The Origins and Development of Agriculture in East and Southern Africa

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a widerange of content in a trusted digital archive. We use information technology and tools to increase productivity andfacilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available athttps://about.jstor.org/terms

Wenner-Gren Foundation for Anthropological Research, The University of Chicago

Press are collaborating with JSTOR to digitize, preserve and extend access to Current

Anthropology

This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTC

All use subject to https://about.jstor.org/terms

3 The Origins and Development of Agriculture in East and Southern Africa y David Seddon INTRODUCTION FOR ANY DISCUSSION of the origins and subsequent develop- ment of an idea or technique, a detailed chronological framework is required. Only in Rhodesia has systematic archaeological investigation of "protohistoric" sites been carried out for any appreciable length of time (cf. Summers 1966:463); in other regions investigations have been few or have begun only in the last decade (Fagan 1966a:453; Robinson 1966a: 1 71), while in some areas work can hardly be said to have started even now (Oliver 1966:374). This situation, however, is changing fast (cf. Fagan 1967a: 10 and passim), and a general pattern is now emerging as a result of adequate dating of excavated material in con- junction with analysis of botanical, ethnographic, and linguistic evidence (cf. Fagan 1961 a, 1963a, 1965a; Deacon 1966:50-66; Phillipson 1965a). In attempting to construct this pattern, the prehistorian depends on three different kinds of evidence: (a) direct archaeological evidence-the remains of domesticated plants (in the form of pollen, seeds, or macroscopic remains) in context; (b) indirect evidence-all other material discovered in an archaeological context that, by its nature, suggests the presence of agriculture and a food- producing economy; (c) evidence provided by botanical, ethnographic, and linguistic studies. J. DAVID SEDDON was born in London in 1943 and took his first degree in Archaeology and Anthropology at the University of Cambridge in 1964. He was a Junior Lecturer at the School of African Studies, University of Cape Town, in 1964-65, and Lec- turer in the Department of Geography and Environmental Studies at the University of the Witwatersrand,Johannesburg in 1965-66. He has carried out fieldwork in England and Wales, Iran, and South Africa. As part of the African Studies Programme at the University of the Witwatersrand, he has been especially concerned with pre-European Iron Age settlement patterns and distribution in the Transvaal. His publications include work on the Early Stone Age and on the Iron Age of South Africa. The indirect evidence of material artefacts without the proof of cultivated plants and permanent settlement or of domestic stock is often misleading, and the short list of technological traits that occur only in association with agricultural economies (prepared by Meighan et al. 1958) demonstrates the difficulty of judging subsistence from indirect sources ... (Gabel 1960:438). "Neolithic" elements and cultures have been described from widely distributed localities in Africa, but it has be- come apparent over the past decade that even in the Near East this label is misleading (cf. Braidwood 1957, Braid- wood and Reed 1957, Seddon 1966a, and, for Africa, Hugot in this volume); certain cultural traits can be adopted without changing the basic economy of a group, and the presence at a site of grindstones and axes, or even of pottery and semi-permanent settlement does not, of necessity, demonstrate the existence of agriculture (cf. Forde 1946: 69-100; Klima 1962; Semyenov 1964: 68-70). Our present understanding of the early development of agriculture in Africa, however, depends largely on indirect evidence, and it is on this basis that this review is con- structed. Although the title delimits, in theory, the field of interest, in fact any such survey must consider the related topics of animal domestication, settlement patterns, pottery, and other material artefacts, and the spread of iron-working-these being so closely associated with early agriculture in many of the regions discussed. This situation is not necessarily reproduced in other parts of the conti- nent. The regions considered are those for which some general picture can be provided and even here there is considerable variation in the amount of detail or certainty of the in- formation. Those hardly touched upon, like Angola, Mozambique, and the Congo, though likely to iveal much vital data in the near future (cf. Hiernaux 1959- Vol. 9 * No. 5 December 1968 489This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

1966; Hiernaux and Maquet 1960; Nenquin 1959,1963a), are, as yet, relatively little known. LATE PLEISTOCENE AND POST-PLEISTOCENE BACKGROUND Some recent works have tended to discount the changing climate at the end of the last glaciation as a major factor inducing cultural and economic change in the Near East (Braidwood and Howe 1960, 1962; Reed 1959). Not all are agreed on this point, however (cf. Seddon 1966a), and with regard to Africa Clark (1962a, b) has suggested that the climatic deterioration of the late Pleistocene may have been responsible for the concentration of population in areas made favourable by the presence of permanent water. An increase in temperatures and a decrease in rainfall (which during the Gamblian was probably greater by I than the present mean [Bond 1957; Flint 1959:370]) appears in Africa between ca. 15,000 and 10,000 B.C. Drier conditions are reflected in the geological record at the Haua Fteah, in Cyrenaica, between 10,000 and 8000 B.C. (Higgs 1961, McBurney and Hey 1955), and if this represents a general tendency it is likely to have been associated with a shrinking of water resources in some of the drier parts of the continent (Clark 1954, 1962a; Korn and Martin 1957; Wayland 1954). The concentration of population in well-watered areas may, on the other hand, represent an inevitable stage in the increasing exploitation of natural resources (cf. Braidwood and Howe 1962). In any case, the late- Pleistocene peoples of these areas, unlike their predeces- sors, were sedentary or near-sedentary communities of hunter/gatherers with specialised foraging and collecting techniques which prepared the way for the even more intensive collecting pattern of the Late Stone Age people and their probably greater population density (Clark 1962b). In the woodlands and savannah, vegetable foods, including probably tubers, are likely to have assumed considerable importance (cf. Fagan 1965b :36-41). In the open grasslands, groups of hunters came to rely heavily on certain species for their meat. Many of the sites of the Southern African highveld and Central African savannah produce specialised tools suggesting specialised hunting techniques, and evidence of the use of digging sticks is widespread in East Central and Southern Africa in the Late Stone Age (cf. Clark 1950, 1959; Cooke 1963; Gabel 1963; Goodwin and van Riet Lowe 1929; Mason 1962; Posnansky and Cole 1963). A pattern basic to food production was, therefore, present in parts of East and Southern Africa at least as early as 15,000 B.C. and possibly even before that. The situation somewhat resembles that in Southwest Asia and Central America, where, it is suggested, specialisation and seasonal hunting and collecting laid the basis for food- producing as long as 30,000 years before actual farming (Braidwood and Howe 1960, 1962; Coe and Flannery 1964; Flannery 1965; MacNeish 1964; Perrot 1962). In these areas, planted agriculture appears to have preceded seed agriculture by several thousand years, and it remains possible that this was the case in East and Southern Africa as well (cf. Sauer 1952). There is, however, no convincing evidence from excavated sites that the threshold to food production was crossed before outside influence was felt. EVIDENCE OF EARLY AGRICULTURE EGYPT AND THE SUDAN Direct evidence of the cultivation of wheat, barley, and flax between ca. 5000 and 4000 B.C. is found in the Fayumr A of Egypt (Arkell and Ucko 1965). Other sites to the west and to the south demonstrate the spread of farming concepts in Northern Africa: (a) the suggestion of food production, probably largely pastoral, from the Haua Fteah, in the 5th millennium B.C. (Higgs 1962, McBurney 1960); (b) finds at Meniet, in the Hoggar, of pollen (pos- sibly domestic grain) in the 4th millenium B.C. (Delibrias, Hugot, and Quezel 1957; Firbas 1937, quoted in Erdtman 1943; Hugot and Quezel 1957; Pons and Quezel 1957); (c) the distribution of the "Neolithic" of the Sahara- predominantly cattle-herders but also probably agricul- turalists-whose dates cluster, for the most part, around 3500-3000 B.C., with some possibly as early as the 6th millenium (Clark 1965b, Mori and Ascenzi 1959); and (d) the Naqada I (Amratian) period, for which a reason- able date would seem to be around 3800-3600 B.C. (Braidwood 1957:80; Arkell and Ucko 1965:152) and in which the economy is likely to have been based on agricul- ture and cattle-keeping with hunting and fishing still quite important (Kaiser 1957). Egypt had its first major contact with iron during the Assyrian invasions of the 7th century B.C. (Arkell 1966: 451; Forbes 1950). The Cushites, rulers at that time of Egypt, were forced back into the Sudan and eventually established their capital at Meroe, where it remained from 538 B.C. to ca. A.D. 300 (Arkell 1966:451; Sassoon 1963; Wainwright 1945). Meroitic influences in the form of pottery reached to the borders of Ethiopia and as far south as the Sobat river; it seems possible that it was from here that the knowledge of iron-working spread southward into sub-Saharan Africa, although there is, as yet, no sound archaeological evidence for this supposition (Clark 1965a). It is possible that knowledge of iron-smelting came southward direct from Meroe to Bunyoro in Uganda and from there towards the East African coast (cf. Wain- wright 1954:113-36) but more likely, in view of the pat- tern emerging from radiocarbon determinations, that it spread before the Christian era into West Africa (David- son 1964:70-75; Fagg 1962; Mauny 1952). Many workers feel that a knowledge of iron-working may have assisted the Bantu in their expansion, whether from the savannah regions just south of the Sahara or from the lightly wooded country of the Katanga even farther south (cf. Gray 1962:185-87; Guthrie 1962:281; Murdock 1959; Oliver 1966:366). ETHIOPIA AND THE SOMALI REPUBLIC The "hoe cultivators" of the central and western parts of Ethiopia should probably be regarded as associated with the southern Sudanic regions and West Africa (cf. Bailloud 1959, Clark 1962b, Vaufrey 1947), where incipient food production may have begun before 2000 B.C. Clark (1962b) has suggested an association between these stone hoe cultures and ensete cultivation, as there is a measure of agreement in their distributions. Ensete is widely culti- vated in Ethiopia at the present time (Simmonds 1958, Smeds 1955), and weighted digging sticks are generally used (Simoons l965a. Itis nossible that enet was culti- 490 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

vated in Egypt during the Middle Predynastic period (ca. 3500-3100 B.C.) and was introduced from there into Ethiopia (Laurent-Tackholm 1951). Cultural assemblages farther east are rather different (cf. finds at Agordat near Axum in Eritrea [Arkell 1954]) and suggest a greater emphasis on pastoralism (Clark 1962b). It has been suggested that the stonework from Agordat and some of the pottery indicate a possible con- nection with the late C-group people of lower Nubia- pottery-making pastoralists, probably with subsidiary agriculture, who settled in lower Nubia sometime between 2181 and 2040 B.C. (Arkell 1961) and these people may have been responsible for the introduction into Ethiopia of wheat and barley (Clark 1962b). In eastern Ethiopia, the crops were probably wheat, barley, and teff (Porteres 1962), while in the Gregory Rift area they are more likely to have been finger millet, various sorghums, and Pennisetum (Clark 1962b). The evidence and dating of material in context is still poor, but what little archaeological information there is does not support assertions that this area provided an independent centre of crop domestication, although it clearly was a centre of diversification (Anderson 1960: 75-76; Schnell 1957; Schiemann 1951 :309-13). Contirary views are based primarily on botanical inference (cf. Murdock 1959; Porteres 1950, 1951 a, 1962; Vavilov 1931, 1935, 1949-50). UGANDA, KENYA, AND TANZANIA In East Africa, also, the dating of early agriculture is not precise. Agriculture is assumed, though not proven (Posnansky 1961a), for the makers of dimple-based pot- tery, dated to ca. A.D. 1000 at Nsongezi, in Uganda, where it follows a Late Stone Age industry of ca. A.D. 800-900 (Pearce and Posnansky 1963). Similarities between this pottery and some of the early Zambian ware suggest that it may be earlier than this elsewhere (Fagan 1967a :31-32; Posnansky 1961b). Sites with dimple-based pottery tend to be in areas in which agriculture could have been sup- plemented by hunting and fishing. Iron in small quantities was associated with the pottery at Nsongezi and Kansyore Island; similar pottery was found with tuyeres and furnaces in Ruanda (Hiernaux 1959, 1960; Hiernaux and Maquet 1960), Kasai (Nenquin 1959), and central Tanzania (Smolla 1957), though there is no necessary equivalence in age. (Above the dimple-based level at Nsongezi comes pottery decorated by plain core rouletting; similar ware is dated by association to around A.D. 1350-1500 at the Bigo earthworks in western Uganda [cf. Shinnie 1960]. The tradition lasted only a short time, and the economy was basically pastoralist.) In Kenya, possible evidence for agriculture is associated with pottery assemblages of Sutton's (1964a, b) Class B, which comprises the Gumban A and Hyrax Hill variants of the Stone Bowl culture (also cf. M. Leakey 1943). There is little evidence that sites of this group, once thought to be as early as the 4th millennium B.C., predate the 2nd millenium. Material recovered at Hyrax Hill (M. Leakey 1945) suggests a possible Sudanic origin (Fagan 1965b :45) and an economy that was predominantly pastoral, with hunting and collecting, although grindstones and stone bowls with "traces of carbonaceous elements adhering to the inside" (Fagan 1 965b :45) are suggestive of seed- coIIectinzv and nossiblv domestication. Some of these sites Seddon: AGRICULTURE IN EAST AND SOUTHERN AFRICA produce iron (L. Leakey 1935), but adequate dating is lacking; it is of interest, however, that according "to traditions of the Baganda, iron-working did not begin in their country before A.D. 1000, having been introduced from Bunyoro" (Cole 1964:301). Sutton's Class A, which is fairly localized in the High- lands and Rift areas, includes pottery found with a long sequence of Late Stone Age artefacts (from 800-100 B.C.) and bones (some of domestic cattle) at the Tunnel rock shelter near Fort Ternan. Pottery of a similar appearance found with obsidian implements at the Njoro river cave (Cole 1964:286; Leakey and Leakey 1950) is dated to ca. 1000 B.C. Both sets of finds suggest a mixed tradition of hunting, collecting, and pastoralism. His Class C, com- parable with the Uganda Bigo wares, is probably to be placed within the first half of the 2nd millenium A.D. (cf. Lanet, dated to the end of the 16th century [Posnansky 1962]) and associated with the arrival of further pastoral- ists. There is little evidence of contact with the coast until the 19th century (cf. Freeman-Grenville 1959, McMaster 1963), when links seem to have been stronger with the richer Iron Age culture of Zambia and Rhodesia. Little is known, as yet, of Tanzania apart from Smolla's (1957) description of dimple-based pottery and of what may be an evolution from that pottery type. It would seem that the Late Stone Age hunter/gatherers continued well into the 2nd millenium A.D. and that here also influence from the coast was not felt until the 19th century, with the arrival of trade goods and the movement of Arabs along the routes opened up by the Nyamwezi people of west- central Tanzania in the late 18th century (Oliver and Fage 1962). Agriculture was certainly practised before the arrival of the Arabs, but there is no clue as to its age. Arab trading stations had been established on the coast from as early as the middle of the 1st millennium B.C. (Davidson 1964: 173-75), but settlement became more frequent after the 7th and 8th centuries, and the first stone buildings were constructed around the 11 th century A.D. (Chittick 1965). Contact with the interior may have begun as early as the 7th century A.D., judging by finds at Gokomere, Isamu Pati, and Ingombe Ilede in Rhodesia and Zambia (Fagan 1963b, d, 1965b; Phillipson 1965a) but it was probably fairly sporadic at first. El Mas'udi, writing in the 10th century, describes the Zanj of Waqlimi (who may be among the Iron Age societies of Rhodesia and Zambia) as skilled metal-workers, energetic traders, and elephant-hunters, and notes that, although they culti- vated the banana, the staple of their diet was sorghum and a plant called kalari, which was dug from the ground like a truffle (Mas'udi 1841, quoted in Davidson 1964: 134-37). In order to appreciate the relationship of East Africa to the spread of agriculture, more work is required in the belt that runs south of the Ethiopian highlands from the southern Sudan to the Horn (cf. Arkell 1966:478); for while there are similarities between some of the artefacts of the middle Nile and the Sudan and some of those of East Africa (Fagan 1 965b, Pearce and Posnansky 1963, Sutton 1 964b), there are suggestions of early dates for agriculture and possibily also for iron-working in West Central and West Africa south of the Sahara (Davies 1966: 471; Fagg 1962). Radiocarbon dates suggest that both Vol. 9 ;No. 5 * December 1968 491This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

