Early domesticated cowpea (Vigna unguiculata) from Central Ghana.
Keywords: Ghana, Kintampo, domestication, cowpea, pearl millet, oil palm, subsistence
Recent research has raised new issues on the nature of subsistence activities practised by the Kintampo, a Later Stone Age (LSA) tradition of West Africa dating to 3600-3200 BP (Casey 2005; Gautier & Van Neer 2005; Watson 2005; D'Andrea et al. 2006). The relevant sites for the most part are found in present-day Ghana, distributed from southern coastal regions to the far north (Figure 1). In addition to ceramics, Kintampo peoples possessed a material culture that included daub architecture, rasps, microliths, chipped and ground projectile points, ground stone axes, grinding stones and objets de parure (i.e. bracelets, pendants and quartz beads) (Watson 2005). Settlement patterns are quite varied, ranging from semi-sedentary villages to rock shelters occupied on a temporary basis (Davies 1962; Flight 1968; Dombrowski 1980; Stahl 1985a & b; Casey 1993; 2000; Watson 2005). Kintampo subsistence has been of interest to many scholars since the early 1960s (e.g. Davies 1962; Flight 1976; Posnansky 1984; Andah 1993; Anquandah 1993; Stahl 1993; Casey 2000; 2005; Watson 2005). Studies to date have demonstrated that a wide array of wild animal and plant resources were utilised by the Kintampo, many of which continued to be procured by later populations until recent times (Carter & Flight 1972: 278; Rahtz & Flight 1974: 28; Stahl 1985b: 138-41; Casey 1993: 85-86; 2000: 33-37; Gautier & Van Neer 2005: 203-6).
Kintampo groups were also in possession of domesticated species including pearl millet (D'Andrea et al. 2001), a forerunner of N'dama cattle and caprines (Carter & Flight 1972: 278; Stahl 1985b: 138-40; Gautier & Van Neer 2005: 202; Watson 2005: 25). Furthermore, specialised techniques, involving arboricultural practices have been suggested for the exploitation of oil palm by Kintampo peoples of south-central Ghana (D'Andrea et al. 2006). Here we report the identification of an additional domesticated species, the cowpea, surviving in the form of charred remains recovered from rock shelters at the 'B-sites' (Boase-sites) (Figure 1) in Central Ghana. We then discuss the implications of these new finds for interpretations of prehistoric West African subsistence during the fourth millennium BP.
[FIGURE 1 OMITTED]
Archaeobotanical research at the B-sites
The B-sites comprise six rock shelter loci excavated under the auspices of the Kintampo Archaeological Research Project (KARP), which began in 1998 (Watson & Woodhouse 2001; Watson 2003; 2005). They are situated in the Buokem Hills, approximately 27km south-west of the modern village of Kintampo in Central Ghana (Figure 1) (Watson 2005: 7-10). The local habitat consists of semi-deciduous forest/savanna ecotone, similar to that in which the nearby rock shelters K1 and K6 are located (Flight 1968; 1976; Rahtz & Flight 1974; Stahl 1985b). The B-sites probably represent a single multi-component site occupied at various rimes by Kintampo and another ceramic LSA population known as the Punpun (Watson 2005: 10). Details of stratigraphy, excavation methods and flotation sampling procedures are summarised elsewhere (Watson 2005: 7-10; D'Andrea et al. 2006).
The B-sites have been intensively sampled for archaeobotanical remains (Watson 2005: 10; D'Andrea et al. 2006; here Table 1). A total of 40 flotation samples was examined which produced five identified taxa including oil palm endocarp and kernels (E. guineensis), incense tree endocarp (C. schweinfurthii), cowpea (Vigna unguiculata var. unguiculata), seeds belonging to the pea family (Fabaceae), and pearl millet (Pennisetum glaucum) (the term 'seeds' here includes a number of botanical elements such as cotyledons, fruits, pericarp, etc.). Other categories include unidentifiable seeds, probable oil palm endocarp (cf. E. guineensis), probable incense tree endocarp (cf. C. schweinfurthii), unidentifiable endocarp and unidentifiable plant remains. These archaeobotanical data indicate that the loci may represent, in part, specialised sites for the production of palm oil (D'Andrea et al. 2006). One cowpea specimen from B6B, Horizon 4 was submitted for radiocarbon dating by AMS (Accelerator Mass Spectrometry), with a resulting date of 3410 [+ or -] 60 bp, 1830-1595 cal BC at 95.5% c.i. (TO 11883). This date confirms a Kintampo association and is consistent with other AMS determinations obtained from B6B (Table 2).
