Includes bibliographical references (leaves 333-355). / This thesis advocates the application of an interdisciplinary approach to the historical archaeology of Tswana towns of the late Moloko period in South Africa, and asserts the importance of examining such sites on a case by case basis against the defined backdrop of their unique historical, political and biophysical contexts. The early 19th century Tlokwa capital of Marothodi, in the Pilanesberg region of South Africa, forms the focus of a study through which the value of this approach is demonstrated. The historical, political and biophysical context of the site is explored, with an emphasis on Tlokwa oral traditions. Archaeological investigation reveals details of settlement organisation, while preliminary ceramic analysis contributes to an understanding of ancestral identity, indicating a possible affinity with early Fokeng lineages stemming from Northern Nguni origins. All of the above is relevant to one particular aspect of production. The organisation of both iron and copper production at Marothodi is explored and analysed against the wider contextual backdrop of the capital. The intensification of metallurgical output, and the adaptation of Tswana cultural codes to the unprecedented demands of living in an aggregated community, demonstrate the degree to which historical context could influence the organisation of production, and consequently the archaeological expression of the town. In summary, this research suggests a period of ascendant political status for the Tlokwa at this time in the history of the chiefdom; a conclusion that could only have been reached through a combination of historical, biophysical, ethnographic and archaeological data.

Bibliography: leaves 238-259. / A method of interpreting the seal body part representation from archaeological sites is presented and applied to three Holocene archaeological assemblages from the west coast of South Africa. The approach that is developed integrates several different methods that have previously been applied to terrestrial species, but that, with few exceptions, have not be.en employed in the analysis of seal remains. Most of the existing taphonomic indices cannot be applied to seals because of their unique physiology. Appropriate field observations and laboratory measurements are used to construct taphonomic indices that can be widely applied to seal bone assemblages. These include: a hardness index that mediates bone destruction through mechanical attrition, a utility index that mediates differential transport of body elements, and two indices that mediate the impact of carnivore ravaging - the carcass consumption sequence, and the carnivore destructive template. A new approach that caters for the simultaneous application of several taphonomic indices to an assemblage, where previously they have been applied individually or in pairs, is developed. In addition to the taphonomic indices, a method of determining ontogenic age is presented, and the potential limits of seal storage are explored.

Bibliography: pages 279-298. / The primary aim of this research is to assess the role of shellfish in prehistoric diet and lifestyle with specific reference to the numerous archaeological sites in the vicinity of Eland's Bay (18°19'E 32°19'S). To give a wider regional context, available data on seven other sites along the southwest coast of South Africa have been included. The quantification, analysis and comparison of the faunal data from all these sites have proved informative on the adequacy and balance of prehistoric coastal diet and on aspects of human ecology and behaviour.

Includes bibliographical references (leaves 274-325).

Bibliography: pages 54-60. / In recent years it has become increasingly apparent that an important aspect of southern African late Stone Age technology was the use of glues or adhesives to heft implements, and this correlates largely with the viability of microlithic industries. The number of sites yielding evidence of hafting have increased considerably since excavators have become aware of its presence. For the most part, the remains consist of odd lumps or traces on implements, but there are about a dozen fairly complete moulded pieces that allow some comment to be made as to the hafting strategy, and this technique is considered. To date, no success has been registered in identifying the actual ingredients used, and the main objective of this paper is to indicate that thin layer chromatography can be used to this end.

