¹ Previous reviews of the southern African radiocarbon dates which have appeared in this Journal are by Maggs, T. (xviii, 2, 1977, 161–91)Google Scholar, Soper, R. C. (xv, 2, 1974, 175–92)Google Scholar, Sutton, J. E. G. (xiii, 2, 1972, 1–24)Google Scholar and Phillipson, D. W. (xi, 1, 1970, 1–15).Google Scholar

² The terms ‘Early Iron Age’, ‘Middle Iron Age’ and ‘Late Iron Age’, as well as ‘Late Stone Age’, which have been used for classifying southern African archaeological sites in the past, have not been employed in this review. This is in the belief that such a system, although essential in earlier stages of research where it was necessary to establish basic sequences, is currently confusing the interpretation of southern African prehistory by introducing a priori concepts of population movement and cultural breaks. Instead the loose economic categories of ‘hunting and gathering’ and ‘farming’ have been used and chronological developments within natural landscape units stressed.

³ G. Mgomezulu has a survey of recent radiocarbon dates from eastern Africa in preparation.

⁴ Antiquity, xlvi (1972), 265.Google Scholar

⁵ Lerman, J. C., Mook, W. G. and Vogel, J. C., ‘C14 in tree rings from different localities’, in Olsson, I. U. (ed.) Radiocarbon Variations and Absolute Chronology. Nobel Symposium 12 (Stockholm, 1970).Google Scholar

⁶ An outline of the physiographic structure of southern Africa is given by Wellington, J. H., Southern Africa, a Geographical Study (Cambridge, 1955).Google Scholar

⁷ Maggs, T., ‘Some recent radiocarbon dates from eastern and southern Africa’, J. Afr. Hist. xviii, 2 (1977), pp. 161–91.CrossRefGoogle Scholar

⁸ Sandelowsky, B. H., ‘Mirabib – an archaeological study in the Namib’, Madoqua x (4), (1977), 221–83.Google ScholarSandelowsky, B. H., van Rooyen, J. H. and Vogel, J. C., ‘Early evidence for herders in the Namib’, S. Afr. archaeol. Bull. xxxiv (1979), 50–1.CrossRefGoogle Scholar

⁹ Sandelowsky, , Van Rooyen, and Vogel, , ‘Early evidence’.Google Scholar

¹⁰ Pta-1012 dates to the mid-fourth millennium and was cited by Maggs (‘Dates’). This reading is confirmed by Pta-1348.

¹¹ Pta-1368 reads some two millennia older than other samples from the same context. The excavator considers this to be the result of contamination.

¹² Sandelowsky, , ‘Mirabib’.Google Scholar

¹³ L. Jacobson, pers. comm.

¹⁴ Sandelowsky, , Van Rooyen, and Vogel, , ‘Early evidence’.Google Scholar

¹⁵ See Maggs, , ‘Dates’Google Scholar for a review of the chronology for the introduction of sheep in the Southern Cape Region.

¹⁶ Jacobson, L. and Vogel, J. C., ‘Radiocarbon dates for two Khoi ceramic vessels from Conception Bay, South West Africa/Namibia’, S. Afr. J. Sci. lxxv (1979), 230–1.Google Scholar

¹⁷ Sandelowsky, , ‘Mirabib'’.Google Scholar

¹⁸ H. de Villiers, ‘Report on human skeletal remains from a grave (Gorob) in the central Namib desert’, in Sandelowsky, B. H., ‘Mirabib – an archaeological study in the Namib’, Madoqua, x (4), (1977), 221–83.Google Scholar

¹⁹ L. Jacobson, pers. comm.

²⁰ Carr, M. J., Carr, A. C. and Jacobson, L., ‘Hut remains and related features from the Zerrissene Mountain Area: Their distribution, typology and ecology’, Cimbebasia, ii (11), (1978), 237–58.Google Scholar

²¹ Jacobson, L., pers. comm.Google Scholar

²² Jacobson, L., pers. comm.Google Scholar

²³ Sampson, C. G., The Stone Age archaeology of Southern Africa (New York, 1974).Google Scholar

²⁴ Deacon, J., ‘Patterning in the radiocarbon dates for the Wilton/Smithfield complex in southern Africa’, S. Afr. archaeol. Bull. xxix (1974), 3–18.CrossRefGoogle Scholar

²⁵ Humphreys, A. J. B., ‘The Holocene sequence in the northern Cape’. Unpublished Ph.D. thesis, University of Cape Town (1979).Google Scholar

²⁶ Maggs, , ‘Dates’.Google Scholar

²⁷ Humphreys, , ‘Holocene sequence’.Google Scholar

²⁸ Humphreys, A. J. B., ‘The re-excavation of Powerhouse Cave and an assessment of Dr Frank Peabody's work on the Holocene deposits in the Taung area’, Ann. Cape Prov. Mus. (nat. Hist.) xi (1978), 217–43.Google Scholar Some of the determinations from this site were reviewed earlier (Maggs, ‘Dates’).