[FIGURE 2 OMMITTED]
B-Sites cowpea remains
A total of 16 fragmentary specimens identified as domesticated cowpea (V. unguiculata var. unguiculata) is present in the B-sites flotation samples. An additional 11 specimens have been identified as probable (cf.) cowpea (Table 1). Illustrations of Bsites cowpeas are presented in Figures 2A-C and 3A & B. Unfortunately only three fragile specimens were complete, making it possible to obtain no more than minimal metrical data. All three specimens fall into the same ranges as domesticated cowpeas identified from archaeological sites in Burkina Faso (Table 3) (Kahlheber 2004). The B-sites specimens display several features indicative of domesticated cowpea (Figure 4A). They lack lateral compression present in some accessions of the wild progenitor, V. unguiculata var. spontanea (Figure 4B). Instead charred cotelydons are asymmetric in outline and orate in cross-section. The hilum is broad and relatively large (unlike those of Sesbania, Tephrosia and Rhynchosia), located at or near the centre of the seed (unlike many wild Vigna species), and situated very dose to the chalaza (Figure 5). The ventral side tends to be flattened and the dorsal side extended, resulting in lower B/H (breadth/height) ratios in comparison to wild V. unguiculata. Other possible Fabaceae candidates native to tropical forest margin areas have been excluded based on differences in seed size and shape, including Bambara ground nut (Vigna subterranea) and Kersting's ground nut (Macrotyloma geocarpum) (Figure 4C & D). Another feature that appears to be characteristic of charred cowpea is wrinkles present on the inner surface of cotyledons, as illustrated in Figure 2C; however the diagnostic value of this criterion is unknown (Kahlheber 2004).
[FIGURE 3 OMITTED]
Cowpea ecology and domestication history
Cowpea is a widely grown crop in the semi-arid and sub-humid zones of Africa and Asia. It is the most significant legume crop grown in Africa today, particularly in West and Central Africa which typically dominate world production (Faris 1965: 433; Rawal 1975; Lambot 2002: 368). Cowpea constitutes a major dietary source of protein for many sub-Saharan populations. The cotyledons contain approximately 22-27 per cent protein and provide a well-balanced complement of essential amino acids (Busson et al. 1959: 13-14; Lambot 2002: 368-9). When mixed with cereals, protein quality is significantly improved. Leaves are an excellent source of beta-carotene, ascorbic acid and folacin (Johnson & Raymond 1964: 71-2; Bressani 1985: 356-7; Burkill 1995: 479; Nielsen et al. 1997: 328-9). Cowpeas are consumed by humans and livestock in a variety of ways. In human diets, the seeds are eaten after boiling or roasting and they can be ground into flour to make breads. Green pods are consumed fresh and the leaves are grown as a nutritious potherb or are dried and made into soups. Various parts of the plant are used medicinally, and in making dyes and textiles (Hebbs & Hill 1946; Burkill 1995: 479). As a green fodder, cowpea is particularly advantageous because individual plants continue to grow and produce leaves even after they have been cut back several times (Irvine 1956: 35; Rachie & Roberts 1974: 61; Westphal 1974: 231; Burkill 1995: 479).
[FIGURE 4 OMITTED]
Cowpea is primarily a savanna species, highly adapted to marginal environments and depauperate field conditions where other crops do not perform well. It is a dependable food security crop in semiarid and forest-margin regions of tropical West and Central Africa (Rachie & Roberts 1974: 45; Westphal 1974: 213; Skerman et al. 1988: 479; Lambot 2002: 368). Early maturing varieties can produce a harvest in as little as two or three months with green pods ready after 50 days (Purseglove 1968: 326; Rachie & Roberts 1974: 45). Sizeable yields can be achieved without the use of fertilizers and with minimal tending (Skerman et al. 1988: 479-80). Rapid flowering may represent an adaptation to short and variable growing seasons of the northern savannas. Cowpea is commonly intercropped with pearl millet and sorghum (Sorghum bicolor) and also cultivated in pure stands (Rachie & Roberts 1974: 45; Purseglove 1968: 326; Westphal 1974: 230; Lush & Evans 1981: 583).