agriculture and iron-working may have spread into East Africa from the southwest (cf. Oliver 1966) and that affinities between dimple-based ware and channelled (Machili) ware from Zambia demonstrate a common origin (cf. Posnansky 1966). It is probable that there was little contact between the iron-working farmers of Zambia (typified by the Kalomo culture people, who were possibly Bantu from the lower Congo [Fagan 1963d]) and the people of Uganda during the first half of the 2nd millennium A.D., despite certain pottery similarities (cf. Cole 1964:322; Fagan 1961b; Oliver and Fage 1962: 32-33). ZAMBIA AND MALAWI When and from where precisely the first agriculturalists moved into Zambia and Malawi is as yet unknown (cf. Fagan 1966:453-62, 479; Posnansky 1966). The earliest Iron Age sites in Zambia are dated to around the begin- ning of the Christian era (cf. Situmpa Forest Station, Machili, A.D. 96 + 200 [C-829] and Lusu, Seshake District, 186+ 150 B.C.), and the associated pottery has been termed Zambezi channelled ware or Situmpa ware (Clark 1962a; Clark and Fagan 1965; Fagan 1963b, c, d; Inskeep 1962, 1963; Posnansky 1961b; Phillipson 1965a). Channel-decorated pottery is found at other sites (Fagan and van Noten 1964), one of the most important being at the Kalambo Falls, in the extreme north of Zambia (Clark 1962c, 1964a). The site of Kalambo, where the channelled ware tradition persists from ca. A.D. 550 to the 16th century (Fagan 1966a; cf. Clark 1964a:973), appears to have been an extensive village or group of villages. Although the only evidence of agriculture is the presence of grindstones, a few bovid teeth, and some carbonised beans, it seems likely that these people were mixed farmers. Recent work at the site of Dambwa, Livingstone has revealed a settlement with iron slag, a few iron tools, and pottery which suggests some affinity with the Situmpa tradition (Daniels 1965; Phillipson, personal communica- tion and 1965a). The radiocarbon date of A.D. 620+ 110 (SR-62) (Phillipson 1965a) may be evidence of a con- tinuity between the people of Machili and Kalambo and those of the Kalomo culture (in this connection see Inskeep 1962, 1963; Fagan 1967a:31). Hoe agriculture is well attested for the Kalomo people, who were widely spread from the borders of southern Barotseland in the west to the Gwembe valley in the east, and from the Victoria Falls as far north as Choma (Fagan 1963d, 1966). They lived for the most part on artificially built-up mounds whose depth occasionally reaches 10 feet, providing a valuable stratigraphic sequence of pottery and fauna (Fagan 1967a). Radiocarbon estimates from Isamu Pati suggest that the time range of this culture is between the 7th and the 13th centuries A.D. (Fagan 1966a, Phillipson 1965a). Iron-working was practised on a small scale, but the rest of the material culture is poor. Besides hoe agri- culture, the economy included stock-breeding, hunting, and collecting (Inskeep 1962). Kaffir corn (sorghum) was cultivated early with iron hoes and possibly with digging sticks weighted with bored stones (Fagan 1962, 1967a: 63-64); fruit and wild seeds were probably collected (Fagan 1963d, 1966a :457); hunting continued throughout, although cattle and sheep or goats became more common as time went on. A rotation cycle of about 12 years-about half that of the present-day Plateau Tonga (Allan 1965)- has been postulated, on the basis of the middle levels, for the Isamu Pati. It would seem that the economy of these people was basically similar to those of several modern groups (Fagan 1967a). At Sebanzi Hill, Lochinvar (Fagan and Phillipson 1965), occupied from a little before A.D. 1300 to perhaps as late as the 17th century (Phillipson 1965d), numerous grindstones and rubbers suggest cereal cultivation, as does the presence of an iron hoe in level 5. Domestic stock predominated in the meat supply. Fishing provided a significant part of the diet, in contrast to that of the Kalomo people of the Batoka plateau to the south. Pot- tery like that of the middle levels at Sebanzi is found at Kangila (dated to ca. A.D. 1400), where cattle and small stock represent 74.2% of the bones (Fagan 1967a). No settlement pattern could be discerned at this site, and dwellings were probably of a temporary nature (Fagan 1963d) again a contrast with the Kalomo people. The early history of this tradition is little known, though there are suggestions that the pottery from the lower levels of Ingombe Ilede, a site dated to between the 7th and the 9th centuries A.D. (Phillipson 1965a), shares some features with Kangila ware (Fagan 1965b, 1966a). The people of Ingombe Ilede cultivated sorghum and kept cattle, goats, and dogs; they used metal hoes and were elephant- hunters and traders. Evidence of contact with the east coast is seen in the gold burials with pottery vessels and rich trade goods dated to around A.D. 850 and in the imported cloth and Conus seashells (Fagan 1963b, d, 1966a; Phillipson 1965a). Outside the southern province there is as yet little information, and the few sites known suggest different traditions (cf. Clark and Schofield 1942, Daniels 1964, Fagan and Phillipson 1965, Phillipson 1965a, b, c). In Malawi recent work has provided very interesting material (Inskeep 1965, Robinson 1966a). From a site near Mwamasapa village in the Ngonde area at the north- ern end of Lake Malawi, charred sorghum seed and a possible millet seed (Pennisetum typhoides) were found with pottery, glass beads, iron arrowheads, tuyeres, and slag (Robinson 1966a). Robinson suggests possible affinities with material from Hyrax Hill and the Bambata ware of Rhodesia, but these, for the time being, are rather specula- tive. A different pottery tradition found north of the Rufira river may be a late derivative of the Kalambo-type channel-decorated ware Fagan (1 963b:159-60) and may be contemporary with, or earlier than, the material from near Mwamasapa. RHODESIA The earliest evidence of agriculture in Rhodesia is associated with the appearance of stamped and chan- nelled pottery (Robinson 1961 a), the beginning of mining, and the occurrence of female and animal figurines. (Another early pottery tradition, Bambata ware [cf. Bernhard 1961; Schofield 1940; Summers 1966:464], possibly originating in the northeast [cf. Robinson 1965b, 1966c] is frequently associated with Late Stone Age implements, but never with iron. While it could be a pre- iron-working pottery tradition, it is more likely to represent the adoption of certain cultural traits by indigenous hunter/gatherers. The stamped and channelled pottery ainnears to he unrelated to the Bambata ware and. 492 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

contexts, of earlier date [Robinson 1961 b, 1964, 1966b]). Several variants of this tradition are known, partly local and partly overlapping in time. The Gokomere complex, which ranges from ca. A.D. 200 to A.D. 800 (Gardner, Wells, and Schofield 194 1; Summers, Robinson, and Whitty 1961:190; Robinson 1963:169, 1966d:474) includes stock and cultivated cereals (Robinson 1961 a, 1963). The Ziwa variant (Summers 1950, 1958), which lasts from ca. 300 to ca. 1000 A.D. (Summers 1966:464), includes agriculture with some hunting and collecting but probably no cattle (Summers, Robinson, and Whitty 1961). This variant may have entered from the northeast and is probably closely related to the Leopard's Kopje culture. The Leopard's Kopje culture (Robinson 1959, 1965a) has been divided into three phases (Robinson 1966b). Sites of phases I and III usually occur in areas with fertile soil, while phase II sites are found even where the soil is poor. The consequent view that phase II may have placed more emphasis on pastoralism than the others is supported by the presence of numerous figurines of domestic stock at phase II sites. The location of sites on the heavy red soil in phase III may suggest that more efficient agricultural implements were being used (Robin- son 1966b:25). Phase I is a link between the Gokomere tradition and phase II; phase II shows the influence of an unrelated tradition first referred to by Schofield (1948: 101-12) as "Shona"; phase III has many of the character- istics of material from the upper levels of Mapungubwe in the Transvaal. The whole complex seems to have lasted from about A.D. 700 until 1300 (Robinson 1966d: 474). Farther east is a group of cultures associated with terrace cultivation. (Summers [1966:467] has suggested that the terracing at Inyanga, at least, may owe its origin to a late facies of the Ziwa tradition.) Poorly dated, the cultures of this group probably extend from around the 16th and to the 19th century. They seem to represent an entirely agricultural tradition, showing little of the richness of the stone-wall traditions of Zimbabwe and Mapungubwe (Summers 1960:282-83, 1961; Whitty 1959). Several of the rock paintings of Mashonaland may represent domestic animals (Summers 1959:33, 34) and various root plants (pp. 79, 83), while a few may show fields of grain (pp. 65, 107). No dates can yet be given to these works of art, however, and so they remain of marginal significance. SOUTH AFRICA Farmers and iron-workers came into South Africa across the Limpopo, probably some time before A.D. 900 (the earliest dates for the Iron Age in this area come from the eastern Transvaal and range from ca. A.D. 800 to 1000 [van der Merwe, personal communication; cf. Mason and van der Merwe 1964]) and almost certainly came into contact with the Late Stone Age hunters, who survived until the beginning of the 2nd millennium A.D. in certain parts (cf. Mason 1962). These people built large stone- walled villages, mined and worked metal, made pottery, kept stock, and grew cereal crops. The apparently rapid increase in population at this time may be due both to the stable food supply and to gradual infiltration from the north (CElark 1959). Skeletal remains from a number of sites (cf. Fouche 1937; Galloway 1959; Gardner 1955, Seddon: AGRICULTURE IN EAST AND SOUTHERN AFRICA breeding between locals and incomers, which repeats the pattern observed from East Africa (Cole 1964) to the south (Brothwell 1963). At Mapungubwe (A.D. 1050+65, Gardner 1963) and Bambandyanalo (A.D. 1380 + 60 and 1420 + 60, Fagan 1964:355) in the northern Transvaal, there are signs of continuity with cultures in the south of Rhodesia at the end of the 1st millennium A.D. and beginning of the 2nd millennium. Both sites have produced remains of domestic cattle and sheep or goats (Mason 1965). Direct evidence of agriculture is found at Mapungubwe and Klipriviers- berg: from the former site "there is evidence of kaffir corn (one or another strain of Andropogon sorghum) and beans (cow-peas or Vigna sinensis)" (Pole-Evans 1937) and from the latter of probable sorghum (Mason 1968). On the whole, the evidence from the Iron Age in the Transvaal and Orange Free State-the large walled settlements, the grindstones, iron hoes, adzes, and spears found at excavated sites, as well as direct evidence of tex- tile manufacture (Wells 1935)-points to a mixed farming economy with supplementary hunting and collecting (see Mason 1962 for references). This pattern continued up until the 19th century (Campbell 1815, 1822; Kirby 1965; Seddon 1966b). Not a great deal is known, as yet, of early farming communities in other parts of South Africa, although stone walling in the Orange Free State suggests a similar economy (Mason 1962). SOUTHWEST AFRICA Clark (1959, 1964b, followed by Fagan 1965b, Oliver 1966) has suggested that there may be connections between the stone bowls or mortars found in Southwest Africa (Fock 1956, 1960, 1961; MacCalman 1962) and those of East Africa. This hypothesis is supported by finds of similar stone vessels in northeastern Zambia that are apparently associated neither with the Late Stone Age nor with the early Iron Age villages (cf. Fagan 1965b :46). The makers of these bowls may have spread diagonally across southern Central Africa to southern Angola and northern Southwest Africa, taking with them pottery and livestock, and possibly iron-working and agriculture, as early as the first few centuries A.D. (Clark 1964b). Contact with local Stone Age hunters may have taken place this early (cf. MacCalman 1965), although the Late Stone Age lasted in some places until the present day (Mac- Calman and Grobbelaar 1965, Silberbauer 1965). Traditionally the earliest farmers and iron-workers to reach the area were the Ovambo, who arrived, along with the Herero (nomadic pastoralists), around A.D. 1550 (Rudner 1966:474; Vedder 1937) and found the Hotten- tots and the iron-using Bergdama already in occupation (Rudner 1957, Pearce 1960). Actual evidence for the East African-Southwest African contact is still slender, and more work is needed in the intervening regions. METHODS OF AGRICULTURE In a few cases there has been quite detailed analysis of the nature of agriculture at early farming sites (cf. Fagan 1967a:62-87), but for the most part it is necessary to draw heaviAlv on nresent-rlav methods for illumination - Comnari- V ol) 9 . - December 1 ' 5596 Vol. 9 * N%o. 5 * December 1968 493This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

sons are often crudely made, and the use of a system of classification such as has been set out by Conklin (1961) might be profitable. Investigations in Zambia have shown that the channelled-ware makers settled along the scarps overlooking the rivers and the milapo (broad, shallow, grass-covered valleys) in the same way as do the present- day people of the Barotse plain (cf. Clark and Fagan 1965), the forest margins along the milapo being most productive, owing to better soil and greater moisture. A number of different agricultural systems, varying with soil type (Trapnell and Clothier 1957), are in use in Zambia at present, but the staple crops are, for the most part, bull- rush millet, sorghum, and maize, underplanted with vegetables (cf. Colson 1949, 1960; Jaspan 1953; Scud- der 1962). The economies of early farmers in Zambia and in Rhodesia are likely to have been similar to those of the present, except that, having no maize (Miracle 1965), they probably used a method close to the Shanjo "bush- clump" system, where land selection determines the sites (Clark and Fagan 1965). ORIGINS OF FOOD PLANTS In the savannah belt that runs from Mali and Niger to the upper Nile and then southeastwards to northern Uganda and parts of Kenya, native cereals (the millets: Sorghum, Pennisetum, and Eleusine) are found in early farming and iron-working contexts. It is likely that agri- cultural concepts and techniques had reached the savannah area and been applied to the indigenous cereals by the 3rd or 2nd millennium B.C. (Clark 1962b, Oliver 1966, Oliver and Fage 1962, Porteres 1962). South of this zone, most of the staple foods of agricultural peoples today (apart from the millets and native yams) were introduced: the banana and Asiatic yam perhaps between 1,000 and 1,500 years ago and maize, cassava, and sweet potatoes much later from America, during the 16th and 17thf centuries. The Indonesians may have reached Madagascar with bananas and yams as early as the first three centuries A.D (Christie 1957; Gray 1962:182), but by that time it appear. that there were already early farmers in Zambia and pos. sibly Rhodesia. The spread of food-producers and iron- workers seems to be closely associated with the Bantu expansion from West Central Africa (Oliver 1966, Posnansky 1961 c), not from the east coast opposit( Madagascar as one might have expected had the intro. duction of the Asiatic yam and the banana provided as great an impetus for population movements as has beer suggested (cf. Gray 1962:182, Posnansky 1961c). Jones (1959) has suggested an early circumnavigation of th( Cape by the Indonesians, but evidence for the earl) appearance of the banana in West Africa is not good, and of the 13 types of yam listed by Dalziel (1937), only twc are certainly Asiatic and both of these are probably late (Miege 1952; Morgan 1962:236). It is more likely that the idea of cultivation spread from the savannah country and woodlands in West Africa, where local yams may have been planted in association with cereals (cf. Fagan 1965b: 44; Morgan 1962:139), and that bananas, though intro- duced via the east coast at an early date (cf. Davidson 1964:134-37; Posnansky 1961c; Simmonds 1959), had little effect on the spread of farming into Southern Africa. That the movement of peoples through East Central and Southern Africa was associated with iron-working is strongly supported by the excavated material. Although metal seems to have been used more for hunting equip- ment than for agricultural implements (cf. Fagan 1966a: 457), it is likely that the acquisition and development of both farming and iron-working techniques were largely responsible for the spread from West Central Africa of the early agriculturalists. Comments by HOMER ASCHMANN* Riverside, Calif., U.S.A. 17 v 68 To a non-Africanist, the three related papers presented here seem to indicate that studies of African prehistory have reached a stage of maximum excitement and at the same time maximum confusion. The data of archaeology are becoming too abundant to permit uninhibited theorizing but are still so spotty that absence of evidence for agriculture of significant antiquity in a given region need not be accorded powerful importance. In West Africa, at least, there seems to be an almost perfect inverse correlation be- tween regional precipitation and antiquity of the evidence for agriculture. That this can be explained in terms of prob- abilities of preservation is not a sufficient response. Ethnographic and botanical and ecological evidence can and must be given greater weight. That many quite recent archaeological sites in eastern and southern Africa show little or no evidence of agricultural activity should come as no surprise. Ethnographically known tribes in the region have supported themselves almost entirely by other activities, and almost exclusively pastoralist or farming tribes are well known. Parallel patterns can be presumed to have existed over a long period of time. The cultural or en- vironmental cause for this economic specialization along tribal lines constitutes a fascinating problem, but one only vaguely related to the origin of agriculture. Of more significance is the fact that the agricultural complexes of Africa south of the Sahara are singularly different from those of the Middle East, both in tech- niques of cultivation and in crop assem- blages. Although various domestic animals were kept, they were not used for plough- ing except in Ethiopia and thus were not a crucial part of the cropping complex. The annual plants cultivated, in complete contrast to those of the Middle East, were adapted to growth in a season of high temperature and high humidity. This is true of the sorghums and millets of the southern edge of the Sahara as well as the root crops of wet forest areas. Except in cool and relatively dry highlands, the Middle Eastern cereals fail completely because of their inability to withstand the pathogens that flourish with high tempera- ture and humidity. A new crop assemblage had to be developed, or introduced from areas such as tropical America or South Asia with similar climatic patterns. When and where such domestication or intro- duction first took place seems to remain an unanswered question, but better answers would certainly form a basis for under- standing the complex diffusions of ideas and migrations of peoples that constitute Africa's long history. I find Hugot's use of climatic change in the Sahara Desert as a basis for identifying that region as a center of agricultural origins quite unsatisfying. That there have been periods wetter than the present is most probable, but the extant flora and the landforms and drainage patterns suggest that perhaps precipitation was doubled or at most tripled. Since present annual rainfall values are well below 50 mm. such an increase would leave the region a full desert, with a few more oases and better Dasture at favored sDots in its 494 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies, Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM interior and a retreat of its northern and southern margins. It was still a barrier and a refuge, though perhaps a less difficult and more populous one than today. Only the date palm is peculiarly adapted to it, and Southwest Asia seems to have a stronger claim as the original center of its domestication. by JOHN BLACKING* j_ohannesburg, South Africa. 2 VI 68 I find no complaints with the way the authors have gone about their summaries, except that in general I object to the use of the term agriculture for what seems to be in most cases horticulture. It may seem old- fashioned, but I think that some verbal distinction between these very different processes of growing crops should be maintained. by B. BRENTJES* Halle/Saale, Germany. 8 v 68 The idea of asking three scholars to produce a summary of African agricultural develop- ment was a good one. All three reports demonstrate, however, that our knowledge is still too sketchy to be conclusive. The archeology of Africa, which has only begun to develop in recent years, permits an acceptable approximation of a summary only for partial geographical areas (par- ticularly for Rhodesia). On the other hand, some problems are not even dis- cussed in these articles. For example, Seddon fails to discuss the Punt problem. The basis for the Egyptian Punt expedi- tions of the 2nd millennium B.C. must have been trade between an agricultural people in Egypt and either a non-nomadic hunting people or an impoverished population of fishermen in Ethiopia or Somalia. An analysis of the Egyptian Punt descriptions and a subsequent search for the archeological traces of this civiliza- tion-which was sustained, at least in the period of the New Empire, by black Africans-would be most revealing. Davies leaves out the problem of the preliminary stages of the Nok culture of Nigeria, already in its Iron Age (from the 1 st millennium B.C.). The primacy of the Near East in agricultural history should probably receive more emphasis than it has hereto- fore been given by Africanists. One of the things that the Near East teaches is the close relationship of animal-breeding and field cultivation. The semi-nomadic culture still frequently found in the Sudan (nomadic cattle-raising with seasonal settlement for grain-growing) might be considered as the foundation for the older herding pattern of the Sahara; but this complex is missing in the reports of all three contributors. Seddon's assumption that the concen- tration of population clusters resulting from the climatic effects of the receding Ice Age is a precondition for agriculture finds no support in Near Eastern data; likewise, other sources offer no archeo- logical evidence whatsoever that planted agriculture preceded seed agriculture. In North Africa, the Sahara highlands were undoubtedly inhabited by herders of sheep and cattle, who probably also raised grain, previous to the settlement of the Nile Valley. Similarly, in the Near East, the mountainous country was settled by tribes of cattle-breeders from the 10th millennium B.C., whereas in the Meso- potamian valley cattle-breeding did not appear until the 6th millennium and agriculture until the 8th at the earliest. The date of 3800-3600 B.C. for Negade I is surely too early; it should probably be 3400-3100 B.C., for the Negade II culture contains indications of Jemdet-Nasr seals, dated at 2800-2700 B.C. The allegedly small yield of the wild grain harvest mentioned by Davies has been refuted by Harlan (1967), who collected 1,400-2,250 grams per hour with his bare hands in Turkish wild-grain fields and increased the yield to 2,450 grams with the aid of a sickle. Considering the extent of modern wild-grain fields in the Near East, the harvesting of wild-grain over many centuries was in all likelihood a stage of transition to agriculture, along with the breeding of sheep and goats, which developed in the 10th to 9th millennium B.C. Davies' interpretation of the Tumbian pickaxes is supported by the discovery of wedge-shaped handaxes (Faustkeile), ob- viously used as picks, in the Ghassul culture (4th millennium) of Palestine (Koeppel, Mallon, and Neuville 1934-40). Stone picks can be traced in the Near East 1,000 years before Fayum and over 2,000 years before Shaheinab, in Tell Hassuna I A (ca. 5800 B.C.) (Lloyd and Safar 1945). The oldest certain description of durra comes from a Sherkar rock picture at Jebel Queili, east of Khartoum (Brentjes 1965). Hugot has overlooked the influence of the growth of grass in nomadic cattle- raising, which undoubtedly made a sig- nificant contribution to North Africa's becoming a desert. The oldest known sickle comes from Zawi Chemi Shanidar (10th millennium) in northern Iraq. The oldest evidence of agriculture in Egypt comes from Fayum A and must be dated about 4600 B.C. (Ehrich 1965). The history of the Sahara barley is surely not quite as simple as Hugot assumes. Wild barley, the only wild grain extending as far as Libya, could easily have been a very early (pre- Egyptian) basis for early agrarian cultures (Harlan and Zohary 1966). It is only in this respect, however, that the Sahara can be considered a "cradle of cultivated plants"; agriculture certainly came to Africa from the Near East. The Sahara can be excluded with still greater certainty as a center for the domestication of animals. Domestication is millennia older in the Near East than in Africa; at present the oldest evidence (domesticated sheep) is in Zawi Cherni Shanidar, dated at 9200 B.C. (Solecki 1964). by KARL W. BUTZER* Chicago, Ill., U.S.A. 20 v 68 These articles are timely, and should provide both stimulus for discussion and guidelines for on-going studies. Fortun- ately or unfortunately, the tempo of current investigation is such that it is impossible to keep in touch with the results of all unpublished excavations, and details are sometimes rendered obsolete by publications in press at the time of writing. So, for example, it is regrettable that Desmond Clark's (1967) reviewof Neolithic culture in sub-Saharan Africa was in- accessible to Seddon and Davies. Northeastern Africa remains crucial for the question of earliest agriculture on the continent. McBurney's (1967: Chap. 9) final report on Haua Fteah provides a good record of the earliest Neolithic in Cyrenaica, beginning ca. 5000 + 250 B.C. This culture is clearly intrusive there, and Mori's (1965:221-24, 241-42) C14 date of 5590 + 220 B.C. for the early "Lower Hearth Complex"- apparently with domesticated cattle and sheep-at Uan Muhaggiag, southwestern Libya, may ultimately support McBurney's suggestion that agricultural traits were derived from the Tunisian sector. Full publication of the results of J. Shiner's and F. Wendorf's Shamarkian sites in Nubia is due at any time, and should provide new chrono- logical and cultural data for the gap between the Fayum "A" and Shaheinab. Similarly, the puzzling implications of specialized, late Paleolithic riverine cult- ures in southern Egypt (Butzer and Hansen 1968, Chap. 4; Hobler and Hester 1969; and the reports in preparation by F. Wendorf and others) still remain to be evaluated in a more general perspective. Taickholm's (1941-54, vol. 3:529-44) case for Ensete edule in Predynastic Egypt is a poor one, being limited to representa- tions of a highly stylized plant on pottery. Fibre once suspected to pertain to Ensete edule has proved to belong to another species or even to an unknown genus. J. R. Harlan (personal communication) is unconvinced that the Predynastic decorated ware refers to Ensete, and I have considerable doubts, for both climatic and Vol. 9 - No. 5 * December 1968 495This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