²⁹ Humphreys, , ‘Holocene sequence’.Google Scholar

³⁰ Horowitz, A., Sampson, C. G., Scott, L. and Vogel, J. C., ‘Analysis of the Voigtspost Site, O.F.S., South Africa’, S. Afr. archaeol. Bull. xxxiii (1978), 152–9.CrossRefGoogle Scholar

³¹ Humphreys, ‘Holocene sequence’.

³² The first formal statement of the western Cape seasonality hypothesis was made by J. E. Parkington, who has carried out a considerable amount of research in this area which is of relevance. For the general model, see: Parkington, J. E., ‘Seasonal mobility in the Late Stone Age’, African Studies, xxxi (1972), 221–43.Google Scholar

³³ Maggs, , ‘Dates’.Google Scholar

³⁴ Robertshaw, P., ‘Excavations at Paternoster, south-western Cape’, S. Afr. archaeol. Bull. xxxii (1977), 63–74.CrossRefGoogle Scholar

³⁵ Robertshaw, P., ‘Archaeological investigations at Langebaan Lagoon, Cape Province’, In van Zinderen Bakker, E. M. (ed.), Palaeoecology of Africa, Vol. 10 (Rotterdam, 1978).Google Scholar

³⁶ Robertshaw, P. and Poggenpoel, C., pers. comm.Google Scholar

³⁷ Smith, A. E. and Ripp, M. R.An archaeological reconnaissance of the Doorn/Tanqua Karoo’, S. Afr. archaeol. Bull. xxxiii (1978), 118–33.CrossRefGoogle Scholar The determinations, as published, have subsequently been corrected and, in addition, two more samples have been processed (Smith, pers. comm.).

³⁸ Maggs, , ‘Dates’.Google Scholar

³⁹ Opperman, H., ‘Excavations in the Buffelskloof rock shelter near Calitzdorp, southern Cape’, S. Afr. archaeol. Bull. xxxiii (1978), 18–38.CrossRefGoogle Scholar

⁴⁰ Opperman, ‘Buffelskloof’.

⁴¹ At the nearby site of Boomplaas the earliest Albany, dated to about 10,000 B.c., is underlain by assemblages of the Robberg Industry, which are absent at Buffelskloof. This longer sequence adds further support to Opperman's interpretation. See Deacon, H. J., Where Hunters Gathered, South African Archaeological Society, 1976.Google Scholar

⁴² Schweitzer, F. R. and Wilson, M. L., ‘A preliminary report on excavations at Byneskranskop, Bredasdorp district, Cape’, S. Afr. archaeol. Bull. xxxiii (1978), 134–40.CrossRefGoogle Scholar

⁴³ Schweitzer, and Wilson, , ‘Byneskranskop’.Google Scholar

⁴⁴ Schweitzer, and Wilson, , ‘Byneskranskop’.Google Scholar

⁴⁵ J. C. Vogel, ‘Radiocarbon dating of the Border Cave sequences; An evaluation of the Pretoria readings’, in Beaumont, P. B., Border Cave. Unpublished M. A. thesis, University of Cape Town (1978).Google Scholar

⁴⁶ An exception is I. Plug's examination of the assemblages from Bushman's Rock Shelter. See Plug, I., ‘Die latere Steentydperk van die Boesmansrotsskuiling in Oos-Transvaal’, Unpublished M.A. thesis, University of Pretoria (1978).Google Scholar

⁴⁷ Davies, O., ‘Excavations at Shongweni South Cave: The oldest evidence to date for cultigens in southern Africa’, Ann. Natal Mus. xxii (1975), 627–62.Google Scholar Cited in Maggs, ‘Dates’.

