report on the 'Middle Stone Age' open-air site of Duinefontein 2 (Melkbosstrand, south-western Cape Province, South Africa). S. Afr. archaeol. Bull. 31: 12-20. Klein, R.G. 1978. The fauna and overall interpretation of the 'Cutting 10' Acheulian site at Elandsfontein (Hopefield), south-western Cape Province, South Africa. Quat. Res. 10: 69-83. Klein, R.G. 1983. Palaeoenvironmental implications of Quaternary large mammals in the Lambrechts (eds.), Fynbos palaeoeFynbos region. In H.J. Deacon, Q.B. Hendey & cology: a preliminary synthesis: 116-138. S. Afr. Nat. Scient. Progr. Rept. No. 75. Klein, R.G. & K. Cruz-Uribe 1984. The analysis of animal bones from archaeological sites. Chicago: Univ. of Chicago Press. Rightmire, G.P. 1984. The tempo of change in the Evolution of Mid-Pleistocene Homo. In E. Delson (ed.), Ancestors: the hard evidence: 255-264. New York: Alan R. Liss. Singer, R. & J. Wymer 1968. Archaeological investigations at the Saldanha Skull Site in South Africa. S. Afr. archaeol. Bull. 23: 63-74. Singer, R. & J. Wymer 1982. The Middle Stone Age at Klasies River Mouth in South Africa. Chicago: Univ. of Chicago Press. Volman, T.P. 1984. Early prehistory of southern Africa. In R.G. Klein (ed.), Southern African prehistory and palaeoenvironments: 169-220. Rotterdam: Balkema.
219