of the Ethiopian Rift country could b successfully grown in the lower Nil Valley. by BRIAN M. FAGAN* Santa Barbara, Calif., U.S.A. 17 v 6t The literature on the origins of fooc production in sub-Saharan Africa is grow ing rapidly. Unfortunately, the pace o synthesis far exceeds that of systemati4 field research, except, perhaps, in th( Saharan regions, where much valuabl field research remains unpublished. In th( interests of space I will confine myself to < few comments on Seddon's paper, whic] covers ground with which I have first hand experience. It is clear that the paper was submittee to CA some two years ago, for there havy been many developments in the field sinc( it was written. A large number of nev radiocarbon dates have been released fo: Late Stone Age and Iron Age sites in east ern and southern Africa which modify sub stantially the picture given in this pape: (Fagan 1967b, 1968b). The latest radio carbon dates do not suggest that agricul ture and iron-working spread from th( south into East Africa. Dimple-based war( sites such as Urewe and Kwale have beer dated to the 3rd to 4th centuries A.D (Fagan 1967b, 1968b), while Phillipson'" new dates from Zambian Early Iron Age sites suggest that iron was establishec there at much the same time (Phillipsor 1968). The Machili date of A.D. 96+219 (C-829), so frequently cited in the literature as evidence for iron-working on the banks of the Zambezi in the 1st century A.D.. appears somewhat earlier than the othe dates, although its standard error is verb large. Iron-working also appeared ir Rhodesia at roughly the same time (Sum- mers 1967). Everything points to a rapid introduction of iron-working, and perhaps agriculture, into much of eastern and southern Africa at the beginning of the Christian era. This event may have been connected with the expansion of Bantu- speaking peoples from a nuclear area to the south of the Congo forest, but we lack archaeological evidence for early iron- working in the Katanga area. The excava- tions at Sanga revealed advanced metal- lurgy in the region by the 9th century (Nenquin 1963b); it is logical to suppose that iron-working has a long history in Katanga. The statement that "Dimple-based ware" sites tend to be concentrated in areas where agriculture could be supplemented by fishing and hunting is open to question. There is little evidence for this, nor have we any valid economic data from "Dimple-based ware" sites to support such statements. Iron Age economies were presumably adapted to the environments in which they were practised, leading to a quite natural concentration on hunting or fishing in some areas if it was of advantage. So few Early Iron Age sites are known from East Africa that statements of this nature are most unwise. Quite a few of those which have been found happen to be sited near water; the association may be coincidental. Contact with the interior of South Central Africa, as evidenced by finds of sea-shells and imported glass beads, begins very much earlier than the 7th century adduced by Seddon. Sea-shells at the base of the Gundu mound near Batoka in southern Zambia have been dated to the 5th century (Fagan 1 968a), and a cowrie shell found in the basal levels of the Kalundu mound is associated with radio- carbon dates in the 4th or 5th centuries (Fagan 1967a). There are imported beads at the Mabveni settlement in Rhodesia-a site apparently missed by Seddon-which may be as early as the 3rd century (Robinson 1961 c; Fagan 1966b). In the early days of Iron Age research in Zambia, we tended to lump components together. We are now splitting, and have divided the "Kalomo culture" of the Kalomo mounds (Fagan 1967a) into at least three parts (Fagan 1968b): (1) an Early Iron Age component dating to the early 1st millennium; (2) the so-called Kalomo culture, in the middle levels; and (3) in the upper horizons, after about the 12th century A.D., a possible Tonga component characterized by globular pots. Seddon's statement that there was probably little contact between the iron- working farmers of Zambia and the people of Uganda during the first half of the 2nd millennium seems almost platitudinous. I think that we must think of the African Iron Age in terms of a series of regional sequences of culture, which share many broad points of similarity but differ greatly in detail one with another. Each Iron Age society was an adaptation to its own particular environment; to trace con- nections over large areas of Africa on the basis of the highly inadequate data we have at the present time is premature. I end this comment with another vigor- ous plea for further fieldwork and study of local sequences. The basis of our knowledge of ancient African agriculture will always be the regional sequence of prehistoric culture, carefully and meticu- lously studied. It is time that we all forgot synthesis for a while and went back into the field. We are running out of new ideas, a sure sign that more data are needed. by H. GAMs* Innsbruck, Austria. 24 v 68 A lot of information on cultivated plants is available in a number of regional floras, among them the unfinished Flora of Egy:pt of Taickholm and Drar (1941-54), the Flore du Sahara of Ozenda ( 1958), the N%ouvelle Flore de l'Algerie of Quezel and Santa (1962), and Quezel's (1965) work on the Sahara, Tchad, and Mauritania, Zhukovsky (1968), continuing the work of Vavilov on centers of variability and origin of cultivated plants, has located 14 "microcenters" for African wild plants closely related to cultivated ones-three in the North, five in the East, and five in the South. Rapid progress is being made in palynology (see, for example, Van Campo 1958-65, 1964; Van Campo and Coque 1960; van Zinderen Bakker 1951, 1963, 1965; van Zinderen Bakker and Goetze 1951-53; Kuprianova 1959; Leroi-Gour- han 1957; Miege and Assemien 1964; Assemien 1965). Most reports up to this point, however, fail to distinguish the pollen of cultivated plants, especially cereals. Firbas (1937) has shown that the pollen of cereals can be distinguished by size, and Erdtman (1944) has distinguished the pollen of Avena, Hordeum, Secale, and Triticum by the structure of the pores. More palynological investigations making such distinctions are needed. by FREDERICK C. GAMST* Houston, Tex., U.S.A. 11 v 68 These three papers consolidate most of the information on the origins of plant cultivation in sub-Mediterranean Africa, but they do not treat the broader evolu- tion of food production; that is, the inter- relationships of plant and animal domesti- cation. We should be certain of our interpreta- tions by analogy of the functions of arti- facts. Although analogy based upon ethnographic data is essential to archae- ology (Chang 1967:229-30), it can be vitalized to the point of non-validity. Hugot notes that although bored stones (kwe) may have other purposes, they were certainly used as weights on digging sticks. Davies remarks that such tools may have been used in mining. In highland north- western Ethiopia, a kwe-like stone is used today as a pivot-base for a pointed vertical pole supporting the door of a house. What the other, possibly innumerable, functions of kwe-like lithic forms might have been over the millennia we can only guess. Davies' admonitions concern- ing the idea of "stone hoes" may not be strong enough. These artifacts alone cannot be used as indications of past cultivation, even if we could observe a few good ethnographic analogies. Davies has cited many of the valuable contributions to problems of plant domesti- cation in western Africa, but he has omitted the work of Murdock (1959:64- 70). In a survey article, rejection of fundamental ideas such as those of Mur- dock should be made explicit by citation and scholarly disagreement (cf. Baker 1962; Morgan 1962). Murdock's thoughts on Malaysian crops and migrations of Bantu-speakers are refuted, not too 496 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies, Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM cogently, by Seddon, although he does cite Merrick Posnansky's valuable contribu- tion (1961 c) to this matter. Are there any evidences, other than pottery, of early "farmers" in Zambia and Rhodesia during the first three centuries A.D., and is the spread of these early food-producers associated with the expansion of speakers of Bantu ? Davies notes that African tubers do not require intensive weeding and are not cultivated as a monoculture. We might add that where primitive cultivating techniques are presently used, and un- doubtedly when they were used in the past, horticultural plots are usually not of the "clean-crop" pattern (Anderson 1960: 72; Simoons 1965b). Domesticates and encouraged and permitted weeds of various kinds and sundry uses grow within the plots in variegated patterns, but almost never in straight rows of a single species. Plants such as generically and specifically different cereals and weeds may be sown, harvested, and consumed as mixtures. Since the time the authors wrote of Tumbian "culture" and a continuity of "cultures" based upon distribution of artifacts, it has been recommended (Clark et al. 1966) that we replace the term culture with industrial complex (e.g., Acheu- lean) or industry (e.g., Kenya Capsian) in the interest of greater clarity and precision. We might note that culture should not be considered a tool-kit, but a learned pattern of behavioral rules transmitted from one generation to the next. Widespread distributions of artifacts with similar morphologies, and perhaps similar functions, may indeed be the result of equally widespread migrations of people who made the objects, as Seddon believes of the "stone bowls" of Southwest Africa. Certain artifacts and techniques, however, may be found within different ethnic groups and be widely dispersed because of their diffusion or because of the stimulus diffusion (Kroeber 1940:1-2) of the idea underlying them. Ethiopia, an area of interest to me, has long been recognized as a center of diversity of crops such as wheat and barley (Braun 1877: 9; Solms-Laubach 1899: 29) and wvas thought to be an independent center of agricultural origins by Vavilov (1931: 6; 1949-50:37-39). Actually, Ethiopia is a derived or secondary center of cultivation (perhaps beginning around 3000 B.C.) and of great importance in the culture history of the Old World (Murdock 1959:181, 183; Simoons 1965a:9-10). The varied and archaic Ethiopian crops result from diverse ecological niches (Burkill 1953: 15- 17, 20-21) and continued ancient patterns of cultivation (Anderson 1960:76). Con- trary to Seddon's assertion, Murdock does not consider Ethiopia an independent center of crop domestication ( 1959: 1 81 , 1833), but should be cited as one of the earlier believers in a connection between western African cultivation and that of southern Ethiopia (1959:67, 181). We find no evidences of hoe cultivators or cultivation in the western parts of Ethiopia in the primary source on this matter (Leakey 1943: 193-94). Celts found here may have been used in mining, according to Mary Leakey. On the work of Pere Azais, Bailloud (1959:23-24) does mention stone-hoe cultivators in southern Ethiopia, but the evidence is not convincing. Given the very great diversity of crops noted by Vavilov (1931, 1949-50) and his co-workers (Govorov 1929-30; Ivanov 1929-30), cultivation in Ethiopia of wheat, barley, and indigenous teff probably predates movement of Saharan pastoralists (Clark 1962b: 216) into this region at Agordat (Arkell 1954; 1961:48-52, 78- 79). The first cultivators in north-central Ethiopia were undoubtedly the ancestors of the Agaw peoples (Stiehler 1957:274; Murdock 1959:181-83; Simoons 1965a: 11) whose remnant-enclaves are distributed throughout the region; however, more information on the southern Ethiopian planters is needed. by G. P. GRIGORIEV* Leningrad, U.S.S.R. 8 v 68 Despite the scarcity of data about the beginnings of agriculture in Africa, this summary of it will be very instructive for all prehistorians. On the basis of the facts from Africa, for example, one might suggest for prehistoric Europe the pos- sibility of overlap of indigenous and very primitive cultivation of some plants and the much more extensive agriculture introduced by invaders. Further, it is clear that the beginning of agriculture is the product of a certain level of sociocultural integration and that it also requires the presence of wild predecessors for culti- vated plants. It is evident also, that it is dangerous to draw conclusions about the sociocultural level of a society from its tool-kit alone, because in the Near East agriculture begins with geometric micro- liths, while in Africa (excluding the Maghreb) it comes after them. by ROLF HERZOG* Freiburg, Germany. 5 vI 68 Considering the origins of agriculture in the Sahara, Hugot accepts the assumption that early Neolithic men in that now- deserted part of the world got their knowledge from the East: The Nile valley was an immense nursery- garden, a hinge-line whence cultivated plants from the nuclear area of the Near East were, after adaptation, transported west and south. In attempting to answer this very difficult question, he should have made more use of the results of recent research by Egypt- ologists and prehistorians specializing in the Nile Valley. The publications of Loret and Hartmann are scarcely sufficient. The contribution by Schweinfurth cited is not the most comprehensive in this respect; his 1891 paper seems more adequate. Another highly relevant book that has been omitted is that of Ludwig Keimer (1924), one of Schweinfurth's last assist- ants. Missing also are Tackholm and Drar's (1941-54) Flora of Egypt and quite a number of other recent contributions. The doum palm is not only character- istic of the southern Sahara; it also grows in Upper Egypt and the Eastern Sudan. Its fruit has been found in Predynastic tombs, and the tree is represented in many of the pictures of the ancient Egyptians. Does Hugot consider the date palm one of the gifts of Mesopotamia? Wallert (1962) has presented some good argu- ments against that theory. In the absence of any evidence at all for any cultural or economic import from the Euphrates- Tigris region to the Sahara via Egypt, I cannot accept such a theoretical survival of the pan-Babylonianism which once animated European scholars. by JAMES J. HESTER* Boulder, Colo., U.S.A. 3 v 68 Philip Hobler and I have just reported on findings in the southwestern desert of Egypt near Dungal Oasis which bear on the question of agricultural origins in the Sahara (Hobler and Hester 1969). Since these discoveries have led to the develop- ment of ecological hypotheses not dis- cussed by Hugot in the preceding article, it seems appropriate to review this new evidence and the hypotheses that have been developed. Around the margins of Dungul Playa (an internal drainage basin approximately 10 x 30 miles in extent) were found numerous camp sites which feature a blade industry, sickle flints, and a large quantity of grinding stones. A new cultural designa- tion, the Libyan culture, has been assigned to these manifestations. Although no preserved remains of cultivated plants were recovered, we inferred from the nature of the artifacts and the location of the sites adjacent to the playa that the inhabitants were probably harvesting seed crops and were therefore in a stage of incipient food production. A radiocarbon date from a hearth of 5950 B.C. + 150 years (WSU-316) is nearly 2,000 years earlier than any demonstrated agricultural hori- zon in Egypt. These facts have led to a consideration of the possibility of an independent development of agriculture within the Vol. 9 N /N'o. 5 * December 1968 A 3 497This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