⁴⁸ Davies, O., pers. comm.Google Scholar

⁴⁹ The original model, proposing an ‘eastern stream’ of migration which included, among other types, the earliest pottery described here was proposed by Phillipson, D. W. (see for a summary, The Later Prehistory of Eastern and Southern Africa (London, 1977)).Google Scholar This hypothesis has been criticized, and an alternative version proposed, by Huffman, T. N., ‘African Origins’, S. Afr. J. Sci. lxxv (1979), 233–7.Google Scholar

⁵⁰ Klapwijk, M., ‘An Early Iron Age site near Tzaneen, N.E. Transvaal’, S. Afr. J. Sci. lxix (1973), 324.Google Scholar

⁵¹ Maggs, T., ‘The Iron Age sequence south of the Vaal and Pongola Rivers: Some historical implications’, J. Afr. Hist. xxi (1980).Google Scholar

⁵² These determinations were provided by the Archaeology Section, Eduardo Mondlane University, Maputo. Matola is described by Cruz e Silva, T., A preliminary report on an Early Iron Age site, Matola IV 1/68 (Maputo, 1976).Google Scholar

⁵³ Hall, M. and Vogel, J. C., ‘Enkwazini, fourth century Iron Age site on the Zululand coast’, S. Afr. J. Sci. lxxiv (1978), 70–71.Google ScholarHa‥, M., ‘Enkwazini, an Iron Age site on the Zululand coast’, Ann. Natal Mus. xxiv (1980).Google Scholar

⁵⁴ The concept of ‘linked dates’ is described by Huffman, T. N., ‘The interpretation of Iron Age radiocarbon dates’, Arnoldia, viii (17), (1977). 1–5.Google Scholar

⁵⁵ Maggs, T., ‘Mzonjani and the beginnings of the Iron Age in Natal’, Ann. Natal Mus. xxiv (1980).Google Scholar

⁵⁶ Klapwijk, , ‘An Early Iron Age Site’.Google Scholar

⁵⁷ Hall, M., The Ecology of the Iron Age in Zululand (in preparation).Google Scholar

⁵⁸ Phillipson, D. W. (The Later Prehistory) has suggested that his ‘eastern stream’ communities were without cattle and that this resource was brought south by later movements of people. Although this hypothesis is not tenable as it stands, as cattle are among the fauna from a number of sixth, seventh and eighth century sites in Natal and the Transvaal, it is still the case that cattle seem unknown from sites of the Matola TraditionGoogle Scholar. See also: Thorp, C., ‘Cattle from the Early Iron Age of Zimbabwe–Rhodesia’, S. Afr. J. Sci. lxxv (1979), 461.Google Scholar

⁵⁹ Huffman, T. N. (‘African Origins’)Google Scholar has suggested that there was a complete break between the Matola Tradition and those ceramic styles that followed it and that this hiatus represents the immigration of a new population. It should be noted, however, that some stylistic elements characteristic of the Matola Tradition are also present in pottery marking this ‘second wave’ and it is quite possible to support a counter hypothesis of internal evolution (T. Maggs, pers. comm.).

⁶⁰ Cattle were among the species identified at the sites of Msuluzi Confluence and Ndondonwana, both in the Tugela River valley. Maggs, T., pers comm., and Msuluzi Confluence, a seventh century Early Iron Age site on the Tugela River', Ann. Natal Mus. xxiv (1980).Google Scholar

⁶¹ There has been little agreement on terminology for classifying ceramics of this period. Huffman, for example. refers to a ‘Bambata, Sterkspruit and NC3…stepped continuum’ (‘African Origins’), while Maggs prefers to abandon such labelling completely (Maggs, T., and Michael, M. A., ‘Ntshekane, an Early Iron Age site in the Tugela Basin, Natal’. Ann. Natal Mus. xxii (1976), 705–39).Google Scholar In this review, we follow the older terminology in referring to a ‘Lydenburg Tradition’, a unit which may be taken as the equivalent of Huffman's ‘continuum’.

⁶² Duarte, R., pers. comm.Google Scholar

⁶³ Evers, T. M., ‘Plaston Early Iron Age site, White River District, Eastern Transvaal, South Africa’, S. Afr. archaeol. Bull. xxxii (1977), 170–89.CrossRefGoogle Scholar

⁶⁴ Inskeep, R. R. and Maggs, T., ‘Unique art objects in the Iron Age of the Transvaal, South Africa’, S. Afr. archaeol. Bull. xxx (1975), 114–138.CrossRefGoogle Scholar

⁶⁵ Evers, T. M. and Vogel, J. C., ‘Radiocarbon dates for Iron Age sites at Lydenburg and White River, Eastern Transvaal’, S. Afr. J. Sci. (1980) (in press).Google Scholar

⁶⁶ Evers, and Vogel, , ‘Lydenburg and White River’.Google Scholar

⁶⁷ Maggs, and Michael, , ‘Ntshekane’.Google Scholar

⁶⁸ Maggs, , ‘Msuluzi Confluence’.Google Scholar

⁶⁹ Maggs, T., pers. comm.Google Scholar

⁷⁰ Inskeep, and Maggs, , ‘Unique art objects’.Google Scholar

⁷¹ Cronin, M., pers. comm.Google Scholar

⁷² Again, the term ‘Gokomere Tradition’ has been adopted as a general category following terminological disagreements. It is taken to include a number of regional facies, some of which are mentioned in the discussion which follows.