[FIGURE 5 OMITTED]
Ng (1995: 329) has proposed that humans initially collected wild cowpea as fodder for livestock. He posits that early cattle-herders gathered entire plants by uprooting, an activity which would have created selection pressure for indehiscent pods. Cowpeas were eventually planted for human consumption, at which time people selected for non-shattering pods, reduced seed dormancy and larger seeds and pods. If cowpea resembles other pulses, however, increases in seed and pod sizes may have been delayed (Ladizinsky 1987: 61-4; Fuller & Harvey 2006: 256-7). Short-day cultivars (needing less than 12 hours of daylight in which to flower) probably evolved first in the savannas and were adapted to cereal farming systems. These types are normally found in higher latitudes where maturation occurs toward the end of the rainy season, allowing the plant to take full advantage of available moisture during flowering (Rachie & Roberts 1974: 55; Ng 1995: 329). Day-neutral cultivars evolved later through introgression with wild subspecies present in the humid tropics, and were incorporated into yam-based farming systems (Steele & Mehra 1980: 398; Lush & Evans 1981: 585; Ng 1995: 329). The scenario proposed by Ng (1995) may have some merit, given that livestock were domesticated several thousand years before crops in Africa (Marshall & Hildebrand 2002). However, to date there is no archaeological evidence for domesticated cowpea among early agro-pastoralist populations in the savanna/Sahel regions, including at the site of Birimi which was extensively sampled for macrobotanical remains (D'Andrea & Casey 2002). Furthermore it may be less likely that Kintampo groups were growing cowpea as a fodder based on zooarchaeological data which indicate that livestock were not present in significant numbers (Casey 2005; Gautier & Van Neer 2005).
The domestication history of cowpea remains to be fully elucidated. Complicating factors include the enormous variability of morphological features in both wild and domesticated forms and the wide geographical distribution of the wild progenitor. Hybridisation between cultivated and wild types has also contributed to increasing diversification (Pasquet 1996; 1999; 2000; Coulibaly et al. 2002; Ba et al. 2004). This degree of variation has resulted in the publication of several taxonomic treatments (e.g. Verdcourt 1970; Rawal 1975: 700; Mar,chal et al. 1978; Lush & Evans 1981; Ng & Mar,chal 1985; Pasquet 1993a & b; 1997; 1999; Coulibaly et al. 2002). The absence of the wild progenitor outside Africa indicates that cowpea was domesticated somewhere on the African continent, and several domestication locales have been proposed (Murdock 1959; Faris 1965: 449; Sauer 1969; Vaillancourt & Weeden 1992:1194). Phylogenetic studies and linguistic evidence indicate that the crop originated in tropical West Africa (Faris 1965; Rawal 1975: 706; Mar,chal et al. 1978: 191-6; Ng & Mar,chal 1985; Vaillancourt & Weeden 1992:1198; Panella et al. 1993: 384; Blench 1994-95: 92; Ng 1995: 329), with a second possible centre of origin in Botswana (Panella et al. 1993: 383-4). Recent studies using amplified fragment length polymorphism analysis (AFLP) have concluded that the two most likely areas of domestication are tropical West or north-eastern Africa (Coulibaly et al. 2002: 365). This work and a subsequent study using random amplified polymorphic DNA (RAPD) has ruled out the possibility of domestication in southern and eastern Africa (Ba et al. 2004: 547). A northern or north-eastern origin has been argued based on the absence of true ecologically wild cowpea in West Africa (Pasquet & Baudoin 2001: 187-88; Coulibaly et al. 2002: 365). However, the paucity of available accessions of the wild progenitor from Sudan, Eritrea and Ethiopia (Pasquet 1999:1118), as well as questions about the efficacy of AFLP and RAPD methods to infer crop origins (Allaby & Brown 2003; 2004; Salamini et al. 2004) have precluded a final assessment. More data are necessary, therefore, to demonstrate a specific area of origin for cowpea (Panella et al. 1993: 384; Pasquet 1999: 1118; Coulibaly et al. 2002: 365; Ba et al. 2004: 547). In any case, it has been determined that a single domestication event occurred in a circumscribed region (Panella & Gepts 1992; Pasquet 1999; Coulibaly et al. 2002: 364; Ba et al. 2004: 547).
