sizes and there is little or no evidence for activity variation. It is the valley and pan, or lake-margin, situation of the Acheulean sites that is the most impressive characteristic, which suggests a degree of terrain specialization, or stenotopia. In this respect, there is a major contrast with the behavior of later modern peoples who occupied all available niches in the landscape (H. J. Deacon 1998).

If the earlier Stone Age Acheulean populations represent archaic people, the middle Stone Age represents modern people. The Fauresmith (an outdated description of what were later classified as Acheulean hand axes) included in the earlier Stone Age by Goodwin and van Riet Lowe (1929) was defined as a small biface industry accompanied by points and blades produced from small prepared cores. These cores are smaller than the Victoria West or proto-Levallois prepared cores (van Riet Lowe 1945) that were made in the Acheulean period (100,000 years ago) to produce large biface blanks. In this sense, the Fauresmith has been seen as transitional between the earlier and middle Stone Ages and was placed in the first intermediate period (Clark 1959). Recorded mainly from open-air stations in Free State and Northern Provinces, the Fauresmith is still poorly defined but may date to more than 200,000 years ago. An early modern human skull (Homo helmei), found in 1932 at Florisbad north of Bloemfontein, dated to some 250,000 years ago (Grün et al. 1996), and associated with essentially modern fauna, may relate to this period. The time range represented by the Fauresmith is crucial to the further understanding of modern human origins.

The middle Stone Age is represented in many cave sequences and open stations. Lacking any good idea of the dating of the sites, Goodwin and van Riet Lowe (1929) defined a number of variations or industries and assumed they were artifacts made by different peoples. Thus, names like Mossel Bay, Still Bay, and Howiesons Poort came into the literature. Scholars had to wait for advances in dating techniques, not only radiocarbon but also alternative dating techniques like uranium disequilibrium dating and luminescence dating, before they could obtain relatively precise measures of the ages of the middle–Stone Age substages that were proposed. A key sequence is that at Klasies River, first excavated in the 1960s (Singer and Wymer 1982). This site, on the southern coast of South Africa, was occupied from the beginning of the late Pleistocene period, 115,000 years ago, and shows the use of marine resources like shellfish and seals from that time and the presence of anatomically modern humans (H. J. Deacon 1995).

Contrary to the expectations of Goodwin and van Riet Lowe (1929) and later researchers (Clark 1959), Howiesons Poort, characterized by distinctive backed tools, was found to be in the middle of the sequence and not at the end. It did not represent a transitional substage to the later Stone Age in which backed tools again occurred. The lesson to be learned was that there was no simple typological evolution between the middle and later Stone Ages. The establishment of the stratigraphic position of Howiesons Poort in the Klasies River sequence (Wurz 1999) has been important because Howiesons Poort is a distinctive horizon marker that can be identified in many middle–Stone Age sequences in southern Africa, which allows typological and chronometric correlations to be made. The dating of Howiesons Poort centers on 70,000 years and postdates the Still Bay with its bifacial points and the Mossel Bay with its thick platformed Levallois-type points.

Molecular biological studies published in the 1980s (Stoneking 1993) led to the formulation of the out-of-Africa hypothesis and focused attention on evidence for modern human origins in Africa. Sites with significant human remains, like Klasies River and the less-well-dated Border Cave in KwaZulu-Natal (Beaumont 1980), assumed particular importance. An apparent anomaly is that early modern human remains in Africa are associated with middle–Stone Age artifacts, whereas in Eurasia in the same time range a different species, or deme, the Neanderthals, considered to be nonmodern in their behavior, are associated with similar middle Paleolithic-Mousterian artifacts. This has led to the proposition that early modern humans in Africa were anatomically but not behaviorally modern (Klein 1995). This proposition, called the later