30 20 150 100 350 50 00 50 100 15 200 25? 300 350 40? 25 0- 500M.l\ | I 2 500- 1000 MA V 20 .----. t I00M 0 2 5' *0 +X M. AT A LA : WSX S -13? 10? 5? 0O 50 10 1O I 5O00 25? 300 35 0 FIG. 1. Map of the Sahara emphasizing the topographic highlands time, this zone has functioned as a migration route for plants, savanna zone to the south. from southwestern Asia. 1$vidence support- ing the domestication of certain plants in the central Sahara is of two types: climatic and topographical. The evidence for a wet phase in the time period 6000- 3500 B.C. iS substantial (Butzerl964: 449- 53) and need not be repeated here. The topographical factors have heretofore been neglected and therefore are reviewed in some detail. The central Sahara features a large semicircular plateau surmounted by mountain ranges rising to over 3,000 m. This plateau extends from Darfur in western Sudan through the Ennedi range; an extension runs north to Gebel Oweinat and the Gilf Kebir, while the main portion trends west-northwest to the Tibesti and Hoggar ranges. The latter is then joined to the subtropical savanna zone to the south through the Air ou Azbine highlands (Fig. 1). We believe this crescentic area, adjoining the savanna belt on both ends, has functioned through time as a zone characterized by migration of plants, animals, and man during times of climatic fluctuation. We believe a knowledge of the nature of this "Saharan Fertile Crescent" is crucial to an under- standing of the prehistory and paleo- ecology of the Sahara. During wetter intervals, there would have been a north- ward advance of the savanna-type flora and fauna. During drier intervals, the borders of this zone would have retreated southward with simultaneous isolation of remnants of the savanna biota on the higher mountain ranges. The factors mentioned above, the evidence of seed harvesting at 5950 B.C., the contemporaneous wetter climatic interval, and the opportunity for the higher elevations of the central Sahara to have been populated by a savanna-zone biota all strengthen the possibility of the independent domestication of plants in the Sahara. Evidence supporting such a view is cited byHugot,whostates that atMeniet, in the Hoggar, pollen from cultivated grasses has been recovered from stratified deposits. However, these findings are of later date (3450 + 150 B.C.) than the Libyan culture material from Dungul Plaza. Therefore when Hugot mentions the "Neolithic of Sudanic Tradition," he looks to the Nile Valley for the source of this agricultural tradition, as there is ample evidence of the cultivation of wheat at this time in Pharaonic Egypt. If, on the other hand, the Dungul evidence pertains to incipient cultivation some 2,500 years earlier, then the independent domestica- tion of local grasses such as sorghum and millet in the Sahara seems possible. Once developed, however, such agriculture would have been short-lived, as increasing aridity would have favored reliance on a nomadic pastoral economy within two or three thousand years. by ROBERT A. KENNEDY* Haverfordwest, Wales. 7 v 68 Davies' knowledge of the archaeology of West Africa is far broader and more profound than is suggested by this and his other writings together, particularly as he is essentially a field man. One now, therefore, looks forward to the production during his "retirement" of his greatest work. For my part, I am deeply grateful for his kind help on several occasions, and I offer my criticisms of this paper with respect and caution. Having said this, however, I must sail in with an immediate broadside. To say that "stone tools variously described" as-in particular-gouges are "likely to have been hoes" is sheer dogmatic nonsense. The gouge of Shaheinab, the Fayum, and also Merimde (Baumgartel, personal communication, 1964) and the Tenere is too finely worked and frequently re-worked at the special hollow cutting edge to be anything else than the wood- working adze which it almost certainly was. Similarly, Davies passes a superficial judgment on the "waisted axes and axe- hammers" which virtually amounts to misrepresentation of those he quotes, in- cluding myself. Having made an exhaustive study of these tools from throughout Africa, I find his remarks highly mislead- ing. In West Africa, however, most, but by no means all, of these "waisted" tools were probably hoes. Finally, it seems rash to assume that "explorers of the 14th and 15th centuries knew nothing of Arab navigation in the Atlantic." I doubt this; it seems pretty clear, for instance, that Columbus had a shrewd idea where he was going. Before dogmatizing in this way we should recall the long refusal of Europeans to accept Arab geographers' accounts of the African interior during the 19th century. It would be more profitable to study the navigational aspects. Besides, maize could perhaps have come west-about through arly Indonesian contacts. It may be ;'safer" to assume that maize was brought 498 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies. Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM to Africa by the European "navigators o: the 16th century," but I think it safer still to keep an open mind on this question. I am very glad indeed to see the poini made early in Hugot's paper that th( "great desert was formerly rather a bonc of union" than a barrier to prehistoric man. We need the constant reminder thai modern fences or frontiers did not exisi in Africa in ancient times. Geography is the key. Regarding the so-called Neolithic, foi Africa and the Sahara in particular I prefer to continue using the traditional definition, because we still need the tern in our discussions and study and we all know what is approximately meant by it. New methods for obtaining evidence o: actual agriculture need not interfere with useful nomenclature. Indeed, Hugot shows how their results tend to justify the use o: the term Neolithic. In any case, I am sure he would agree that too few sites have beer properly studied in the vast region undei review; we can only draw a general picture of the cultural phases, and this mainly for the purpose of guiding future research. The Neolithic communities or "villages' certainly tend to occur round ancient lakes, In Tibesti we found a significant distribu- tion of sites along the terraced margins of fossil wadis which may help us to date both (Oxford University Sahara Expedition 1965, unpublished). Hugot's most im- portant statement is that "agriculture is a phenomenon only of the Early Neolithic." I should like to emphasise the Khartoum Neolithic (my Khartoum Neolithic I and II; see Kennedy 1965a') origin or Tra- dition of this Early Neolithic. Hugot has given us a most concise paper of the high standard one has learnt tc expect from him. I am in accord with almost all he says. It is gratifying to see that his conclusions fit or support most of my own. In particular, also, I must endorse his final plea to "extend and systematize research" in the Sahara by re- iterating my own plea (1965b) for an international research team whose main resources in funds and equipment, e.g., vehicles, are not repeatedly wasted by dispersal after each expedition. For me, the strength of Seddon's paper lies in his statements that "no convincing evidence from excavated sites" exists to indicate that agriculture came to East and southern Africa "before outside influence was felt," and that "a knowledge of iron- working may have assisted the Bantu in their expansion," possibly from the "savannah regions just south of the Sahara." Apart from these crucial words I find the paper largely conjectural and yet another which emphasises the gaps in our knowledge. Why is it that nearly all workers seem to ignore the chief gap in our knowledge, that vast and vital region centred on the southwestern Sudan and Tchad? My own work has convinced me that the origin of the Bantu must be seriously considered as being associated with the development of the African Iron Age south of the Sahara. There can be no doubt, moreover, that the Iron Age emerges out of the Late Neolithic in the southeastern Sahara. The much-vaunted "Bantu ex- pansion" almost certainly started much further north than "West Central Africa"; incidentally, this seems to be further supported by the work of Roberts (personal communication, 1962). The phrase "Bantu expansion" has suited those historians who lack an understanding and respect for archaeology, while its glib convenience lends itself to repetition by others. It has simply been adopted to avoid admitting our ignorance and neglect of the vital region referred to above. If Seddon looks to the north and avoids the historians' bandwagon, he will find, no doubt, as I have (Kennedy 1965a) that the earliest Iron Age pottery in southern Africa is closely similar to the transitional Neolithic- Iron Age pottery from southern Ennedi,... [while pottery] of the early Iron Age in West Africa can also be related both to the pottery from the southeastern Sahara and from southern Africa by a common repertory of decorative treatments and motifs. by JON MULLER* Carbondale, ill., U.S.A. 7 v 68 CURRENT ANTHROPOLOGY is to be con- gratulated on bringing these papers to the attention of a wider audience than might be reached by the various regional journals concerned with Africa. The papers show, however, how little progress has been made toward a solution of the basic problems relating to the origins of food production in Africa since earlier summaries such as Clark's "The Spread of Food Production in Sub-Saharan Africa" (1962). They also show just how little we know about this very large area. In West Africa particularly surveys and excavations are needed to provide the data for more complete regional areal syntheses. As the authors have pointed out, the tools used for early agriculture can prob- ably never be separated archaeologically from those used in intensive collecting. Thus, to a very large degree, the final proof of the presence of agriculture must rest on botanical evidence. To recover this proof should not prove so difficult as was once thought. Modern techniques in palynology and the use of sorting tech- niques such as flotation (Streuver 1962) as well as the careful examination of sherds and daub for impressions of grains can give data which a few years ago would have been lost. In addition, careful use of settlement pattern should yield useful hypotheses in some areas. Of these papers, perhaps the most important for the question of ultimate origins of cultivation in Africa is Hugot's paper on the Sahara. Certainly the data presented here as well as those in other compendia (e.g., Erroux 1961 [ ?]) indicate that Near Eastern influence cannot be discounted. Nevertheless, the very crucial question of the origins of the millets remains to be worked out. Until better evidence is forthcoming, the mere presence of wheat and barley cultivation in the northern Sahara in the 4th millennium does not rule out separate domestication of sorghum and millets (though it does make this seem increasingly unlikely). Davies' discussion of West Africa south Df the Sahara is interesting, but disap- pointingly short. Although the literature on the region is by no means extensive (see Hill and Handler n.d., Munson 1966), there is a real need for a critical evaluation of all of this material. Hugot's and Davies' papers further serve to illustrate that it is important in the study of West Africa not Lo define boundaries which too closely reflect present environments and political krontiers. Tools such as digging sticks and fire may ndeed be prerequisites for root-crop igriculture; but, as Davies himself has jointed out elsewhere (1962:291), these -annot serve as absolute indicators of such zultivation. It is of interest that Davies has ~xtended the root crop hypothesis to West Xfrica (1962 and here), but it is surprising what there is no reference to Sauer's (1952: 3-35) speculations on this matter. In general, Davies' discussion seems to oe a "forest" view of the problem. There .ave been a number of attempts to listinguish between the general type of igriculture practiced in the forest and the igricultural systems of the savannah (e.g., Porteres 1962), and it is still uncertain vhether the root crop cultigens have any ^eal time depth. At least one survey based )n physical and archaeological evidence ias placed the expansion into the forest it a very late date (Livingstone 1958). Davies does not discuss the problems -elating to the origins of the savannah -omplex raised independently by Porteres 1950, 1962) and Murdock (1959). The zasic questions which remain to swered are (I1 ) Are forest agriculture and 1 My thesis also contains a pioneer corpus of all Neolithic material from the African continent. The thesis cannot be consulted without my express permission. I am working on its publication in the form of many papers and, condensed, a single-volume book. In the meantime I am always happy to discuss and exchange with colleagues anywhere. Vol. 9 * No. 5 - December 1968 499This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

how? (2) Which is earlier? (3) Is eithe independent of Near Eastern or Asia] influences ? by WALTHER F. E. RESCH* Frankfurt, Germany. 5 v 6, The difficulty in all considerations of th, origin of agriculture in Africa is the presen unsatisfactory knowledge of Africa's pre history. We do not yet have sufficien archaeological data for an exact chronol ogy. Therefore I think it premature t( postulate implications, especially sinci implications tend to take us away fron the facts. I agree, however, that it is per missible, even desirable, to put forth well founded hypotheses (cf. Wissmann 1957) We need to examine the data regardinf (1) a possible primary centre of develop ment of agriculture in Africa and (2 the beginning of pastoral nomadism Archaeological material from the SaharC indicates cattle-breeding (cf. rock pictures and the use of plants as a food resource. Al Seddon correctly says, the first direc evidence of the cultivation of plants i found in Neolithic Egypt. He probabl) refers to the documentation of the use oi cereals by Scharff (1950) and Bruntor (1937): at Merimde, in Lower Egypt mostly Hordeum distichum and Triticun dicoccum, and at Badari, in Upper Egypt Hordeum polystichum. Another occurrence of cereal (Eragrostis abyssinica, i.e., teff) was found in predynastic Egypt (Werth 1954) Schiemann (1943) believes it possible thal Hordeum polystichum and Eragrostis abysinica could have descended from wild forms growing in northeastern Africa. Hordeum distichum, on the other hand, is doubtless oi Asiatic origin (Helbaek 1953). In this connection, it is interesting that the Neo- lithic of Lower and of Upper Egypt show very few affinities (Resch 1967). At the time when Neolithic cattle- breeders migrated into the area of the present Sahara, they already must have had the knowledge of agriculture. This does not exclude the possibility, of course, that they also domesticated other endemic plants. We may conclude therefore, that until we have more data (from pollen analysis, etc.) the question of the origins of agriculture in Africa cannot be answered. We have to take into consideration that wherever Neolithic cattle-herders lived, a knowledge of agriculture may have been present, simply because all the pastoral cultures of northern Africa originally were part of a complex Neolithic farming and cattle-keeping culture (Jensen 1959). Whether agriculture and herding were imported from Asia or developed inde- pendently in Africa is quite a different matter. One approach to this problem is by way of a more thorough examination of the relatively poor archaeological data and stepped-up systematic archaeological research in northern Africa, and re- examination of all our present theories of culture history in the light of new discoveries. by THURSTAN SHAW* Ibadan, Nigeria. 7 v 68 The elucidation of the origins and develop- ment of agriculture in Africa must be one of the most important goals of archaeolo- gists working in the continent. The state to which knowledge of this topic has been brought in Southwest Asia, and the recent spectacular advances in knowledge of it in the New World, constitute a challenge. Therefore the initiative of CA is to be welcomed in commissioning the articles by Davies, Hugot, and Seddon, and we are indebted to them for their extremely useful reviews of the evidence. What emerges, however, is how little we really know and how much is conjecture. How right, for example, is Seddon that a detailed chronological framework is needed for the elucidation of problems of this kind and that only in one area of Africa has this prerequisite been achieved. Here is yet one more reasonfor emphasising the need to establish these chronological frameworks (Shaw 1963a: 27; 1963b: 462- 63). I also find Seddon's division of the evidence into three kinds-(a) direct, archaeological, (b) indirect, archae- ological, (c) botanical, linguistic, and ethnographic-sounder and more mean- ingful than Hugot's equation of "Pre- historic Artifacts + Botanical Proofs + Ab- solute Date = Agriculture (established fact)." I would question the logical validity of this equation, since the "Pre- historic Artifacts" usually constitute the indirect evidence of Seddon (often being ambiguous, as Hugot himself admits in speaking of mortars and pestles), "Botani- cal Proofs" (distribution of wild and cultivated varieties, etc.) seldom prove, and "Absolute Date," while contributing to the much-needed chronological frame- work and the general understanding of the pattern of origins and diffusion, is per se irrelevant to making the practice of agriculture "an established fact." If one scrutinises the data collected by Davies, Hugot, and Seddon, one has to admit that for the whole of Saharan and sub-Saharan Africa, only four unequivocal pieces of direct evidence are given (pollen of "a type of cultivated grass" from Meniet; carbonised sorghum from Isamu Pati and Kalundu; sorghum, and prob- ably Pennisetum, from Mwamasapa; sor- ghum and beans from Mapungubwe), and two "probables" (Adrar Bous and Kli- priviersberg). To these can be added finds of carbonised sorghum from Daima, northeastern Nigeria, estimated to date from the 9th/lOth century A.D. (Connah 1967 :25). This shows how limited our direct evidence is. It would be pedantically foolish to insist that this is the only sound evidence we have and to disregard the mass of circumstantial, indirect evidence; but it is the only "hard" evidence we have, and it behoves us to remember this in interpreting the rest of the data. How conjectural is much of the present state of our "knowledge" is illustrated by the fact that on the yam map of Hemardinquer, Keul and Randles (1967) the cradle area in Central Africa was at first dated to the 5th millennium B.C. and later, by a stroke of the pen, changed to the 2nd millen- nium. When I challenged this and asked for the evidence, W. G. L. Randles wrote: The yam map, based on Professor Oliver Davies's and J. D. Clark's work, has caused a certain uneasiness in a number of people's minds. In the final version still not published, I have redrawn it and postulated a date of "-Ha ?" ("First half of second millennium B.C.") for the nuclear area, then showing arrows of diffusion round and into the forest without further date. It is all still very hypo- thetical and must no doubt remain so until some way is found of identifying and dating yam cultivation remains in a stratified site. It is seldom clear in the literature whether "agriculture" includes or does not include the keeping of domesticated animals or refers only to the growing of crops; and whether the latter includes, in addition to cereals, the growing of vegeta- tive crops and trees. A set of definitions to meet every situation would be long and detailed, but could not the following be agreed upon ? Agriculture: any form of food production, except pure pastoralism. Pastoralism: the keeping of domesticated animals without the growing of any crops. Mixed Agriculture: the keeping of domes- ticated animals and the growing of crops. Cereal Agriculture (or "Seed Agricul- ture"): the growing of cereal crops. Vegetable Agriculture (or "Planted Agri- culture"): the growing of vegetable crops. Tree Agriculture: the growing of tree crops. Thus the terms "vegeculture" and "arbori- culture" are dispensed with; where cereals, vegetables and trees are grown and tended for their crops it is simpler to talk about "cereal, vegetable, and tree agriculture" than about "agriculture, vegeculture and arboriculture. " We might also avoid, as our contribu- tors, but not all other writers, do, a loose use of the word "millet," which is a generalised word like "corn" and should not be used without further definition if a particular crop is being referred to; e.g., sorghum is sometimes included in "millet," sometimes contrasted with it. Let us also get rid of the ugly term kwe- familiar enough in the literature of 500 DUU L Ult-I IJ ? IJLL 1 U L L 3 LL-I Y ii YLDa y LU CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