The B-sites AMS date of 1830-1595 cal BC represents the earliest archaeological evidence to date of cowpea in Africa, and as such, this determination lends some support to a West African origin for this crop. It is difficult, however, to build a strong case based on a single radiocarbon age estimate. The B-sites specimens predate the earliest securely dated cowpea in India at 1700-1500 cal BC (Fuller 2003: 242-3). This points to an early and rapid spread of cowpea to India. However, additional genetic as well as baseline archaeological data are required to determine the location and timing of cowpea domestication in Africa.
Cowpea and Kintampo subsistence
Cowpeas may also be present at K6, where they are described as being very common in Kintampo deposits. Based on their abundance, Flight (1976: 217-18) argues that they were cultivated, but there was insufficient evidence to determine whether seeds came from domesticated plants. He further contends that the small size of the K6 cowpeas was the result of shrinkage due to charring (Flight 1976: 218). In later excavations at K6, Stahl (1985b: 141; 1994: 76) recovered four legume seeds which bore a resemblance to cowpea, but at roughly 5mm in length they were judged to be smaller than available reference specimens. As a result, they were conservatively identified to the family Leguminosae (=Fabaceae). Recent work has shown that a length of 5mm falls within the range of domesticated cowpea land races (Kahlheber 2004). In any case, low numbers of cowpeas in Kintampo archaeological contexts may not necessarily reflect their economic insignificance. Pulses are often under-represented in archaeological sites because they can be highly susceptible to deterioration by breaking down into two cotyledons (Gasser & Adams 1981: 183-4).
It has recently been proposed that the Kintampo people were derived from Sahelian foragers who were acculturated into an agro-pastoral economy during a long period of coexistence of forager-fishers and pastoralists (Watson 2005: 28-32). Modifications of subsistence regimes probably took place as they occupied more humid zones. This would have included the adoption of forest resources such as oil palm and incense tree (D'Andrea et al. 2006), and several species of wild fauna. North of the B-sites at Birimi, the cultivation of pearl millet among Kintampo occupants differed from those of agro-pastoral peoples present further north in the Sahel (Klee et al. 2000: 232; Klee & Zach 1999: 87; Neumann 1999: 78-9; Vogelsang et al. 1999: 60-1; Kahlheber & Neumann, in press). Late-maturing forms of pearl millet were selected from a secondary centre of diversity that developed around Lake Chad (Tostain 1998:481). These forms, which are highly adapted to savanna habitats, were probably introduced to the region near Birimi during an arid phase when the crop was displaced south of the Sahel (D'Andrea et al. 2002: 160-2). Although the area of cowpea origin remains to be determined, if initially domesticated in a savanna habitat, and day-neutral cultivars evolved as the crop was introduced to the forest zone (Ng 1995: 329), the B-sites cowpeas may have arrived in central Ghana as a result of this process. Furthermore these introductions may have taken place over a similar time frame, with the earliest known pearl millet at Birimi dating to 3460 [+ or -] 200 BP (D'Andrea et al. 2001: 343) and the B-sites cowpeas to 3410 [+ or -] 60 BE However, additional ANIS dates of Kintampo pearl millet and cowpea are necessary to determine whether this was a widespread phenomenon.
Subsistence links to savanna, Sahel and northern regions are evident in domesticated species identified in Kintampo contexts, including cattle, caprines, pearl millet and cowpea. Both pearl millet and cowpea are early-maturing, arid-adapted crops that tolerate poor environments, and are able to produce respectable harvests with minimal tending (Skerman et al. 1988: 479-80). Such crops would seem to be particularly well-suited to populations with higher mobility, such as agro-pastoralists occupying savanna regions following the onset of aridification c. 4500 BP, who may have been looking to minimise risk and establish predictable food supplies (Gifford-Gonzalez 2005: 204-6; Marshall & Hildebrand 2002). Biological indicators at Kintampo sites that demonstrate the use of locally available forest resources are all, strictly speaking, wild species, including land snails, freshwater fish, reptiles, large rodents, antelopes, primates, oil palm, incense tree and Celtis (Stahl 1985b: 139-41; Gautier & Van Neer 2005:196-202; D'Andrea et al. 2006). The available archaeobotanical and zooarchaeological evidence attests to a heterogeneous origin of foods procured and produced by the Kintampo, representing a mosaic of effective solutions taken to obtain both wild and domesticated resources.
