the study of the spread of agriculture, both Khoekhoe pastoralism and the settlement of mixed farming communities; and a field outside the scope of this article, the study of the colonial period, the last 350 years.

Human Origins

It has been established since the 1960s that some sedimentary units in two of the best-known sites, Sterkfontein (Kuman 1994) and Swartkrans (Clark 1993), include undoubted Oldowan and Acheulean artifacts and bone tools. These can be dated by the associated faunas to 2 million years old and are among the oldest dated archaeological occurrences. It has been suggested that the number of burned pieces of bone discovered at Sterkfontein (Brain and Sillen 1988) in deposits that may date to about a million years ago may be early evidence of the antiquity of fire-minding if not of fire-making. After half a century of research, these caverns continue to be a focus of archaeological attention as they continue to provide new evidence of early human behavior.

There are few, if any, acceptable Oldowan-aged sites other than the solution cavern occurrences. However, there is a very strong presence of Acheulean biface makers known throughout the subcontinent, and these would date to between more than 1 million years ago and some 300,000 years ago. They are evidence of the establishment of significant human populations on the subcontinent. As a result of alluvial diamond diggings, the Vaal River terraces and gravels north of Kimberley became known as a prolific source of Acheulean artifacts. Systematic research initiated in the 1930s by van Riet Lowe (Söhnge, Visser, and van Riet Lowe 1937; van Riet Lowe 1952) established Canteen Koppie near Barkly West and Riverview Estates near Windsorton as two of the main sites. Expectations that five or more substages of what Goodwin and van Riet Lowe (1929) defined as the Stellenbosch (Acheulean) culture could be recognized on the principle that bifaces took on more refined forms through time have proved too simplistic (H. J. Deacon 1975). The collections were selected from mining dumps and excavated samples (Beaumont 1999; Mason 1988), and more stratigraphic and dating controls became available only later.

Cornelia in Free State Province and Elandsfontein in Western Cape Province are some of the few Acheulean sites that have provided adequate faunal samples, but in neither case can the fauna be directly associated with the artifacts. A date in the range 400,000–700,000 years ago can be ascribed to these faunas (Klein and Cruz-Uribe 1991), and at Elandsfontein, the Saldanha calvarium, which is morphologically similar to the Kabwe skull from Zambia, was recovered in 1952.

Most Acheulean sites are open stations, and other than the solution cavern occurrences, only three cave sites are known to have been occupied in this time range: Cave of Hearths, Montagu Cave, and Wonderwerk. A human mandible was recovered during 1952 excavations at the Cave of Hearths (Mason 1988) as well as a limited fauna, younger than that from Cornelia. Montagu Cave was reexcavated in 1964 (Keller 1973) and produced large artifact samples of Acheulean and younger age but no fauna. Wonderwerk, a large, deep tunnel-like cave but with some seven meters of deposits of different ages, was again disturbed by guano mining, and it includes important Acheulean horizons (Beaumont 1999). Acheulean sites have proved very difficult to study because of poor preservation of context and associations, which has limited progress in developing hypotheses about the meaning of the Acheulean occurrences.

Since the 1960s, Acheulean studies have emphasized the evidence for activity variation in the relative proportions of large, heavy bifacial tools to small flake tools in separate lake margin occurrences in East Africa. In such situations, the context was assumed to be one of little disturbance. As research has progressed, this assumption has not proved justified, and it is apparent the patterning is more readily explained by the selective transport of different sizes and masses of artifacts (Isaac 1977). Most South African open-air Acheulean sites are in valleys and on their margins, and although these areas have been geomorphologically active there has not been same degree of winnowing of artifact