T)n17i.C "uIlnt nndl .Seddoi:n ORCTmNSTT OF ARPTtC-ANT AGC PTOUTUTPIRE SVMPOSUTTTM import from a local vernacular into English, when "digging stick weight" is perfectly reputable and self-explanatory and "large bored stone" may be used in contexts where it is desired not to prejudge function. Seddon refers to the possibility of planted agriculture preceding seed agriculture in East and southern Africa, and while he is quite correct in saying that there is at present "no convincing evidence from excavated sites that the threshold to food production was crossed before outside influence was felt," I am inclined to agree with Davies that there is a good possibility that indigenous yams came to be cultivated in West and Central Africa before, and independently of, the introduction into the area of food-producing ideas from the north. Coursey (1967: 7-10) certainly seems to believe in an early domestication of yams in Africa and points to the pro- hibition in certain areas on the use of iron tools for the digging of yams in New Yam festivals, which strongly suggests that yam cultivation antedates the commencement of the Iron Age. It certainly does seem "likely that agricultural concepts and techniques had reached the savannah area and been applied to the indigenous cereals by the 3rd or 2nd millennium B.C." (Seddon). The area where we must look for the origin of cultivated sorghum is now said to be north of 100 N. latitude and east of 250 E. longitude, but not extending into Egypt (Doggett 1965:58). It seems to be established that sorghum reached India from Africa in the second half of the 2nd millennium B.C. and travelled to the Middle East from there, although it was not grown in Egypt before the Roman Byzantine era (Doggett 1965:61, 62). In Davies' assertion that "in the sub- tropical zone it is easy to fix a line between pre-agricultural cultures and cultures which understood agriculture," the "line" can only be a purely theoretical one. The vital chronological line is extremely hard to detect with any degree of confidence. I think it is quite likely that Sangoan picks were used for gathering tubers. (I doubt, however, that Sangoan man would have been deterred from using them in digging hunting traps, if he had needed to, by its being "backbreaking work" ! Hoeing a yam field or a rice field is "backbreaking work.") Nevertheless, if the Sangoan pick is an earth-digging tool, one does wonder either why earth-digging was subsequently abandoned or what tool was substituted. Davies hazards that the wooden pestle and mortar may go back to proto- agricultural times; it might-but it seems to me to be a piece of equipment more in accord with systematic agriculture. It is in any case one of the great inventions, to which enough importance has not hitherto been attached; for it is the wooden pestle and mortar, above all, which render fibrous tropical tubers digestible. It is not true to say that we can have no evidence of the early introduction of the pestle and mortar because they are of wood, any more than this is true of the digging stick for the same reason (Gabel 1965:52-54); it just depends on the circumstance of preservation; we already have a pestle from Nigeria whose wood has been radiocarbon-dated to A.D. 885 + 120 (Fagg 1965:23). Mortars were used for breaking down herbs in ancient Egypt, and it is surely a wooden, and not a stone, mortar depicted in use in a Theban tomb painting dated to ca. 1900 B.C. (Singer 1965:274). The use of fire in hunting and agricul- tural practices should be remembered; probably the difficulty the baobab has in regenerating north of the forest is due to bush-burning rather than to the competi- tion of tall grass. I no longer think that the numerous ground stone axes found in West African forest regions were used for felling trees. The tiny ones were probably used for wood-carving and woodworking, the larger ones for dealing with roots and small branches and for killing trees by removing the lower bark preparatory to burning, the blunt-edged ones as earth hoes but where the trees of the rain forest were felled to make clearings, I am sure that fire was the instrument. A dead tree some 45m. high was once in the way of one of my excavations in the forest; a labourer with a steel axe made little impression on it in a whole day's work; after keeping a fire burning at its base for less than three days we felled it in exactly the place where we wanted it to fall. It is ambiguous to say, as Hugot does, that "the Epipalaeolothic industries of the Maghreb evolved locally towards a Neo- lithic"; this should mean an independent invention of agriculture in Northwest Africa, which I doubt, and I suspect that all that is being referred to is a response to the infiltration of food-producing ideas from further east. Similarly, to say "the Sahara was a centre for animal domestica- tion" leaves it unclear as to whether the meaning intended is that certain wild animals were domesticated there (which is what it should mean) or merely that it was an area where large numbers of domesti- cated animals were kept in early times (which is presumably what it does mean). The family of the edible Dioscoreaceae has more than two centres of origin, D. trifida being regarded as indigenous to southern Central America and two separate centres being recognised in Asia (Coursey 1967:11). It may also be worth correcting the notion, stated elsewhere (Alexander 1966:2), that D. bulbifera was introduced to Africa from Asia, since the African variety of this species is quite different from the Asiatic and must have been growing in the continent for many thousands of years before there was any agriculture anywhere in the world (J. R. S. Lawton, personal communication, 1968). I have laid emphasis on the paucity of direct evidence; the search for this direct evidence must go on. The possibility of finding macroscopic remains is much subject to chance, but as time goes on there should be an increasing number of well- dated finds of carbonised seeds. In order to get the direct evidence of agriculture that we need it is worth making a special search for water-logged sites, where remains are likely to be preserved. Similarly, there should be set on foot a piece of planned research to study systematically the in- creasing quantity of dated pottery now accumulating, in order to cull from it all the seed impressions available. What can be achieved when this is done has been demonstrated from the pottery of eight phases of the Neolithic in the Tichitt area of Mauretania, as a result of which were recognised impressions of the genera Chloris, Cenchrus, Panicuin, Echinochloa and Brachiaria, species of the last three having been minor African cultigens (Munson 1968). Finally, the study of African palynology should be encouraged and further endowed so that the evidence from pollen may be able to realise its maximum contribution to this particular question. Even though most developed yam culti- gens have come to depend on man's vegetative propagation and are not found in a wild form, for example, D. cayanensis and D. rotundata, and have no need to flower, they do still sometimes do so (J. R. S. Lawton, personal communica- tion, 1968). As the chronological frameworks for different parts of Africa are constructed, further indirect evidence will accumulate and will be able to be interpreted with greater reliability. In the realm of Seddon's third type of evidence, it might well be worthwhile to set up a research project to make a much more thorough and com- prehensive linguistic study of the terms for crops, domesticated animals, and agri- cultural implements and practices in the languages of Africa than it has been pos- sible to attempt hitherto. by J. E. SPENCERf Los Anigeles, Calif. 10 v 68 Hugot, Davies, and Seddon have per- formed a useful service in bringing to- gether known facts, pieces of evidence, and elements of relative chronology concerning the development of cropping practices in Africa. All three presentations blur the pic- ture, however, by confusing elements having to do with origins with elements that belong to~ late nrccssis o~f c'ron lderPlAlrnFen : Vol. 9 N No. 5 * December 1968 501This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Despite its broad meanings, the term "agriculture" is not really applicable to the initial processes of selecting wild plants, domesticating certain plants, and begin- ning the production of plant products. The term "agriculture" belongs, in the techno- logic sense, to those systems of production that involve advanced tools, draft-animal or mechanical power, developed cropping systems, and mature technologies of production, just as the term "manu- facturing" belongs to developed industrial operations. We should recognize that the protocultivation of semi-domesticates be- longs to origins and that ecological fitting-out of cropping systems belongs to a later era of development. It does not clarify the issues to introduce into the discussion of origins, as such, comment on matters that cannot possibly have any- thing to do with origins of plant domestica- tion and crop-growing systems. Thus, Hugot, in commenting on the plough, and Davies, in commenting on maize, discuss a tool technology and a crop plant which have to do with rather late developments now masking the primary problem of attempting to determine how crop-growing began in Africa, what its history of develop- ment has been, and what plants were employed in the initial systematic opera- tions. The human activities related to plants that characterize the Early Neolithic must be separated from those that characterize the mid-era phase of development and from those of the matured Late Neolithic. This holds for operations working with plants themselves, for operations con- cerned with the evolution of non-cultivat- ing tools into crop production tools, for the working out of cropping systems through the substitution of late-era plants for less useful early-era plants, for the integration of crop plants into animal economies and animals into cropping economies, and for such other aspects as the ecologic systemization of cropping combinations suited to particular environ- mental situations. Thus, I would object to Hugot's sentence "Agriculture is a phenomenon only of the Early Neolithic" as a contra- diction in terms, even though I recognize that his point was to suggest that in the Sahara the Late Neolithic was a period of environmental deterioration during which crop-growing activities were subjected to strong environmental stresses. I would suggest that in West Africa the adoption of rice as a crop staple, probably involving the replacement of tubers and other less- used plants, 'cannot possibly be part of the beginnings of crop production, but must rather represent the late-era matura- tion of a regional cropping system. Again the hybridization of already domesticated plants, producing varieties suited to local ecologies, is not an initial-stage activity. The diffusion of crop plants, often involv- ing the acceptance of a superior crop plant already long domesticated or of a more adaptable crop plant alternative to an older, less adaptable one long in use (during conditions of environmental change), is also clearly consequent to the beginnings of cultivation. We should probably be very critical of the notion that any grain crop group existing in a large number of varieties can be the result of the domestication of a single wild species. We now know enough about the complex history of the wheats to avoid the early errors in wheat discussions when we are considering the history of the rices, the sorghums, the millets, and, in Africa, several others of the grain-like genera. I am glad to see that none of the authors is bound by the too-frequent Vavilov syndrome (in which all domestications are referred to the Vavilov plant centers only). We must also avoid, however, the tendency to attribute all domestications to the Early Neolithic. Secondary domestications, both later in time and consequent upon the diffusion of primary crop-production ideologies, must have characterized the whole of the Neolithic in particular marginal, outlying, or distant regions. The adoption of already ongoing cropping practices by non-cultivating societies must have involved operations that were new to those specific societies; but these are operations that are relatively late in the history of crop production. Particularly does this refer to those integrative kinds of practices that led to the development of true agriculture-the animal-powered systems using advanced tools, advanced cropping systems, advanced dietary sys- tems, and advanced technologies in plant production and yield-processing. Only by careful differentiation of the kinds of activities that relate to origins, development systems, and integrative processes can we learn to separate begin- nings from late trends. In almost no part of the earth do we have sufficient well-catalogued information to character- ize the sequential growth of the origins and development of agriculture, but we must not go on mixing simplistically all kinds of activities together in our studies of the origins, developments, and matura- tions of Neolithic agriculture. by RICHARD A. YARNELL* Atlanta, Ga., U.S.A. 10 v 68 Each of the writers attempting to elucidate the origins and development of agriculture is apparently handicapped by a lack of sufficient evidence. Without adequate remains of the plants under discussion, it is difficult to go much beyond a discours in speculation. Nevertheless, the summaric of currently available evidence shoul prove useful. The quantity of plar remains reported from the Sahara i indeed surprising, but the scarcity c such material from West Africa is dis appointing. Though Murdock (1959) is cited b Seddon, none of the authors has attempte an evaluation of his interpretations. Edga Anderson, Burkill, and Vavilov also ar largely ignored; yet Davies sees fit t recognize the contributions of Cartei Heyerdahl, and Merrill. All three article are largely oriented to an archaeologica point of view to the detriment of th botanical, even when plants are discusse A total geographical orientation, empha sizing ecology, would be more meaningful Admittedly, however, the archaeologica data have largely been absent in the earlie interpretations and are absolutely vital t our understanding of agricultural evolu tion. Currently most lacking is direc archaeological evidence of the specifi relationships between plants and mar though this type of evidence is difficult t, produce, especially for the earlier stage of agricultural development. Hugot states that the combination c certain utensils . . . would lead one strongly to believ that their users ate vegetable products. Bu one cannot decide if their harvesting wa directed to wild or cultivated plants. This sort of orientation toward the plant involved seems to be fundamental nc only in the thinkingof theseCA authors bu in that of many others as well. Plants cai easily be both wild and cultivated at th same time if wild refers to the conditioi of the plants. "Cultivation" ordinaril refers to activities of man; what happens t, the plants as they are altered genetically physiologically, and morphologically as; result of man's activities is called "domesti cation." Agriculture, properly considered would include both cultivation and activi ties effecting propagation, the forme probably preceding the latter in th, evolution of agriculture. The earlies cultivated plants very likely were "wild' or (probably more accurately) "weedy.' Even if the term "domesticated" is sub stituted for Hugot's "cultivated," we ar still faced with the problem of intermediate which seem to be always present iI evolutionary sequences. Thus the state ment might well read "One cannot a present know the extent to which th harvested plants were domesticated (o had achieved cultigen status)." The poin would be trivial were it not for the fac that Hugot's statement is typical of s many others. 502 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies. Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM Replies by 0. DAVIES Our symposium has been unable to cover the subject as fully as might have been hoped. Most coastal territories south of the equator are little explored, and, further, we were unsuccessful in finding an Egyptol- ogist to co-operate. Hugot and I avoided domestic animals, because the abundant evidence, especially from the Sahara, would have made the articles too long. At the same time, few of our critics except Shaw discuss tuber agriculture, a subject much less studied than cereal agriculture and for that reason more exciting. No commentator referred to a recent article (Barrau 1967) discussing food- gathering and protoagriculture in the humid tropics of Oceania. Barrau points out that selection, protection, and en- couragement of plants indicate proto- agriculture; but true agriculture appears only when plants are removed from their natural habitat, so that their environment and reproduction are entirely under human control. This criterion does not apply universally in Africa; yams continued to be cultivated where they were indigenous. It is, however, useful for certain plants, such as the baobab. Muller's distinction between forest and savannah agriculture is misleading because the terms are inaccurately defined. With forest must be grouped wooded savannah (up to about 100 N. in Ghana), which has fairly small trees, well spaced, and gallery forest along rivers. This zone is easy to burn, and, as the distribution of archaeo- logical material from the earliest Holocene shows, easy to penetrate. It has enough rain to grow yams; but long droughts and shallow soils overlying crystalline rocks or lateritic crusts make it unsuitable for forest trees and bushes. The zone has recently extended southward; but its core has prob- ably not moved so long as climatic con- ditions have remained stable. Clearings in the high forest may not be older than true agriculture practised by by DAVID SEDDON The comments fall into three categories: those requiring a detailed reply, those a more general reaction and those, by their nature, needing little direct response. With- in this final category are (a) comments which merely reiterate, in other words, what the authors have said, (b) comments making minor points with which I am in agreement and (c) those that add new information or provide further references. Careful readers will be able to distinguish between (a), (b) and (c), but I am par- ticularly grateful to Butzer, Fagan, Gains, Gamst, Hester and Shaw for what they add-it is only a nitv that there is so little close-knit societies. A clearing around Kumasi is associated with the Kintampo Neolithic (the beginning of which is dated about 1400 B.C., according to Flight, but which may have lasted a long time). There is evidence in Ghana for other clearings associated with early types of stone axe. There are many clearings in Gaboon. Forest clearance has been plausibly dated in the Lower Congo to 3080 ? 170 B.P. (Lv-46). One point in reply to Brentjes: we do not know the beginnings of the Nok culture in Nigeria. The earliest reliable date is that from Taruga, I-1459, 2230 ?120, clearly well after the beginnings of agriculture in West Africa. The earliest date from Daima is not much older, 1-2372, 2400 ?95 (Fagan 1967). One can only reserve judge- inent on the much older dates from Nsukka University Farm, in view of the reversal of stratification (Fagan 1967). Seddon describes the introduction to southern Africa of the complex agriculture- cattle-iron. Commentators have associated this with the Bantu expansion into terri- tories previously enjoying a hunting and food-gathering economy. The complex may have originated in invasion from the northern savannah through Nigeria (as suggested by the undated cattle-painting at Birnin Kudu) and the Cameroons to the Lower Congo. Whether iron-working was brought by the first immigrants or was a later accretion we do not know. Russegger (1844:286) described very primitive meth- ods of iron-smelting in Kordofan, sugges- tive of continuous tradition from Pharaonic Egypt. Oliver (1966) locates the Bantu cradle in northeast Angola. I am inclined to put it farther west, pending investigation of large settlement mounds with apparently primitive material in the valley west of Leopoldville (Kinshasa). At Ntereso (Northern Ghana), invaders apparently brought cereals (as attested by sherd impressions), cattle, and probably iron about 1250 B.C., and fused with the local Kintampo people. In general, how- discussion from those at present working in the field on the problems of early farming in Africa. The relevant literature is at the moment largely inaccessible to me and I shall, of necessity, confine myself here primarily to general reactions. Two major themes, both stated at the beginning of the paper, which appear in nearly all the comments, are those of the patchiness and inadequacy of the data and of the rapid changes that the general pic- ture is undergoing as a result of more recent excavation and analysis. Indeed, Fagan (probably my severest critic and one nearer than most to the original ever, there is no evidence north of the equator for development by invasion, and dates are far older than in eastern and southern Africa. As Hester suggests, there may have been local domestication of cereals in the eastern Sahara, and a fortiori of tubers in other areas, where hoe was established instead of plough and garden instead of field. Though we have no verifi- cation from radiocarbon dates, there appears to be, near the middle Niger and in southern Mauritania, a continuity of heavy digging tools from the Sangoan to the Neolithic. If certain plants were domesticated in- dependently in Africa, cattle must at some stage have been introduced from abroad. Short-horned and long-horned breeds may have arrived separately and by different routes. The former do not travel well, so will have spread along valleys rather than across the Sahara, which even at the period of optimal climate must have had wide barren stretches. By the 6th or 5th century B.C. there were in West Africa, according to Herodotus (Book 2, Chap. 32-33, probably in the Djourab; Book 4, Chap. 195, perhaps on the upper Niger), communities capable of forming big villages and of mining and trading gold. Not much later, the highly developed Nok culture had iron, probably cattle, and almost certainly agriculture. In the 1st millennium A.D., cities and or- ganised states appeared in the Niger valley, probably by radiation from the Roman empire; they could not have existed without agriculture. Two final points, in reply to Kennedy: I have had to rely on the published photo- graphs of the gouges from Shaheinab and elsewhere. A detailed examination is needed to decide if all are wood-working tools. Secondly, texts which have been quoted in the controversy about pre- Columbian maize in the Old World should be scrutinised according to the standards of source-criticism which have been de- veloped for classical literature. material) uses these two problems to ques- tion, in effect, the raison d 'tre of the review. That the article now "contains nothing new" may possibly be true for Fagan and number of others, and it would seem that syntheses over such broad areas are liable to run some two years behind r esearch articles in specialised archaeological jour- nals. Other printed remarks and informal letters, however, suggest that such an over- view is useful and interesting to many. Clearly it is important that "fieldwork and study of local sequences" should continue to take place and also that the results should be reported as soon as possible, but syn- thesis is necessary and one of the major Vol. 9 * No. 5 * December 1968 503This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