Farmers or foragers?
How is one to characterise such a wide-ranging set of subsistence strategies evidently practised by Kintampo peoples? Recently, investigators have drawn attention to the diversified nature of the subsistence economy (Casey 2005: 236-9; Watson 2005: 30-1; D'Andrea et al. 2006), and some have argued that it is not possible to define the Kintampo and other African prehistoric societies (Casey 2005: 240; Neumann 2005: 249) as strictly 'foragers' or 'farmers'. This is a situation that is not unique to Africa, which, like many other regions of the world, provides unambiguous examples of prehistoric foraging and farming societies, along with cultures that do not fit neatly into such categories (cf. Smith 2001; Zeder 2006). The existing evidence suggests that the Kintampo economy, at the very least, can be characterised by 'low level food production with domesticates' (Smith 2001: 14-17); however it would seem premature at this stage to arrive ata conclusion on the economic importance of wild vs. domesticated resources. Despite new data available on species utilised by the Kintampo, overall only a handful of sites has been systematically sampled for plant and animal remains. Archaeobotanical studies have been completed at only three sites (Stahl 1985 a: 141; D'Andrea et al. 2001; 2006; D'Andrea & Casey 2002), while zooarchaeological interpretations (Stahl 1985a: 210-4; Gautier & Van Neer 2005: 201-2) are based on very low MNIs (Minimum Number of Individuals) of both domesticated and wild animals (Stahl 1985a: Table 4.12). Although the agro-pastoral component of the economy is now known to have been significantly more complex than was previously understood, at this point in time there appears to be no firm basis upon which to conclude that either farming or foraging was the dominant subsistence activity practised by the Kintampo.
Laboratory analysis and AMS determinations completed on the B-sites macro-remains were funded by a Simon Fraser University Institutional Grant to ACD. Caroline Lefebvre and Meridith Sayre (Simon Fraser University) assisted in the sorting of B-sites samples, through financial support provided by the Simon Fraser University Work Study Programme. Ali drawings were completed by Barbara Voss (University of Frankfurt). We benefited immensely from several informative discussions about cowpeas with Dr Don Faris (formerly of the Research Branch, Agriculture Canada). In addition, we are grateful to Dr Thierry Vanderborght (National Botanic Garden of Belgium, Meise) who provided Vigna reference specimens for comparative purposes. We thank Dr Laurence A. Pavlish (University of Toronto) who read and commented on various drafts of this paper and Shannon Wood (Simon Fraser University) for assistance with Figure 1. Helpful suggestions were also made by two anonymous reviewers. Dr R,my Pasquet (Institut de Recherche pour le D,veloppement, Nairobi) kindly sent us difficult-to-acquire reference specimens, and we appreciate his patience as we discussed at length the B-sites cowpeas.
ALLABY, R.G. & T.A. BROWN. 2003. AFLP Data and the Origins of Domesticated Crops. Genome 46: 448-53.
--2004. Reply to Comment by Salamini et al. on 'AFLP Data and the Origins of Domesticated Crops'. Genome 47: 621-2.
ANDAH, B.W. 1993. Identifying early farming traditions of West Africa, in T. Shaw, P. Sinclair, B. Andah & A. Okpoko (ed.) The archaeology of Africa: food, metals and towns: 240-54. London: Routledge.
ANQUANDAH, J. 1993. The Kintampo complex: a case study of early sedentism and food production in sub-Sahelian West Africa, in T. Shaw, P. Sinclair, B. Andah & A. Okpoko (ed.) The archaeology of Africa: food, metals and towns: 253-60. London: Routledge.
BA, F.S., R.S. PASQUET & P. GEPTS. 2004. Genetic Diversity in Cowpea [Vigna unguiculata (L.) Walp.] as Revealed by RAPD Markers. Genetic Resources and Crop Evolution 51: 539-50.
BLENCH, R. 1994-1995. Linguistic Evidence for Cultivated Plants in the Bantu Homeland. Azania XXIX-XXX: 83-102.