The fossil locality Langebaanweg, is world renowned for its vast abundance and diversity of fauna. Langebaanweg is the only fossil site in South Africa that preserves remains from the Late Miocene to Early Pliocene. The Mio-Pliocene is a period of major climatic change, with global temperatures falling, and increased aridity and seasonality across the African continent. Langebaanweg, located on the coastal platform of Southern Africa, is ideally placed to investigate how these major changes in climatic conditions in the Mio-Pliocene affected ecology, environment and animals in south-western Africa. The majority of fossils from Langebaanweg were recovered from the early Pliocene Varswater Formation. Specifically, the two major fossil bearing horizons, the Langeberg Quartz Sand Member (LQSM) and the overlying Muishond Fontein Pelletal Phophorite Member (MPPM). The Varswater Formation has been reconstructed as a fluvio-estuarine environment with localised tidal-flats and marshes. The relationship between the MPPM and the LQSM and their relative ages and depositional environments is controversial. This thesis sought to investigate the depositional environment of the MPPM and LQSM in Langebaanweg E-Quarry, with the aim of identifying the relationship and comparative ages of the two members. In order to investigate this, three test pits were excavated at Langebaanweg and the lithology, sedimentology and taphonomy of remains recovered from the test pits was studied. The test pits were located in the sivathere bonebed, the 1976/2 excavation and in a new locality positioned 200 m to the south of the 1976/2 excavation. The depositional environment of the bonebed and 1976/2 were investigated as well as their relationship to each other. The taphonomic pathways by which bones came to be interred in each test pit was identified. Sedimentologically, the MPPM and LQSM were found to be almost identical and the only major distinguishing characteristic was the lack of phosphate in the latter deposit. This indicates that the two horizons were deposited in the same setting and/or by the same depositional process. Additionally the MPPM and LQSM sediments were dominated by fine to very fine sand grains and muds and silts were virtually absent (except in the MPPM in one test pit). This, together with the texture of the sediments, lead to the conclusion that the MPPM and LQSM, in Langebaanweg EQuarry, had a marine origin. Langebaanweg E-Quarry is here interpreted as a barrier-island estuary formed on a wave-dominated coast. Given the consistency of sediments, taxa and taphonomic characters between the MPPM and the LQSM they are interpreted as representing a single unit or depositional event where the upper parts (MPPM) have been secondarily phosphatised. Under this scenario, there is no unconformity or age gap between the MPPM and the LQSM. Investigations of the depositional environment of the 1976/2 excavation concur with Hendey's (1980) reconstructions for a river channel. The results of this study reject previous reconstructions of the sivathere bonebed as a lag deposit of a river channel (Roberts et al. 2011; Smith & Haarhoff 2006). This study shows that the bonebed was deposited in a low energy freshwater pool associated with the river. No evidence was found for an aquifer-fed spring as suggested by Brumfitt et al. (2013). The MPPM sediments of the test pit located to the south of 1976/2 displayed a dominance of fine muds and silts with a possible high organic content. This was interpreted as a possible marsh habitat. This is the first study to identify a marsh environment at Langebaanweg since the 1980's and all other examples of this type of environment have been destroyed by mining. The identification of this locality will provide exciting new opportunities for future research. Taphonomic analysis has shown that the influence of fire and predation on faunal remains excavated in this study was small and internment in all cases was quick. As a final note, this thesis conducted a review of previous research and excavations at Langebaanweg E-Quarry by Dr Brett Hendey and identified the location of some fossil collecting localities that had been lost. These were plotted onto aerial photographs, which will be beneficial to future researchers trying to identify old locals or identify new deposits.

Bibliography: pages 54-60. / In recent years it has become increasingly apparent that an important aspect of southern African late Stone Age technology was the use of glues or adhesives to heft implements, and this correlates largely with the viability of microlithic industries. The number of sites yielding evidence of hafting have increased considerably since excavators have become aware of its presence. For the most part, the remains consist of odd lumps or traces on implements, but there are about a dozen fairly complete moulded pieces that allow some comment to be made as to the hafting strategy, and this technique is considered. To date, no success has been registered in identifying the actual ingredients used, and the main objective of this paper is to indicate that thin layer chromatography can be used to this end.

This study sought to understand the archaeology of the Zimbabwe Culture capital of Khami through synchronic and diachronic analyses of its material culture. The research employed a number of methodological approaches that included a review of historic documents, surveying and mapping, excavations, museum collection analysis, and artefact studies, in order to collect datasets from various sections of the site, including the walled and the nonwalled areas. The main indication is that there is a great deal of similarity in material culture distribution across the whole site. An analysis of objects by stratigraphic sequence exposes continuity and change in local and imported objects. Dry stone-wall architectural data suggests that the site was constructed over a long period, with construction motivated by a number of expansionary factors. The study confirms that Khami began as a fully developed cultural unit, with no developmental trajectory recorded at Mapungubwe or Great Zimbabwe, where earlier ceramic units influenced later ones. Consequently, this study cautiously suggests that Khami represents a continuity with the Woolandale chiefdoms that settled in the south-western parts of the country and in the adjacent areas of Botswana. On the basis of the chronological and material culture evidence, Khami is unlikely to have emerged out of Great Zimbabwe. However, more research is needed to confirm these emergent conclusions, and to better understand the chronological and spatial relationships between not just Woolandale and Khami sites but also Khami and the multiple Khami-type sites scattered across southern Zambezia.

Includes abstract. Includes bibliographical references.

Includes abstract. / Includes bibliographical references (leaves 106-124). / This dissertation aims to increase our understanding of Papio by: 1) examining variation within and between baboon species; 2) investigating morphological integration in the cranium, and; 3) determining what evolutionary processes produce diversification among baboon crania. Twenty-five three-dimensional cranial landmarks on six-hundred and sixty four baboon crania from seven species were digitized. Forty-seven linear distances, chosen to cover the skull with minimal redundancy, were calculated from these coordinate data.