TABLE 1 A. ANIMALS B. PLANTS AA Hunting AB Fishing BB Gathering WILD Stage 1. Herding Collecting Stage 2. l l Stock raising Growing DOMESTI- Stage 3. CATED difficulties facing those not immediately involved in working on the problem is the almost incredible dispersal of information in small, local journals. If it is felt that the delays in publication usually experienced by large formal journals attempting to provide broad syntheses are too great, the simpler and faster method of duplicating on a roneo machine could be used; i.e., all those working on a particular topic perhaps could write a short summary every year or even six months for CURRENT ANTHRO- POLOGY, who could roneo the reports and sell them at a small price. Discussion of problems such as this may be considered unnecessary "meta-archaeology" but the proper and timely dissemination of infor- mation is as important as its gathering, in archaeology as in other subjects, and com- ments on these reviews suggest that the problem is a real one. A third theme of considerable general importance can be extracted from the dis- cussion-that of the nature and reliability of evidence and inference from excavated materials and from ethnographic and other studies. This is a topic of crucial importance to the archaeology of early farming and one that has recently received some long overdue attention (cf. Chang 1967a, 1967b). All too frequently, how- ever, the logical bases for inference are not rigorously applied, and what results is a shaky construction founded on little more than guesswork. A broad review must, of necessity, be a compromise in this respect, but, despite suggestions that it is as a whole "largely conjectural," the introduction emphasises the vital necessity of appreciat- ing the different types of evidence and their reliability or "power." With regard to direct botanical evidence, it is essential to realise that only this provides absolutely reliable evidence; the interpretation of material artifacts is always fraught with difficulties of classification and comparison. The "indirect inference" possible from ethnographic, contemporary botanical, and other studies is more of a general guide than anything to be too heavily relied upon. (For a discussion of "direct" and "indirect" inference in archaeology and some of the problems of artifact typology, see Seddon 1969). Specific examples of the difficulties and dangers of artifact "identification" may be found among the comments to the review (although wilder and more "conjectural" explanations of artifacts may be founc without much effort in the general archaeo logical literature [for just one example se( Seddon 1966c]), and the problems associ- ated with loose ethnographic analogies are also evident. The fourth theme is that of the sub- division, reappraisal or criticism of the ad- mittedly loose general category of "agri- culture." In view of the present inadequacy of relatively reliable data, much discussion of finer points is premature but the various suggestions are worth bearing in mind foi the future. I merely add my own here (sec Table 1) for what they are worth. "Growing" will be associated with dif- ferent styles of cultivation, depending partly on the type of plant cultivated and partly on social and cultural traditions. It might be useful, however, to divide "plants" provisionally into (1) seed/grasses (2) root-tuber/plants (3) fruit/plants (4) fruit/trees (5) fruit/bushes (6) leaf-stalk/ plants (7) leaf/bushes (cf. Murdock 1959: 181-84). Economies with both A and B in Stage 1 have been termed "hunters and gatherers"; those with both A and B in Stage 3, "mixed farmers"-the word "farming" being reserved for general refer- ence to fully Stage 3 economies. The type of land utilisation and style of cultivation are associated with and will dictate whether the term "agriculture" or "horticulture" is used. The term "foraging" could be used as AC to describe the collection of snails, grubs, and other small creatures (hardly to be regarded as hunting) as well as the search for carrion and dead meat. Finally, a few more specific replies or remarks about particular comments. I can- not agree with Brentjes that "one of the things the Near East teaches is the close relationship of animal breeding and field cultivation (see p. 425)." We just do not have enough information to support this if he really does believe in "tribes of cattle- breeders from the 10th millennium B.C." Regarding Fagan's contribution, it may be worth quoting a few more recent views. Posnansky says (I1968: 1; 1968: 3) I have argued in favour of a movement of the Dimple-based/Channelled ware peoples from an area to the west of the Zambian plateau, to the Zambezi valley and north to the lake Victoria region. The impression gained is that the Dimple- Based ware and Channelled ware have a relationship but are not the same type of pottery. Both may have derived from a com- mon source and in this connection it is interest- ing to note that the Kasai pottery is closer to the Dimple-Based wares ... than the Chan- nelled wares. According to Phillipson (1968: 21 1) The Early Iron age people appear to have been responsible for the introduction into Zambia of pot-making, metallurgy and, less certainly, food production.... It is suggested that the early iron age people slowly spread into Eastern Africa from an area west of Lake Tanganyika during the first few centuries A.D. I accept Fagan's strictures regarding the distribution of Dimple-Based pottery sites and the economic interpretation to some extent, but it is of interest that Posnansky, who is responsible for the excavation of many of these pottery sites wrote in 1961 (Posnansky 1961c: 185) It is tempting on the basis of this regional differentiation to think in terms of the Dimple- Based pottery users as agriculturalists or pos- sibly people with a mixed economy, in the sense that they were migratory, who still aug- mented their food supply by hunting, fishing and food-gathering.... and again, in 1968 (Posnansky 1968: 3) One further feature of the Dimple-Based and Channelled wares is not that they are found widely scattered, but that geographically they are restricted to certain environments, along lake fringes and by rivers like the Zambezi and Kagera valleys, where a mixed economy of foraging and agriculture could have been possible. Evidence for contact between the coast and south central Africa now seems to suggest beginnings in the 4th to 5th century, but for Mabveni Robinson (1961c: 97) notes that ... the only alternative evidence of date, the three glass beads, is open to more than one interpretation and an early or late date might equally well be argued by the bead types. The inclusion in the bracketed references of Murdock as a supporter of the "Ethiopia as an independent centre" school was a slip for indeed he suggests that "at some time prior to 3000 B.C. the Negroid ances- tors of the Prenilotes penetrated the plateau from the west, bringing with them agri- culture of a Sudanic type" (Murdock 1959: 181). Finally I feel that Shaw's reminder of the importance of fire in certain agri- cultural practices is worth underlining. Traces of ash and burnt wood could be picked up in excavation and there is an abundant literature on contemporary and near-contemporary "slash-and-burn" agri- culture or horticulture. So far it has been possible to reconstruct to some extent the settlements of prehistoric farmers in eastern and southern Africa (in some areas, at least); it would be exciting if progress could be made towards an understanding of the systems of land use and actual agricultural practices. 504 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

7-1--.;-- -InN -A- -D TTT T References Cited ABERG, E. 1950. Bai ley and wheat from Saggara Pyramid in Egypt. Annals of the Royal Agricultural Society of Sweden, XVII, Uppsala 17:59-63. ALEXANDER, JOHN. 1966. The domestication of yams. Research Seminar on Archaeology and Related Subjects, London University Institute of Archaeology. Mimeographed. [TS -] ALLAN, W. 1965. The African husbandman. Edinburgh and London: Oliver and Boyd. ALLISON, P. A. 1962. Historical inferences from the effect of human settlement on the vegetation of Africa. jrournal of African History 3:241-49. ANDERSON, EDGAR. 1960. "The evolution of domestication," in Evolution after Darwin. Edited by Sol Tax, vol. 2, pp. 67-84. Chicago: University of Chicago Press. ARKELL, A. J. 1949. Early Khartoum. London: Oxford University Press. - - -. 1953. Shaheinab. London: Oxford University Press. -- -. 1954. Four occupation sites at Agordat. Kush 2:33-62. 1961. A history of the Sudan. London: Athlone Press. -- -. 1966. The Iron Age in the Sudan. CURRENT ANTHROPOLOGY 7:451-52. ARKELL, A. J., and P. J. UCKO. 1965. Review of predynastic development in the Nile Valley. CURRENT ANTHROPOLOGY 6:145-56. ASSEMIEN, P. 1965. Les d6pots du quaternaire r6cent dans la basse vall6e du S6n6gal. Determination palynologique. Association se'negalaise pour l'etude du Quaternaire de l'Ouest africain 8. [HG*] AUBREVILLE, A. 1962. Savanisation tropicale et glaciations quaternaires. Adansonia 2:16- 84. BAILLOUD, G. 1959. La pr6histoire de l'Ethi- opie. Mer Rouge-Afrique orientale, Cahiers de l'Afrique et de 1'Asie, pp. 15-43. Paris. BAKER, H. G. 1962. Comments on the thesis that there was a major centre of plant domestication near the headwaters of the River Niger. Journal qf African History 3:229- 33. [FCG * ] BALOUT, L. 1952. Pluviaux, interglaciaires et pr6histoire saharienne. Travaux de l'Institut de Recherches Sahariennes 8:9-12. - - -. 1955. Prdhistoire de l'Afrique du Nord. Paris: A.M.G. BARADEZ, J. 1949. Fossatum Africae: Recherches aeriennes sur l'organisation des confins sahariens a l'epoque romaine. Paris: A.M.G. BARRAU, J. 1967. De l'homme cueilleur a l'homme cultivateur. Cahiers d'histoire mon- diale 10: 277-92. BELLAIR, P., and D. PAUPHILET. 1959. L'age des tombes pr6islamiques de Tejerhi (Fezzan). Travaux de l'Institut de Reclierches Saharaiennes 18: 183-85. BERNARD, E. A. 1962. Interpr6tation astro- nomique des pluviaux et interpluviaux du Quaternaire Africain. IBERSOM, Bruxelles 40:67-96. BERNHARD, F. 0. 1961. Ziwa wares of Inyanga. NADA 36:84-92. BLANC, A. C. 1939. Une lettre sur les corrdla- tions pal6olithiques. L'Anthropologie 49:611- 15. BOND, G. 1957. The geology of the Khami Stone Age sites. Occasional Papers of the Museum of Southern Rhodesia, 21lA(3) :44-55. BRAIDWOOD, R. J. 1957. Jericho and it setting in Near Eastern history. Antiquit 31:73-84. BRAIDWOOD, R. J. and B. HOWE. 1960 Prehistoric investigations in Iraqi Kurdistan Studies in Ancient Oriental Civilization 31 - - -. 1962. "South-western Asia beyond th( lands of the Mediterranean littoral," ir Courses toward urban life. Edited by R. J Braidwood and G. R. Willey. Viking Func Publications in Anthropology no. 32. BRAIDWOOD, R. J. and 0. A. REED. 1957 The achievement and early consequences o food-production: A consideration of th( archaeological and natural-historical evi dence. Cold Spring Harbor Symposia o0 Quantitative Biology 22: 19-31. BRAUN, A. 1877. Die Pftanzenreste des dgyplischei Museums in Berlin (Vortrag von 1871). Berlin, Wiegardt. [FCG ( BRENTJES, B. 1965. Fels- und Hdhlenbilde Afrikas. Leipzig. [BB ] BROTHWELL, D. R. 1963. Evidence of earl) population change in Central and Southerr Africa: Doubts and problems. Alan 132: 101 - 4. BRUNTON, G. 1937. Mostagedda and the Tasiar, culture. London: Quaritch. [WR * ] BURKILL, I. H. 1950-51. The rise and decline of the greater yam. Advancement of Science 7:443-48. - - -. 1953. Habits of man and the origins ol the cultivated plants of the Old World. Proceedings of the Linnean Society of London 164:12-42. BussON, F., P. JAEGER, R. CARBIENER, and J. LANZA. 1959. Contribution a l'6tude botanique de Coleus rotzundifolius. journal d'Agriculture tropicale et de Botanique applique'e 6:675-79. BUTZER, K. W. 1957. Mediterranean pluvials and the general circulation of the Pleisto- cene. Geografiska Annaler 39:48-53. ---. 1961. "Les changements climatiques dans les r6gions arides depuis le Pliocene," in A history of land use in arid regions, pp. 35-64. Paris: UNESCO. -- -. 1962. "The Pleistocene sequence in Egypt and its implication for the pluvial- glacial correlation in the Sahara." Actes du IVe Cong7es .Panafricain de Prehistoire, Leopold ville, 1952, vol. 8 (40), pp. 133-39. - - -. 1964. Environment and archaeology. Chicago: Aldine Press. [JJH ] BUTZER, K. WV., and C. L. HANSEN. 1968. Desert and river in Nubia: Geomorphology and prehistoric environments at the Aswan Reservoir. Madison: University of Wisconsin Press. [KWB * ] CADENAT, P. 1960. Une faucille pr6historique a Columnata. Libyca 8:238-52. CAMPBELL, J. 1815. Travels in South Africa. London. - - -. 1822. Travels in South Africa ... the second journey. Vol. 1. London. CAMPS, G. 1966. Nouvelles dates donn6es par le carbone- 14 concernant la prehistoire r6cente en Algerie (Capsien superieur et Neolithique). Bulletin de la Societe Pre- historique Franfaise, Comptes rendus des seances mensuels 58(2):84-88. CARTER, G. F. 1964. Archaeological maize in West Africa. Man art. no. 95. CHANG, K. C. 1967. Major aspects of the interrelationship of archaeology and ethnol- ogy. CURRENT ANTHROPOLOG.Y 8:227-43. [FCG W] - - -. 1967b. Rethinking archaeology. Random House, New York. CHAVAILLON, J. 1964. Lesformations quaternaires du Sahara Nord-occidental. Paris: Centre National de Recherche Scientifique. CHEVALIER, A. 1932. Les productions v6g6tales du Sahara et de ses confins Nord et Sud. Passe, present, avenir. Revue internationale de Botanique Appliquee et d'Agriculture Tropicale 133-34, 669-924. -- -. 1936. Contribution 'a I'tude de quelques especes africaines du genre Dioscorea. Mu- seum d'Histoire naturelle Bulletin 2 (8) :520-51. -- -. 1938. Le Sahara, centre d'origine de plantes cultivees. La vie dans les regions de- sertiques Nord-tropicales de I'ancien monde. Societe de Biogeographie, Memoires 6:307-22. ---. 1949. Un igname sauvage a tubercules comestibles. Revue de Botanique appliquee 19:609-12. CHITTICK, N. 1965. The "Shirazi" coloniza- tion of East Africa. Journal of African History 6:275-94. CHRISTIE, A. H. 1957. An obscure passage from the Periplus. Bulletin of the School of Oriental and AJrican Studies 19:345-53. CLARK, J. D. 1950. The Stone Age cultures of Northern Rhodesia. Cape Town: South African Archaeology Society. ---. 1954. The prehistoric cultures of the Horn of Africa. Cambridge: Cambridge University Press. -- -. 1959. The prehistory of Southern Africa. London: Penguin Books. ---. 1962a. "Africa south of the Sahara," in Courses toward urban life. Edited by R. J. Braidwood and G. R. Willey. Viking Fund Publications in Anthropology no. 32. -- -. 1962b. The spread of food-production in sub-Saharan Africa. Journal of African History 3:211-28. ---. 1962c. "The Kalambo Falls prehistoric site: An interim report." Actes diu IVe Congres Panafricain de Prehistoire, Tervuren. Edited by G. Mortelmans and J. Nenquin, pp. 195-202. Tervuren: Mus6e Royal de l'Afrique Centrale. ---. 1964a. Prehistoric culture and Pleisto- cene vegetation at the Kalambo Falls, Northern Rhodesia. Nature 4923:971-75. - - -. 1964b. Stone vessels from Northern Rhodesia. Man 64:69-73. -- -. 1965a. "Changing trends and developing values in African prehistory." African Affairs, Report on Spring Conference 1964. -- -. 1965b. Comment on: Review of pre- dynastic development in the Nile Valley, by A. J. Arkell and P. J. Ucko. CURRENT ANTHROPOLOGY 6:158-59. - - -. 1966. Precision and definition in African archaeology. South African Archaeological Bulletin 2 1: 114-2 1. [FCG * ] - - -. 1967. "The problem of Neolithic culture in subsaharan Africa," in Background to evolution in Africa. Edited by W. W. Bishop and J. D. Clark, pp. 601-27. Chicago: University of Chicago Press. [KWB f ] CLARK, J. D., and B. M. FAGAN. 1965. Charcoals, sands and channel-decorated pottery from Northern Rhodesia. American Anthropologist 67:354-71. COE, M., and K. V. FLANNERY. 1964. Microenvironments and Mesoamerican pre- history. Science 143:650. COLE, SONIA. 1964. The prehistory of East Africa. London: WTeidenfeld and Nicolson. Vol. 9 - No.- 5 * December 1968 505This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