BRESSANI, R. 1985. Nutritive value of cowpea, in S.R. Singh & K.O. Rachie (ed.) Cowpea: research, production and utilization: 353-9. New York: John Wiley & Sons.
BURKILL, H.M. 1995. The Useful Plants of Tropical West Africa. Volume 3. Kew: Royal Botanic Gardens.
BUSSON, F., R. CARNIENER & B. BERGERET. 1959. Etude de la Fraction Protidique des Graines de Vigna unguiculata Walp. Qualitas Plantarum et Materiae Vegetabiles 6:11-5.
CARTER, P.L. & C. FLIGHT. 1972. A Report on the Fauna from the Sites of Ntereso and Kintampo Rock Shelter Six in Ghana: With evidence for the Practice of Animal Husbandry During the Second Millennium BC. Man 7(2): 277-82.
CASEY, J. 1993. The Kintampo Complex in Northern Ghana: Late Holocene Human Ecology on the Gambaga Escarpment. Unpublished PhD Thesis, University of Toronto.
--2000. The Kintampo Complex: The Late Holocene on the Gambaga Escarpment, Northern Ghana (BAR International Series 906, Cambridge Monographs in African Archaeology). Oxford: Archaeopress.
--2005. Holocene occupations of the forest and savanna, in A.B. Stahl (ed.) African archaeology: 225-48. Oxford: Blackwell.
COULIBALY, S., R.S. PASQUET, R. PAPA & P. GEPTS. 2002. AFLP Analysis of the Phenetic Organization and Genetic Diversity of Vigna unguiculata L. Walp. Reveals Extensive Gene Flow Between Wild and Domesticated Types. Theoretical and Applied Genetics 104: 358-66.
D'ANDREA, A.C. & J. CASEY. 2002. Pearl Millet and Kintampo Subsistence. African Archaeological Review 19(3): 147-73.
D'ANDREA, A.C., M. KLEE &J. CASEY. 2001. Archaeobotanical Evidence for Pearl Millet (Pennisetum glaucum) in Sub-Saharan West Africa. Antiquity 75: 341-8.
A.C. D Andrea (1), S. Kahlheber (2), A.L. Logan (3) & D.J. Watson (4)
(1.) Department of Archaeology, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6
(2.) J.W. Goethe-Universitat, Institut fur Archeologische Wissenschafien, Gruneburgplatz 1, D-60323 Frankfurt, Germany
(3.) Museum of Anthropology, University of Michigan, 1109 Geddes Avenue, Ann Arbor, Michigan, USA 48109
(4.) 183 Castlemilk Drive, Glasgow G45 9JT, UK
Table 1. Macrobotanical remains from 134C, B5C and 136B. Seed Count Pennisetum Vigna cf. V. glaucum unguicilata unguicculata CONTEXT LOCUS B4C Layer 1 -- -- -- Layer 2 -- -- -- Layer 3 -- -- -- B4C Total 0 0 0 LOCUS B5C Horizon 1 -- -- -- Horizon 2 -- -- -- Horizon 3 1 -- 2 B5C Total 1 0 2 LOCUS 13613 Horizon 1 -- -- -- Horizon 2 -- 1 Horizon 3 -- -- 0 Horizon 4 -- 15 6 B6B Total 0 16 16 TOTAL 1 16 11 ALL SITES Unidentifiable Total Seed CONTEXT Fabaceae Seed Fragments Count LOCUS B4C Layer 1 -- -- 0 Layer 2 -- -- 0 Layer 3 -- 1 1 B4C Total 0 1 1 LOCUS B5C Horizon 1 -- 14 14 Horizon 2 -- -- 4 Horizon 3 -- 25 28 B5C Total 0 43 46 LOCUS 13613 Horizon 1 -- -- 0 Horizon 2 5 Horizon 3 11 3 Horizon 4 6 48 B6B Total 922 56 TOTAL 9 66 