⁷³ Pers. comm., Archaeology Section, Eduardo Mondlane University, Maputo.

⁷⁴ Pers. comm., Archaeology Section, Eduardo Mondlane University, Maputo.

⁷⁵ Huffman, T. N., ‘The origins of Leopard's Kopje: An 11th century Difaquane’, Arnoldia viii (33), (1978), 1–23.Google Scholar

⁷⁶ Rightmire, G. P. and Van, N. J. der Merwe, ‘Two burials from Phalaborwa and the association of race and culture in the Iron Age of southern Africa’, S. Afr. archaeol. Bull. xxxi (1976), 147–52.CrossRefGoogle Scholar

⁷⁷ Rightmire and Van der Merwe, ‘Two burials’.

⁷⁸ Robey, T., ‘Mpambanyoni: A late Iron Age site on the Natal south coast’, Ann. Natal Mus. xxiv (1980).Google Scholar

⁷⁹ Soper, , ‘Dates’.Google Scholar

⁸⁰ Vogel, , ‘Border Cave sequence’.Google Scholar

⁸¹ Maggs, T., pers. comm.Google Scholar

⁸² Hall, M. and Maggs, T., ‘Nqabeni, a Late Iron Age site in Zululand’, S. Afr. Archaeol. Soc. Goodwin Series, iii (1979), 159–76.CrossRefGoogle Scholar

⁸³ Hall, M., ‘An Iron smelting site in the Hluhluwe Game Reserve, Zululand’, Ann. Natal Mus. xxiv (1980).Google Scholar

⁸⁴ N. J. Van der Merwe, pers. comm.

⁸⁵ Hall, and Maggs, , ‘Nqabeni’.Google Scholar

⁸⁶ T. M. Evers, pers. comm.

⁸⁷ Maggs, , ‘Dates’.Google Scholar

⁸⁸ Garlake, P., ‘An investigation of Manekweni, Mozambique’, Azania. xi (1976). 25–48.CrossRefGoogle Scholar

⁸⁹ Henthorne, D. I., Parkington, J. and Reid, R. C., ‘An archaeomagnetic survey of Mgungundlovu’, S. Afr. Archaeol. Soc. Goodwin Series, iii (1979), 149–58.CrossRefGoogle Scholar

⁹⁰ Henthorne, , Parkington, and Reid, , ‘Mgungundlovu’.Google Scholar

⁹¹ Cited in Maggs, ‘Dates’. See also Maggs, T., Iron Age Communities of the Southern Highveld (Pietermaritzburg, 1976).Google Scholar

⁹² Taylor, M. O. V., ‘Late Iron Age settlements on the northern edge of the Vredefort Dome’, Unpublished M.A. thesis, University of the Witwatersrand (1980).Google Scholar

⁹³ Type N is an architectural style of stone building, with a surrounding wall and an inner ring of primary enclosures linked by secondary walling. Distribution is concentrated on the Vaal and Klip Rivers, extending north and southwards to areas of prominent hills. See Maggs, Iron Age Communities.

⁹⁴ Type Z sites are also built of stone and are associated with the Kubung, an offshoot of the Rolong. See Maggs, , Iron Age Communities.Google Scholar

⁹⁵ Taylor, , Late Iron Age Settlements.Google Scholar

⁹⁶ Taylor, , Late Iron Age Settlements.Google Scholar

⁹⁷ Taylor, , Late Iron Age Settlements.Google Scholar

⁹⁸ Klapwijk, , ‘An Early Iron Age site’.Google Scholar

⁹⁹ Earlier determinations from Silver Leaves were cited by Maggs (‘Dates’) and the new dates have been provided by M. Klapwijk, pers. comm.

¹⁰⁰ Phillipson, , The Later PrehistoryGoogle Scholar, and Huffman, , ‘African origins’.Google Scholar

¹⁰¹ Huffman, T. N., pers. comm.Google Scholar

¹⁰² Huffman, T. N., pers. comm.Google Scholar

¹⁰³ Bambata pottery has been regarded as an adoption by hunting and gathering communities (see Phillipson, , The Later PrehistoryGoogle Scholar) but has subsequently been reincluded within the Lydenburg cluster (Huffman, ‘African origins’). Despite this rather crucial change in status it has never been adequately defined.

¹⁰⁴ Huffman, T. N., pers. comm.Google Scholar

¹⁰⁵ Similar sites at Coronation Park, Salisbury, have been dated to the tenth century a.d. See Soper, ‘Dates’.