- --. 1965. The NVeolithic revolution. London: British Museum. COLSON, E. 1949. Life among the cattle-owning Plateau Tonga. Occasional Papers of the Rhodes-Livingstone Museum 6. - - -. 1960. The social organisation of the Gwembe Tonga. Manchester: Manchester University Press. CONKLIN, H. C. 1961. The study of shifting cultivation. CURRENT ANTHROPOLOGY 2:27- 61. CONNAH, GRAHAM. 1967. "Progress report on archaeological work in Bornu 1964-1966," in Northern History Research Scheme, Second Interim Report. Edited by H. F. C. Smith, pp. 20-31. Zaria: Ahmadu Bello University and University of Ibadan. [TS*] COOKE, C. K. 1963. Report on excavations at Pomongwe and Tsangula caves, Matopo Hills, Southern Rhodesia. South African Archaeological Bulletin 18:75-171. COURSEY, D. G. 1968. Yams. London: Long- mans. DALZIEL, J. M. 1937. The useful plants of West Tropical Africa. London: Crown Agents. DANIELS, S. G. H. 1964. Monunments Commission annual report for 1963. Livingstone. - -. 1965. "The statistical determination of pottery types from the recently excavated site of Dambwa, near Livingstone." Pro- ceedings of the Central African Scientific and Medical Conference, Lusaka, 1963. Oxford. DAVIDSON, B. 1964. Old Africa rediscovered. London: Gollancz. DAVIES, 0. 1962. "Neolithic cultures of Ghana." Actes dut IVe Congres Panafricain de Pr6/histoire, Tervuren. Edited by G. Mortel- mans and J. Nenquin, pp. 291-301. Tervuren: Koninklijk Museum voor Mid- den-Afrika. [JM * ] - --. 1964a. The Quaternary in the coastlands of Guinea. Glasgow: Jackson. - --. 1964b. The Lower and Middle Palaeo- lithic in West Africa. Rivista di Scienze preistoriche 19: 1-21. ---. 1966. Comment on: The Iron Age in Sub-Saharan Africa, by A. J. Arkell and R. Summers. CURRENT ANTHROPOLOGY 7:470-71. DEACON, J. 1966. An annotated list of radio- carbon dates for sub-Saharan Africa. Annals of the Cape Provinzcial Museums 5. DECHAMBRE, E. 1950. Le Sahara. Centre primitif de domestication. Comptes rendus de la Socite' de la Bioge6ographie 236-38, 147-51. DELIBRIAS, G., H. J. HUGOT, and P. QUE1ZEL. 1957. Trois datations de sediments recents pour le radio-carbone. Lybica 5:267-70. DOGGETT, H. 1965. "The development of the cultivated sorghums," in Crop plant evolution. Edited by Sir Joseph Hutchinson, pp. 50- 69. Cambridge: Cambridge University Press. [TS *] DOIZE, R. L. I 948. Les boules de pierre et les pierres perforees du Musee du Congo. Muse'e royal de Congo belge (Tervuren), Annales in 40 I (i) : part iii. ERDTMAN, G. 1943. An introduction to pollen analysis. Waltham, Mass. - - -. 1944. Sadeslagens pollenmorphologi (Pollen morphology of the cereals). Svensk Botanisk Tidskrift 38:73-80. [HG A ] EHRICH, R. W. Editor. 1965. 2nd edition. Chronologies in Old WForld archaeology. Chicago and London. [BB W] ERROUX, J. 1961 (?) Les ble's des oasis sahari- ennes. Universite d' Algerie, Institut de Recherches Sahariennes, mEmoire 7. FAGAN, B. M. 196la. Radio-carbon dates for sub-Saharan Africa 1. Journal of African History 2:137-39. - - -. 196 1b. Pre-European iron working in Central Africa. Journal of African History 2: 199-210. - - -. 1962. Bored stones from the Kalomo culture. South African Archaeological Bulletin 17:67-197. -- -. 1963a. Radio-carbon dates for sub- Saharan Africa 2. Journal of African History 4: 127. -- -. 1963b. The Iron Age sequence in the Southern Province of Northern Rhodesia. Journal of African History 4:157-77. -- -. 1963c. Letter to the editor. South African Archaeological Bulletin 18:60. - - -. 1963d. The Kalomo/Choma Iron age project (1960-3): Preliminary report. South African Archaeological Bulletin 18: 69. -- -. 1964. The Greefswald sequence: Bambandyanalo and Maungubwe. Journal of African History 5:337-61. -- -. 1965a. Radio-carbon dates for sub- Saharan Africa 3. Journal of African History 6: 107-16. - --. 1965b. Southern Africa during the Iron Age. London: Thames and Hudson. -- -. 1966a. The Iron Age of Zambia. CURRENT ANTHROPOLOGY 7:453-62. - - -. 1966b. Radio-carbon dates for sub- Saharan Africa 4. Journal of African History 7:495-506. [BMF * ] -- -. 1967a. Iron Age cultures in Zambia I. Katorno and Kangila. London: Chatto and Windus. [BMF A ] -- -. 1967b. Radio-carbon dates for sub- Saharan Africa 5. Journal of Af rican History 8: 513-2 7. [BMF * ] - --. 1968a. "The Iron Age Sequence in the southern parts of Zambia," in Studies in African prehistory. Edited by J. D. Fage and Roland Oliver. Cambridge: University Pre?s. In press. [BMF ] - - -. 1968b. Radio-carbon dates for sub- Saharan Africa 6. Journal of African History 9(4). In press. [BMF ] FAGAN, B. M., and D. W. PHILLIPSON. 1965. Sebanzi: the Iron Age sequence at Lochin- var, and the Tonga. Journal of the Royal Anthropological Institute 95, part 2. FAGAN, B. M., and F. VAN NOTEN. 1964. Two channel-decorated pottery sites from N. Rhodesia. Man 64:8. FAGG, B. E. B. 1962. "The Nok terra-cottas in West African art history." Actes du IVe Congres Panafricain de Prehistoire, Tervuren. Edited by G. Mortelmans and J. Nenquin, pp. 445-50. Tervuren: Musee Royal de l'Afrique Centrale. - - -. 1965. Carbon dates for Nigeria. Man 65:22-23. [TS*] FIRBAS, F. 1937. Der pollenanalytische Nachweis des Getreidebaus. Zeitschrift Botanik 31:447-78. [HG*] LAKSBERGER, C. 1928. The emmers (Triticum dicoccum Schrank) of ancient Egypt and of modern times. Bulletin of Applied Botany, of Genetics and Plant Breeding 19. 'LANNERY, K. V. 1965. The ecology of early food production in Mesopotamia. Science 147:1247. LINT, R. F., 1959. Pleistocene climates in Eastern and Southern Africa. Bulletin of the Geological Society of America 70: 343-74. O0CK, G. J. 1956. Stone bowls fromn S.W. Africa. South African jtournal of Science 52: - - -. 1960. Another stone bowl from Southern Africa. South African Archaeological Bulletin 15:114. -- -. 1961. Steintopfe im sudlichen Afrika. journal of the South West African Scientific Society 15:41-46. FORBES, R. J. 1950. Metallurgy in antiquity. London. FORDE, D. 1946. Habitat, economny and society. London: Methuen. FOUCHA, L. 1937. Mapungubwe: Ancient Bantu civilisation on the Limpopo. Cambridge: Cambridge University Press. FRANCHET, L. 1932. Les premiers instruments de labour et leur evoluttion. Extraits des comptes rendus de la 4e semaine du machinisme a la ferme, Lausanne. FREEMAN-GRENVILLE, G. S. P. 1959. "The times of ignorance: A review of pre-Islamic and early Islamic settlement on the East African Coast," in Discovering Africa's past. Uganda Museum Occasional Paper, no. 4. GABEL, C. 1960. Seminar on economic types in pre-urban cultures of temperate wood- land, arid and tropical areas. CURRENT ANTHROPOLOGY 1: 437-38. - --. 1963. Lochinvar Mound, a Later Stone Age camp site in the Kafue Basin. South African Archaeological Bulletin 18: 40-48. - - -. 1965. Stone Age hunters of the Kafiue. Boston: Boston University Press. [TS*] GALLOWAY, A. 1959. The skeletal remains of Bambandyanalo. Johannesburg: Witwaters- rand University Press. GARDNER, E. W. 1935. The Pleistocene Fauna and Flora of Kharga Oasis, Egypt. Quarterly Journal of the Geological Society of London 91, 4: 479-518. GARDNER, G. A. 1955. Mapungubwe 1935-40. South African Archaeological Bulletin 10: 73-77. -- -. 1958. Mapungubwe and the Second Volume. South African Archaeological Bulletin 13:123-32. - --. 1963. Mapungubwe. Volume 2. Report on excavations at Mapungubwe and Bambandyanalo in N. Transvaal from 1935-1940. Pretoria: van Schaik. GARDNER, G. A., L. H. WELLS, and J. F. SCHOFIELD. 1941. The recent archaeology of Gokomere. Transactions of the Roval Society of South Afrita 28:219-53. GOODWIN, A. J. H., and C. VAN RIET LOWE. 1929. The Stone Age cultures of South Africa. Cape Town. GoVOROV, L. I. 1929-30. The peas of Abys- sinia. A contribution to the problem of the origin of cultivated peas. Bulletin of Applied Botany, of Genetics and Plant Breeding 24:420- 31. [FCG ] GRAY, R. 1962. Report on the Third Con- ference on African History and Archaeology, July 1961. journal of African History 3: 175- 91. GUTHRIE, M. 1962. Some developments in the prehistory of the Bantu languages. journal of African History 3:273-82. HARLAN, I. R. 1967. A wild wheat harvest in Turkey. Archaeology 20:197-201. [BB * ] HARLAN, I. R., and D. ZOHARY. 1966. Distribution of wild wheats and barley. Science 153: 1074-86. [BB * ] HARTMANN, F. 1923. L'agricuilture cienne Egypte. Paris: Librairie Imprimeiers reunies. HELBAEK, H. 1953. Archaeology and agri- cultural botany. University of London, Insti tute of Archaeology, Annual Report 9:44-[WR~] 506 URRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies, Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM - - -. 1955. Ancient Egyptian wheats. Pr( ceedings of the Prehistoric Society, N.S., 21 (11) 93-95. HEMARDINQUER, JEAN-JACQUES, MICHAE KEUL, and W. G. J. RANDLES. 196 Ignames. Documentation ge6ographique et histo2 ique. Paris. Centre de Recherches Historique de I'Ecole Pratique des Hautes Etudes. [TSW HEYERDAHL, T. 1964. Plant evidence fo contacts with America before Columbus Antiquity 38:120-33. HIERNAUX, J. 1959. Recent research at protc historic sites in Ruanda, in the Belgiai Congo (Katanga Province) and in Ugand (Kibiro). Uganda Museum Occasional Papei 4:26-30. - - -. 1966. Comment on: The Iron Age i] sub-Saharan Africa, by A. J. Arkell, E Fagan, and R. Summers. CURRENT ANTHRO POLOGY 7:47 1. HIERNAUX, J., and E. MAQUET. 1960. Culture prehistoriques de 1',age des metaux au: Ruanda-Urundi et au Kivu (Congo Belge -2. Mdrnoires de l'Acaddmie Royale de Scien;es d'Outremer, Classes des Science Naturelles et Medicales, Bruxelles, N. S., 10(2) 1-88. HIGGS, E. S. 1961. Some Pleistocene faunas o the Mediterranean coastal areas. Proceeding of the Prehistoric Society 27: 144-54. -- -. 1962. A metrical analysis of some pre historic domesticated animal bones fron Cyrenaican Libya. Man 62:200. HILL, M. H., and J. S. HANDLER. n.d The stone age in West Africa: P review of the archaeological literature MS. Southern Illinois University. [JMf HOBLER, PHILIP, and J. J. HESTER. 1969 Prehistory and environment in the Libyar Desert. South African Archaeological Bulleti7 69. In press. [JJH * ] HUGOT, H. J. 1962. Premier apercu sur la pre- histoire du Tener6 du Tefassasset. Document, Scientifiques. Missions Berliet Tinere'-Tchad Paris: A.M.G., edit. 149-68. - -. 1963. Recherches prdhistoriques dau. I'Ahaggar nord-occidental. Memoire du C.R.A.P.E. 1. - - -. 1966. Une industrie nouvelle: le Ndolithiqwu fadelien. In press. HUGOT, H. J., and P. QUE~ZEL. 1957. A propos de quelque graines fossiles du gisement pre- historique de Meniet. Bulletin de la Socie'ti de Histoire Naturelle de l'Afrique du Nord 48: 370- 73. INSKEEP, R. R. 1962. Some Iron Age sites il Northern Rhodesia. South African Archaeo- logical Bulletin 17: 136-80. - --. 1963. Letter to the editor. South African Archaeological Bulletin 18:32. ---. 1965. Preliminary investigations of an Iron Age cemetery at Nkudzi Bay, Malawi. National Museum of Zambia, Special Paper. IVANOV, A. 1929-30. The sorghum of Northeastern Africa and Southwestern Arabia. Bulletin of Applied Botany, of Genetics and Plant Breeding 24:273-300. [FCG ] JASPAN, M. A. 1953. The Ila/Tonga peoples of North- Western Rhodesia. London. JEFFREYS, M. D. W. 1954. The history of maize in Africa. South African J7ournal of Science 50: 197-200. - --. 1955. "Some notes on the Neolithic of West Africa." Third Panafrican Congress Proceedings, no. 40. - - -. 1957. The origin of the Portugucsa word zaburro as their name for maize. Bulletin de l'Institutfranfais d'Afrzque noire 19B: I 11-36. - - -. 1963. How ancient is West African maize? Africa 33:115-31. - - -. 1966. Bored stones. South African Archaeological Bulletin 21:186. JENSEN, A. E. 1959. Altvdlker Sud-Athiopiens. Stuttgart: Kohlhammer. [WR*] JONES, A. M. 1959. Indonesia and Africa: The xylophone as a culture-indicator. Journal of the Royal Anthropological Institute 89:155-68. KAISER, W. 1957. Zisr Inneren Chronologie der Naqadakultur. Archaeologia Geographica 6:69-77. KEIMER, L. 1924. Die Gartenpflanzent im Alten Agypten. Hamburg. [RH A ] KENNEDY, R. A. 1960. Necked and lugged axes in Nigeria. Antiquity 34:54-58. - -. 1965a. The spread of Neolithic modes of life in Africa. Unpublished B. Litt. disserta- tion, Oxford University, Oxford, England. [RAK * ] ---. 1965b. Comment on: Review of pre- dynastic development in the Nile Valley, by A. J. Arkell and Peter J. Ucko. CURRENT ANTHROPOLOGY 6:160. [RAK ] KIRBY, P. R. 1965. SirAndrew Smith. Cape Town. KLIMA, B. 1962. "The first ground plan of an Upper Palaeolithic loess settlement in Middle Europe and its meaning," in Courses toward urban life. Edited by R. J. Braidwood and G. R. Willey. Viking Fund Publications in Anthropology no. 32. KOEPPEL, R., R. MALLON, and R. NEUVILLE. 1934-40. Teleilat Ghassul. 2 vols. Rome. [BB*c] KORN, H., and H. MARTIN. 1957. "The Pleistocene of South West Africa." Pro- ceedings of the Third Panafrican Congress on Prehistory. Edited by J. D. Clark. London: Chatto and Windus. KROEBER, A. L. 1940. Stimulus diffusion. American Anthropologist 42: 1-20. [FCG ] KUPRIANOVA, L. A. 1959. Pollen and spores of West African mangrove swamps. Doklady Akademiya Nauk SSSR 129:428-30. English translation in Proceedings, American Geological Institute, 1960, pp. 1186-88. [HG*r] LASSERRE, G. 1964. Du mais sauvage au mais cultive. Cahiers d'Outremer 17:314-24. LAURENT-TACKHOLM, V. 1951. The plant of Naqada. Annales du Service des Antiquites de l'Egypte 51:299-312. LEAKEY, L. S. B. 1935. The Stone Age races of Kenya. Oxford: Oxford University Press. LEAKEY, L. S. B., and M. D. LEAKEY. 1950. Excavations at the Njoro River Cave. London: Oxford University Press. LEAKEY, M. D. 1943. Note on the ground and polished stone axes of East Africa. journal of the East Africa and Uganda Natural History Society 17: 182-95. - - -. 1945. Report on the excavations at Hyrax Hill, Nakuru, Kenya Colony. Transactions of the Royal Society of South Africa 30:271-409. LEROI-GOURHAN, A. 1957. Note sur les pos- sibilites qu'apporte l'analyse pollinique aux etudes climatologiques en Afrique du Nord. Bulletin de la Socidte^palynologique Franfaise 54: 524-25. [HGc ] LIVINGSTONE, FRANK B. 1958. Anthropological implications of sickle cell gene distribution in West Africa. American Anthropologist 60: 533-62 . [JM W ] LLOYD, S. and F. SAFAR. 1945. Tell Hassuna. Journal of Near Eastern Stuidies 4:295-89. [BB*] LORET, V. 1892. La flore pharaonique. Paris. McBURNEY, C. B. M. 1960. The Stone Age of North Africa. London: Penguin Books. -- -. 1967. The Haua Fteah (Cyrenaica) anzd the Stone Age of the southeast Mediterranean. Cambridge University Press. [KWB ] McBURNEY, C. B. M., and R. W. HEY. 1955. Prehistory and Pleistocene geology of Cyrenaican Libya. Cambridge University Press. MACCALMAN, H. R. 1962. A further stone bowl from South West Africa. South African Archaeological Bulletin 17:237. - - -. 1965. Carbon-14 dates from South West Africa. Souith African Archaeological Bulletin 20:215. MACCALMAN, H. R., and B. J. GROBBELAAR. 1965. Preliminary report of two stone- working Ova-Tjimba groups in the Northern Kaokoveld of South West Africa. Cimbebasia 13:1-39. McMASTER, D. N. 1963. Speculations on the coming of the banana to Uganda. Uganda Journal 27: 163-75. MAcNEISH, R. S. 1964. Ancient Meso- american civilization. Science 143:531. MANGELSDORF, P. C. 1958. Reconstructing the ancestor of corn. American Philosophical Society Proceedings 102:454-63. MASON, R. J. 1962. The prehistory of the Trans- vaal. Johannesburg: University of the Wi- watersrand Press. - - -. 1965. The origin of South African society. South African Journal of Science 66. MASON, R. J., and N. J. VAN DER MERWE. 1964. Radio-carbon dating of Iron Age sites in the Southern Transvaal: Melville Koppies and Uitkomst Cave. South African Journat of Science 60: 142. MAUNY, R. 1952. Essai sur l'histoire des metaux en Afrique Occidentale. Bulletin de l'Institutfranfais d'Afrique noire 14:545-95. - - -. 1953. Notes historiques autour des principales plantes cultivees d'Afrique occidentale. Bulletin de l'Institut franfais d'Afrique noire 15:684-730. MERRILL, E. D. 1954. The biology of Cook's voyage. Chronica Botanica 14 (5-6). MIEGE, J. 1952. L'importance economique des ignames en Cote d'Ivoire. Repartition des cultures et principales varieties. Revue internationale de Botanique Appliquee et d'Agri- culture Tropicale 32:142-55. - - -. 1954. Les cultures vivrieres en Afrique occidentale. Cahiers d'Outremer 7:25-50. MIEGE, J., and P. ASSEMIEN. 1964. Biblio- graphie sur la palynologie (pollens actuels et fossiles) et les problemes de paleobotanique en Afrique occidentale et au Sahara. Association senegalaisepour l'etudes du Quaternaire de l'Ouestafricain 4: 8-10. [HG*] MIRACLE, M. P. 1965. Introduction and spread of maize in Africa. Journal qf African History 6:39-55. MONOD, T. 1964. "The late Tertiary and Pleistocene in the Sahara," in African ecology and human evolution. Edited by F. C. Howell and F. Bourliere. London: Methuen. MORGAN, W. B. 1962. The forest and agri- culture in West Africa. Journal of African Hi^story 3:235-39. MORI, F. 1965. Tadrart Acacus. Torino: G. Einaudi. [KWB*] Vol. 9 * No. 5 * December 1968 507This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