103 ALL SITES Elaeis cf. E. guineensis E. guineensis guineensis CONTEXT endocarp kernels endocarp LOCUS B4C Layer 1 3.30 -- 0.85 Layer 2 2.69 -- 1.12 Layer 3 0.13 -- -- B4C Total 6.12 0.00 1.97 LOCUS B5C Horizon 1 4.87 -- 2.10 Horizon 2 29.10 -- 2.37 Horizon 3 15.22 -- 0.00 B5C Total 49.19 0.00 4.47 LOCUS 13613 Horizon 1 26.05 -- 1.31 Horizon 2 46.15 -- 2.55 Horizon 3 47.20 -- -- Horizon 4 500.70 0.18 2.75 B6B Total 620.10 0.18 6.61 TOTAL 675.41 0.18 13.05 ALL SITES Weight (g) Canarium cf. E. schweinfurthii schweinfurthii Unidentiafiable CONTEXT endocarp endocarp Endocarp LOCUS B4C Layer 1 -- -- -- Layer 2 -- -- -- Layer 3 -- -- -- B4C Total 0.00 0.00 0.00 LOCUS B5C Horizon 1 1.62 0.09 3.29 Horizon 2 0.42 0.62 7.74 Horizon 3 1.03 0.20 21.28 B5C Total 3.07 0.91 32.31 LOCUS 13613 Horizon 1 0.28 0.01 9.24 Horizon 2 0.56 -- 12.72 Horizon 3 -- -- 0.38 Horizon 4 2.21 0.38 30.12 B6B Total 3.05 0.39 52.46 TOTAL 6.12 1.30 84.77 ALL SITES Total Unidentiafiable CONTEXT Endocarp Charcoal Plant Remains LOCUS B4C Layer 1 4.15 1.99 0.07 Layer 2 3.81 10.93 -- Layer 3 0.13 11.67 0.77 B4C Total 8.09 24.59 0.84 LOCUS B5C Horizon 1 11.97 3.83 0.02 Horizon 2 40.25 2.91 0.02 Horizon 3 37.73 8.54 nil B5C Total 89.95 15.28 0.04 LOCUS 13613 Horizon 1 36.89 3.30 0.18 Horizon 2 61.98 1.68 0.28 Horizon 3 47.58 0.30 -- Horizon 4 536.34 1.26 0.35 B6B Total 682.79 6.54 0.81 TOTAL 780.83 46.41 1.69 ALL SITES Average E. guinnensis CONTEXT Density Sediment LOCUS B4C Layer 1 0.22 20.15 Layer 2 0.37 10.25 Layer 3 0.02 8.00 B4C Total -- 38.40 LOCUS B5C Horizon 1 0.09 66.19 Horizon 2 1.51 32.26 Horizon 3 0.38 67.64 B5C Total -- 166.09 LOCUS 13613 Horizon 1 2.38 14.00 Horizon 2 3.60 11.52 Horizon 3 4.44 10.65 Horizon 4 12.53 27.54 B6B Total -- 63.71 TOTAL -- 268.19 ALL SITES Table 2. Locus B6B AMS determinations. Unit Horizon Lab. No. Material 2 5 GX 28710 b.v. collagen 2 4 TO 11883 cowpea cotyledon 2 4 TO 11238 oil palm endocarp 1 2 GX 29106 charcoal Unit Depth (cm) Date (uncal. bp) 2 50 3230 [+ or -] 40 2 70 3410 [+ or -] 60 2 70 3460 [+ or -] 50 1 101 3470 [+ or -] 40 Unit cal. BC * cal. BP * 2 1610-1410 3560-3360 2 1830-1595 3780-3545 2 1885-1680 3835-3630 1 1890-1680 3840-3630 * 95.5% confidence interval. Table 3. Metrical data for West African V. unguiculata var. unguiculata. Dimensions (mm) B-sites Burkina Faso (Kahlheber 2004) Length 5.0-6.3 3.7-9.7 Breadth 3.6-4.6 2.4-6.6 Height 3.4-4.2 3.0-6.1 Length:Height 1.5-1.6 0.9-2.1 Breadth:Height 0.9-1.1 0.7-1.4 Length:Breadth 1.4-1.8 1.0-2.3
|Printer friendly Cite/link Email Feedback|
|Author:||D'Andrea, A.C.; Kahlheber, S.; Logan, A.L.; Watson, D.J.|
|Date:||Sep 1, 2007|
|Previous Article:||Ottoman bows--an assessment of draw weight, performance and tactical use.|
|Next Article:||The state of theocracy: defining an early medieval hinterland in Sri Lanka.|