¹⁰⁶ Huffman, T. N., pers. commGoogle Scholar. The sample was from an earlier excavation by K. Robinson.

¹⁰⁷ Hanish, E., pers. comm.Google Scholar

¹⁰⁸ See, for example, Fagan, B. M., Southern Africa during the Iron Age (New York. 1965)Google Scholar, or Robinson, K. R., ‘The Leopard's Kopje culture: its position in the Iron Age of southern Rhodesia’, S. Afr. archaeol. Bull. xxi (1966), 5–51.CrossRefGoogle Scholar

¹⁰⁹ Huffman, T. N., ‘The Leopard's Kopje tradition’, Mem, natl. Mus. Rhod. vi (197) 4, 1–150.Google Scholar

¹¹⁰ Huffman, ‘Eleventh century Difaquane’.

¹¹¹ Sinclair, P., pers. comm.Google Scholar

¹¹² Huffman, T. N., pers comm.Google Scholar

¹¹³ Huffman, T. N., pers. comm.Google Scholar

¹¹⁴ Hanisch, E., pers. comm.Google Scholar

¹¹⁵ Hanisch, E., pers. comm.Google Scholar

¹¹⁶ Plug, I., Dippenaar, N. J. and Hanisch, E., ‘Evidence of Rattus rattus (House Rat) from Pont Drift, an Iron Age site in the northern Transvaal’, S. Afr. J. Sci. lxxv (1979), 82.Google Scholar

¹¹⁷ Huffman, , ‘Leopard's Kopje’.Google Scholar

¹¹⁸ Hanisch, E., ‘Excavations at Icon, northern Transvaal’, S. Afr. Archaeol. Soc. Goodwin Series, iii (1979) 72–9.CrossRefGoogle ScholarVoigt, E., ‘Faunal remains from Icon’, S. Afr. Archaeol. Soc. Goodwin Series. iii (1979), 80–5.CrossRefGoogle Scholar

¹¹⁹ Prendergast, M. D., pers. comm. and ‘A new furnace type from the Darwendale Dam Basin’, Rhodesian Prehistory vii (1975), 16–20.Google Scholar

¹²⁰ M. D. Prendergast, pers. comm.

¹²¹ Prendergast, M. D., ‘Stone-reinforced furnaces from Masembura Tribal Trust Land, Rhodesia’, Rhodesian Prehistory vii (1977), 17–18.Google Scholar

¹²² Prendergast, M. D.. ‘Chisvingo Hill furnace site, northern Mashonaland’, S. Afr. Archaeol. Soc. Goodwin Series, iii (1979), 47–51.CrossRefGoogle Scholar

¹²³ Gardner, G. A., ‘Mapungubwe’, Pretoria (1963).Google Scholar

¹²⁴ Vogel, J. C., ‘Radiokoolstofdatering van nedersettings uit die ystertydperk op Greefswald’, Unpublished paper.Google Scholar

¹²⁵ Vogel, , ‘Radiokoolstofdatering’.Google Scholar

¹²⁶ Vogel, , ‘Radiokoolstofdatering’.Google Scholar

¹²⁷ Vogel, , ‘Radiokoolstofdatering’.Google Scholar

¹²⁸ Vogel, , ‘Radiokoolstofdatering’.Google Scholar

¹²⁹ Fouche, L., Mapungubwe, ancient Bantu Civilization on the Limpopo (Cambridge, 1937).Google Scholar Gardner, Mapungubwe.

¹³⁰ Voigt, E., ‘The faunal remains from Greefswald as a reflection of Iron Age economic and cultural activities’, Unpublished M.A. thesis, University of Pretoria (1978).Google Scholar

¹³¹ For example, Fagan, B. M., ‘Zambia and Rhodesia’, in Shinnie, P. L. (ed.), The African Iron Age (Oxford, 1971).Google Scholar

¹³² The chronology established on the basis of previous radiocarbon dates is summarized in Garlake, P. S., Great Zimbabwe (London, 1973).Google Scholar

¹³³ Huffman, T. N. and Vogel, J. C., ‘The controversial lintels from Great Zimbabwe’, Antiquity liii (1979), 55–57.Google Scholar

¹³⁴ Huffman, T. N., pers. comm.Google Scholar

¹³⁵ Huffman, T. N., pers. comm.Google Scholar

¹³⁶ Sinclair, P., pers. comm.Google Scholar

¹³⁷ Vogel, , ‘Radiokoolstofdatering’Google Scholar; Huffman, T. N., pers. comm.Google Scholar