MORI, F., and A. ASCENZI. 1959. La mum infantile di Uan Muhuggiag. Rivista Antropologia, Istituto Italiano di Antropologi Roma 46:125-48. MORTELMANS, G. 1947. A propos de quelqu pierres percees remarquables du KatanE central. Sociite belge d'Anthropologie et Pu!histoire, Bulletin 58:151-71. MUNSON, PATRICK J. 1966. An annotate bibliography of the literature pertaining the Neolithic archaeology of West Afric the Soudan and the western Sahara. M: University of Illinois. [JM i -- -. 1968. Recent archaeological research i the Dhar Tichitt region of south-centr Mauretania. West African Archaeologic, Newsletter 10. In press. [TS fi MURDOCK, G. P. 1959. Africa: Its peoples at their culture history. New York: McGraw-Hil MYERS, L. 1958. New skeletal remains froi the Springbok Flats. South African Joiurnn of Science 54:167-75. NAUROIS, R. DE, and F. Roux. 1965. L( mangroves d'Ai'icennia les plus septentrior ales de la Cote occidentale d'Afriqu Bulletin, Institut de la Fondation d'Afriqz Noire 27:943-54. NENQUIN, J. 1959. Dimple-based pots froi Kasai, Belgian Coingo. Man art. no. 90. -- -. 1963a. Notes on some early potter cultures in North Katanga. J7ournal African History 4:19-32. - - -. 1963b. Excavations at Sanga, 195; Tervuren: Musee royal de l'Afriqu Centrale. [BMF* NEUVILLE, R. 1934. Les dibuts de l'agricultuwe de la faucille prdhistorique eni Palestine. Extrait de Recherches, Societe Hebraique d'Ex ploration et d'Archeologie. OAKLEY, K. P. 1961. "On Man's use o fire," in Social life of Early Man. Edited b, S. L. Washburn, pp. 176-93. Viking Fun Publications in Anthropology no. 31. OKrY, G. E. 0. 1960. Indigenous Nigeriai food plants. West African Science Associatio) J_ournal 6(2): 117-2 1. OLIVER, R. 1966. The problem of the Banti expansion. J_ournal of African History 7:361- 76. OLIVER, R., and j. D. FAGE. 1962. A shor history of Africa. London: Penguin Books. OZENDA, P. 1958. Flore du Sahara septentriona et central. Paris: Centre National de Re cherches Scientifiques. [HG* PANYELLA, A. 1959. La prehistoria de Fernando Po. Instituto de Estudios africano, Archivo, 13(49):17-33. PEARCE, S. V. 1960. The appearance of iron anm its use in protohistoric Africa. London University of London Press. PEARCE, S. V., and M. POSNANSKY. 1963. The re-excavation of Nsongezi Rock- shelter, Ankole. Uganda journal 2 7: 85-94. PERCIVAL, J. 1936. Cereals of ancient Egypt and Mesopotamia. Nature 138:270-73. PERROT, J. 1962. Palestine-Syria-Cilicia," in Courses toward urban life. Edited by R. J. Braidwood and G. R. Willey. Viking Fund Publications in AnthropoJogy no. 32. PHILLIPSON, D. WV. 1965a. Zambian radio- carbon dates. Archaeologia Zambiana 5:2-4. - - -. 1965b. Archaeological investigations near Chitambo Mission. Zambia, June, pp. 26-33. - - -. 1965c. The Chitambo expedition. Archaeologia Zambiana 3:2-4. - --. 1965d. Early smoking pipes from Seb- anzi, Zambia. Arnoldia 1(40). - - -. 1968. The Early Iron Age in Zambia: Regional variants and some tentative con- clusions. j'ournal of African History 9(2): 19 1- 211. [BMFW] POLE-EVANS, I. B. 1937. "Report on vegetable remains," in Mapungubwe: Ancient Bantu civilization on the Limpopo. Edited by L. Fouche. Cambridge: Cambridge University Press. POMMERET, Y. 1966. Les houes et haches taill6es de la region de Libreville. Sociitepre- historique et protohistorique gabonaise 2(5) :9-5 109. PONS, A., and P. QUEZEL. 1957. Premiere etude palynologique de quelques paleosols sahari- ens. Travaux de l'Institut de Recherches Saharaiennes 16: 15-40. PORTERES, R. 1950. Vieilles agricultures de ]'Afrique intertropicale. L'Agronomie Tropicale 5:489-507. -- -. 1951 a. Geographie alimentaire--Ber- ceaux agricoles et migrations des plantes cultivees en Afrique intertropicale. Comptes Rendus de la Societi de Biogeographie 239: 16-2 1. -- -. 1951 b. Eleusine coracana Gaertner. Butlletin de l'Institut fianfais d'Afrique noire 13:1-78. ---. 1955a. L'introduction du mais en Afri- que. J_ournal d'agriczulture tropicale et de Botanique appliquee 2: 22 1-3 1. . 1955b. Les cereales mineures du genre Digitaria. J_ournal d'agriculture tropicale et de Botanique appliqude 2:620-75. - - -. 1958. Le millet teff de l'Abyssinie. Journal d'agriculture tropicale et de Botaniquie appliqude 5:454-63. -- -. 1959a. Les Sorghos. J_ournal d'agriculture tropicale et de Botanique appliqude 6: 68-84. -- -. 1959b. Le mals ou "bl des Indes." Journal d'agriculture tropicale et de Botanique appliquee 6:84-105. - - -. 1959c. Les Riz. J_ournal d'agricultntre tropicale et de Botanique appliqude 6: 189-233. - - -. 1959d. Les Mils a chandelle ou Mils penicillaries. Journal d'agriculture tropicale et de Botanique appliquee 6:290-302. - - -. 1959e. Les millets digitaires. Journal d'agriculture tropicale et de Botanique appliqude 6:305-9. -- -. 1962. Berceaux agricoles primaires sur le continent africain. Journal of African History 3:195-210. POSNANSKY, M. 1961 a. Pottery types from archaeological sites in East Africa. Journal of African History 2:177-98. 1961b. Iron Age in East and Central Africa-points of comparison. South African Archaeological Bulletin 16: 134-36. - - -. 1961 c. Bantu genesis. Uganda J_ournal 25:86-93. -- -. 1962. "The Neolithic cultures of East Africa." Proceedings of the Fourth Pan-African Congress for Prehistory, pp. 375-81. ---. 1966. Comment on: The Iron Age in sub-Saharan Africa, by A. J. Arkell, B. Fagan, and R. Summers. CURRENT ANTHRO- POLOGY 7:473. -- -. 1968. "Bantu Genesis-Archaeological reflexions." Journal of African History, 9, 1: 1-11. POSNANSKY, M. and G. COLE. 1963. Recent excavations at Magosi, Uganda: A pre- liminary report. Man 63:133. QUEZEL, P. 1965. La vdgetation du Sahara, du Tchad et la Maure'tanie. Geobotanica selecta 2: 333. Stuttgart: G. Fischer. [HGAc] QUEZEL, P., and C. MARTINEZ . 1962. Premiers re'sultats de l'analyse palynologique de se'diments recueillis au Sahara me'ridional a' l'occasion de la Mission Berliet Tine're'-Tchad. Paris: A.M.G. QUE~ZEL, P., and S. SANTA. 1962. Nouvelle Flore de l'Algirie L Paris: Centre National de Recherches Scientifiques. [HG*] REED, C. A. 1959. Animal domestication in the Near East. Science 130:1629. RESCH, W. F. E. 1967. Das Rind in den Felsbild- darstellungen Nordafrikas. Wiesbaden: Steiner. [WR*] REYGASSE, M. 1950. Monuments funeraires prd- islamiques de l'Afrique du Nord. Paris: A.M.G. ROBINSON, K. R. 1959. Khami ruins. Cam- bridge: University Press. -- -. 196 Ia. "Excavations on the Acropolis Hill," in Zimbabwe excavations, 1958. Edited by R. Summers, K. R. Robinson, and A. Whitty. Occasional Papers of the National Museum of Southern Rhodesia 3(23A): 159-92. -- -. 1961 b. Archaeological report: Buffalo Bend Expedition. Rhodesian Schools Explora- tion Society (Matabeleland branch), pp. 16-19. - --. 196 1c. An early Iron Age site from the Chibi District, Southern Rhodesia. South .frican Archaeological Bulletin 16: 75-102. [BMF * ] ---. 1963. Further excavations in the Iron Age deposits at the Tunnel site, Gokomere Hill, Southern Rhodesia. South African Archaeological Bulletin 18:155-71. -- -. 1964. Dombozanga Rock Shelter, Mtetengwe River, Beitbridge, Southern Rhodesia. Excavation results. Arnoldia 1(7). -- -. 1965a. The Leopard's Kopje Culture: A preliminary report on recent work. Arnoldia 1(25). -- -. 1965b. A note on Iron Age sites in the Zambezi Valley, and on the escarpment in the Sipolilo District, Southern Rhodesia. Arnoldia 1 (27). -- -. 1966a. A preliminary report on the recent archaeology of Ngonde, Northern Malawi. Journal of African History 7:169- 88. -- -. 1966b. The Leopard's Kopje culture, its position in the Iron Age of Southern Rhodesia. Soutth African Archaeological 21:5-51. -- -. 1966c. Bambata ware: Its position in the Rhodesian Iron Age in the light of recent evidence. South African Archaeological Bulletin 21:81-85. ---. 1966d. Comment on: The Iron Age in sub-Saharan Africa, by A. J. Arkell, B. Fagan, and R. Summers. CURRENT ANTHROPOLOGY 7:473-74. RONCIERE, C. DE LA. 1925. La decouverte de 1'Afrique au moyen-age. Paris. RUDNER, J. 1957. The Brandberg and its archaeological remains. Journal of the Souith West African Scientific Society 12: 7-44. - - -. 1966. Comment on: The Iron Age in sub-Saharan Africa, by A. J. Arkell, B. Fagan, and R. Summers. CURRENT ANTHROPOLOGY 7:474-75. RUHLMANN, A. 1936. Le Tenereen. Societed de Preliistoire du Maroc, Bulletin 10: 3-15. RUSSEGGER, J. 1841-48. Reisen in Europa, Asien und Afrika 1835-41. Vol. 2, part 2, Reise in Aegypten. Stuttgart. iASSOON, H. 1963. Early sources of iron in Africa. South African Archaeological Bulletin 18: 176-80. )AUER, C. 0. 1952. Agricultural origins and dispersal. New York: American Geographical Societv. 508 CURRENT ANTHROPOLOGYThis content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

Davies, Hugot and Seddon: ORIGINS OF AFRICAN AGRICULTURE SYMPOSIUM SCHARFF, A., and A. MOORTGAT. 1950 Agypten und Vorderasien imn Altertumn. Muinchen Oldenbourg. [WR*] SCHIEMANN, E. 1943. Entstehung der Kultur pflanzen. Ergebnisse der Biologie 19:409-552. [WR*] ---. 1951. New results in the history of the cultivation of cereals. Heredity 5. SCHNELL, R. 1957. Plantes alimentaires et vie agricole de l'Afrique noire. Paris: Larose. SCHOFIELD, J. F. 1940. A report on the pottery from Bambata Cave. South African journal oj Science 37:361-72. - - -. 1948. Primitive pottery. South African Archaeological Society Handbook Series, 3. SCHULZ, A. 1916. Der Enmmer des alten Aegyptens. Berichte der Deutschen Botanischen Gesell- schaft 34. SCHWEINFURTH, G. 1887. Die letzten botani- schen Entdeckungen in den Grabern Aegyptens. Engels Botanisches J7ahrbuch 8: 1-16. -- -. 1891. Agyptens auswartige Beziehungen hinsichtlich der Culturgewachse. Paper presented to the Anthropological Society of Berlin. [RH*] SC:UDDER, T. 1962. The ecology of the Gwembe Tonga. Manchester: Manchester University Press. SEDDON, J. D. 1966a. The origins of domesti- catiorn: Some developments in the last five years. South African Archaeological Bulletin 21: 101-7. - - -. 1966b. Kurrichane: A late Iron Age site in the Western Transvaal. African Studies 25:227-31. ---. 1966c. The function of the hand-axe." Man (NS) 1: 244-5. -- -. 1969. Culture, Personality and Society in Prehistory. Institute for the Study of Man in Africa, Johannesburg: In press. SEMYENOV, S. A. 1964. Prehistoric technology. London: Cory, Adams and Mackay. SHAW, THURSTAN. 1963a. Archaeology and Aigeria. Ibadan: Ibadan University Press. [TS ] -- -. 1963b. Field research in Nigerian archaeology. Journal of the Historical Society of Nigeria 30:449-64. [TS*] SHINNIE, P. L. 1960. Excavations at Bigo, 1957. Uganda Journal 24:16-28. SICARD, H. VON. 1966. Comment on: The Iron Age in sub-Saharan Africa, by A. J. Arkell, B. Fagan, and R. Summers. CURRENT ANTHROPOLOGY 7:475- 77. SILBERBAUER, G. B. 1965. Bushman survey report. Bechuanaland Government Printer. SIMMONDS, N. W. 1958. Ensete cultivation in the southern highlands of Ethiopia: A review. Tropical Agriculture 35: 302-7. - - -. 1959. Bananas. London. SrMooNs, F. J. 1965a. Some questions on the economic prehistory of Ethiopia. Journal of African History 6:1-13. - - -. 1965b. Two Ethiopian gardens. Land- scape 14: 15-20. [FCG*] SINGER, CHARLES, E. J. HOLMYARD, and A. R. HALL. 1955. A history of technology. Vol. 1. Oxford: Clarendon Press. [TS * ] SMEDS, H. 1955. The ensete planting culture of eastern Sidand, Ethiopia. Acta Geographia 13. SMOLLA, G. 1957. Prahistorische Keramik aus Ostafrika. Tribus 6: 35-59. SNOWDEN, J. D. 1936. The cultivated races of sorghum. London: Adlard. SOLECKT, R. I. 1964. "Zawi Chemi Shanidar, a post-Pleistocene village site in northern Iraq." Report of the 6th International Congress on the Quaternary, Warsaw, 1961, vol. 4, pp. 405-12. Lodz. [BB*] SOLMs-LAUBACH, H. 1899. Weizen und Tulpe und deren Geschichte. Leipzig: Felix. [FCG*] STIEHLER, W. 1948. Studien zur Landwirt- schaft- und Siedlungsgeographie Aethio- piens. Erdkunde. 2:257-82. [FCG ] STREUVER, S. 1962. Implications of vegetal remains from an Illinois Hopewell site. American Antiquity 27:584-87. [JM*] -- -. n.d. Flotation: A method for recovery of small-scale archaeological remains. Ms. Northwestern University. [JM*] SUMMERS, R. 1950. Iron Age cultures in Southern Rhodesia. South African Journal of Science 47:231-38. ---. 1958. Inyanga. Cambridge: Cambridge University Press. -- -. Editor. 1959. The rock art of the Federation of Rhodesia and Nyasaland. London: Chatto and Windus. -- -. 1960. Environment and culture in Southern Rhodesia. Proceedings qf the American Philosophical Society 104:266-92. - - -. 196 1. Zimbabwe: A Rhodesian mystery. Cape Town: Nelson. - - -. 1966. The Iron Age of Southern Rhodesia. CURRENT ANTHROPOLOGY 7:463- 68. -- -. 1967. "Iron Age industries of southern Africa, with notes on their chronology, terminology and economic status," in Back- ground to African Evolution. Edited by Walter W. Bishop and J. Desmond Clark, pp. 687- 700. Chicago: University Press. [BMF * ] SUMMERS, R., K. R. ROBINSON, and A. WHITTY. 1961. Zimbabwe excavations, 1958. Occasional Papers of the National Museum of Southern Rhodesia 3(23A). SUTTON, J. E. G. 1964a. Annual report 1963- 64. British Institute of History and Archaeology, pp. 10-14. -- -. 1964b. A review of pottery from the Kenya Highlands. South African Archae- ological Bulletin 19:27-35. TACKHOLM, V., with M. DRAR. 1950. Flora of Egypr 2. Bulletin of the Faculty of Science, Cairo University, no. 28. [RH* ] - --. 1954. Flora of Egypt 3. Bulletin of the Faculty of Science, Cairo University, no. 30. [KWB * ] TIXIER, J. 1960. Examen en laboratoire de la "Faucille n? 2" de Columnata. Libyca 8: 253-58. TRAPNELL, C. G., and J. N. CLOTHIER. 1957. Revised edition. The soils, vegetation and agricultural systems of Northwestern Rhodesia. Lusaka. VAN CAMPO, M. 1958-65. Palynologie africaine. 6 vols. Senegal: Institut frangais de l'Afrique noire. [HG*c] - - -. 1964. Representation graphique des spectres polliniques des regions saharien- nes. Comptes Rendus de l'Acade'mie des Sciences 258:1873-76. [HG*] VAN CAMPO, M., and R. CoQuE. 1960. Palyno- logie et geomorphologie dans le Sud Tunisien. Pollen et Spores 2:275-84. [HG*r] VAUFREY, R. 1947. Le Neolithique para- tumbien: une civilisation agricole primitive du Sudan. La Revue Scientifique 3:205- 32. VAVILov, H. I. 1931. The problem of the world's agriculture in the light of the latest investigations. London. - - -. 1935. Bases theoriques de la selection des plantes. 1: Silection generale. Moscow-Lenin- grad. - - -. 1949-50. The origin, variation, im- munity and breeding of cultivated plants. Chronica Botanica 13: 1-364. VEDDER, H. 1937. South West Africa in early times. London. WAINWRIGHT, G. A. 1945. Iron in the Matapan and Meroitic Ages. Sudan Notes and Records 26, part 1: 5-36. - - -. 1954. The diffusion of "Uma" as a name for iron. Uganda Journal 18:113-36. WALLERT, I. 1962. Die Palme im alten Agypten. Berlin. [RH ] WAYLAND, E. W. 1954. Outline of prehistory and Stone Age climatology in the Bechuana- land Protectorate. Academie royale des Sciences coloniales, Section des Sciences naturelles medicales, Memoires 25:203-33. WELLS, L. H. 1935. Remains from a grave in the Klein Letaba district. South African Journal of Science 32:625-32. WERTH, E. 1954. Grabstock, Hacke und Pflug. Ludwigsburg: Ulmer. [WR*] WHITTY, A. 1959. Classification of stone buildings in Mashonaland. South African Archaeological Bulletin 14:57-71. WILLETT, F., 1959. Recent excavations at Old Oyo and Ife. Man art no. 135. - - -. 1962. The introduction of maize into West Africa. Africa 32:1-13. WISSMANN, H. VON 1957. Ursprungsherde und Ausbreitungswege von Pflanzen- und Tier- zucht und ihre Abhangigkeit von der Klimageschichte. Erdkunde 11:81-94, 175- 93. [WR*] ZHUKOVSKY, P. M. 1958. New centres of the origin and new gene centres of cultivated species. Botanical journal (inRussian). 53: 430- 68. [HG*] ZINDEREN BAKKER, G. M. VAN. 1951 a. Archaeology and palynology. South African Archaeological Bulletin 6(23) :1-8. [HG*] -- -. 1951 b. Palynology in Africa, Reports. 8 vols. Bloemfontein. [HG*] - -. 1963. Symposium on early man and environments in South Africa. Palaeo- botanical studies. South African Journal of Science 59: 332-40. [HG*] - - -. 1965. Ueber Moorvegetation und den Aufbau der Moore in Siud-und Ostafrika (On bog vegetation and structure of bogs in Africa). BotanischesJ7ahrbuch 84:215-38. [HG*] ZINDEREN BAKKER, G. M. VAN, and J. A. COETZEE. 1951-53. South African pollen grains and spores. 3 vols. Cape Town. [HG*] Vol. 9 * No. 5 December 1968 509This content downloaded from 146.230.187.153 on Fri, 26 Apr 2019 10:22:17 UTCAll use subject to https://about.jstor.org/terms

4. Permitted Uses of the Content 4.1 Institutional Licensees and/or Authorized Users may search, view, reproduce, display, download, print, perform, and distribute Licensed Content for the following Permitted Uses, provided they abide by the restrictions in this Section 4, Section 5 and elsewhere in these Terms and Conditions of Use: (a) research activities; (b) classroom or organizational instruction and related classroom or organizational activities; (c) student assignments; (d) as part of a scholarly, cultural, educational or organizational presentation or workshop, if such use conforms to the customary and usual practice in the field; (e) authors or other Content creators may at their discretion incorporate their Content (other than Books) into unrestricted databases or website with prior permission from the publisher and other applicable rights holders; (f) linking (g) Data for Research as defined further in Section 14.3 below. and for Institutional Licensees and/or Authorized Users other than Register & Read Users: (h) on an ad hoc basis and without commercial gain or in a manner that would substitute for direct access to the Content via services offered by JSTOR, sharing discrete portions of Content for purposes of collaboration, comment, or the scholarly exchange of ideas; (i) downloading Content (including certain Book chapters where downloading is made available for the Book chapters on the JSTOR Platform) to view, use and display including on a personal digital device (e.g. mobile devices, e-readers, and personal computers); (j) (except for Books) in research papers or dissertations, including reproductions of the dissertations, provided such reproductions are only for personal use, library deposit, and/or use solely within the institution(s) with which the Authorized User and/or his or her faculty readers are affiliated; and (k) fair use under Section 107 of the U.S. Copyright Act, educational exceptions, or other similar provisions of the copyright laws or other intellectual property right laws in the United States. 4.2 The portions of Licensed Content that have been downloaded or printed out by an Authorized User may continue to be used in compliance with these Terms and Conditions even if such license should later terminate or expire. 4.3 Additional Permitted Uses of the Early Journal Content. JSTOR encourages broad use of the Early Journal Content. In addition to all permitted uses described above, users are free to copy, use, and redistribute the Early Journal Content in part or in whole for non-commercial purposes. JSTOR asks that you acknowledge JSTOR as the source of the Early Journal Content; if you use material from JSTOR online, we request that you link directly to the stable URL provided. If you use Early Journal Content offline, we ask that you credit the source as follows: “Courtesy of JSTOR.” Please be considerate of other users and do not use robots or other devices or coordinate activity to systematically download these works as this may be disruptive to our systems. 4.4 Use of Open Access Content. Open Access Content may be subject to different terms of use, such as a Creative Commons license, as indicated in the copyright statement for the content. If an Open Access Content item has different terms posted where the content appears, those terms shall have precedence over these Terms and Conditions of Use.

Nelisiwe Ndlovu

04/26/2019 - 12:32