2012 年 120 巻 2 号 p. 129-149
We designed observational surveys of controlled foraging trips of Baka hunter-gatherers in Cameroon to verify the ‘wild yam question’—i.e. is it possible for human beings to live without agricultural products in a tropical rainforest?—and to examine their foraging lifestyle. We observed two 20-day trips during which no agricultural or commercial food except salt and pepper could be used. The first trip was conducted by six married couples in August, the short dry season, of the year 2003, and the second one by eight married couples in October, the rainy season, 2005. The Baka cooperators obtained 22 species and 43 vernacular names of food in all during both survey periods. No cooperators lost weight from any food shortage in both seasons. Energy intake per consumption-day was estimated at 2528–2865 kilocalories in the dry season, and at 2479–2777 kilocalories in the rainy season. Providing more than 60% of estimated energy intake in both seasons, wild yam tubers proved to be an essential food to enable a foraging life in tropical rainforests. From this survey we could find no evidence that it is impossible to live independently of agriculture in a tropical rainforest although it seemed that the cooperators paid a high energy cost to secure food, especially wild yam tubers. This study implies that a Paleolithic foraging lifestyle in the African tropical rainforest was very likely, although not easy, and that Paleolithic foragers may have been the ancestors of the present ‘pygmy’ hunter-gatherers.
In the late 1980s, arguments arose among researchers of hunter-gatherer societies concerning the hypothesis of the ‘wild yam question,’ i.e. is it possible for human beings to live without agricultural products in a tropical rainforest (Headland and Bailey, 1991)? One of the main reasons for the debate is that tropical rainforests are not rich in food resources. Based on ecological surveys of the Ituri Forest, Hart and Hart (1986) questioned whether hunter-gatherers entered the interior of rainforests as both animal and botanical food resources were poorer than in its fringe or in the forest–savanna ecotone. Headland (1987) argued that human beings could not live independently of agricultural products in tropical rainforests where the availability of starchy food such as wild yam tubers was poor. According to Vincent (1984), tropical rainforests have far less biomass of tuberous plants than savanna. Mammalian biomass also is reported to be less than in drier environments (Barnes and Lahm, 1997). Another reason is the assertion of Bailey et al. that there are neither pure hunter-gatherer groups in tropical rainforests in the present day, nor evidence that they lived there in the past (Bailey et al., 1989). Furthermore, according to Headland (1997), modern-day hunter-gatherer societies have lived with long histories of trade and interaction with the dominant societies around them. In contrast, Bahuchet et al. (1991), Brosius (1991), Dwyer and Minnegal (1991), Endicott and Bellwood (1991), and Stearman (1991) countered the hypothesis based on their own fieldwork in Africa, Borneo, Papua New Guinea, the Malay Peninsula, and the Amazon in a special issue of Human Ecology on the ‘wild yam question.’ Elsewhere, from the ecological viewpoint, Colinvaux and Bush (1991) mentioned that the primary habitants in rainforests, especially the Amazon forest, were likely to have been hunter-gatherers. With regard to Africa, giving an example of the Aka hunter-gatherers, Bahuchet et al. (1991) made the point that the tropical rainforests in the west of the Republic of Central Africa were composed of diverse vegetation and in some areas the productivity of wild yam tubers was sufficient to support a local group of hunter-gatherers. In addition, Hladik and Dounias (1993) argued that the forests in southern Cameroon have diverse yam plants and their productivity could possibly support a foraging lifestyle. Emphasizing the capability of a pure foraging lifestyle in tropical rainforests, however, they did not go far enough to entirely reject the hypothesis of the ‘wild yam question.’ Since then, archaeological records, which have the potential to disprove it, have been accumulated. In the late 1990s the research team of Mercader excavated one site after another in the African tropical rainforests (Mercader et al., 2000, 2001; Mercader and Brooks, 2001; Mercader, 2002; Mercader and Martí, 2003; Mercader, 2003). They found evidence that the sites in the Ituri Forest in the east of the Democratic Republic of Congo were used sequentially from 18800 years BP to 800 years BP and the sites in the forests of southwestern Cameroon and Equatorial Guinea from 40000–30000 years BP to 9000 years BP. These forests did not become entirely savanna vegetation but they left considerable forested areas, i.e. they presented open forest environments, even in the Leopoldvillian cold climate period, and they have presented dense forest environments since 10000 years BP. Mercader’s work raised the possibility of a purely foraging lifestyle at least in African tropical rainforests. However, neither the archeological record, nor the ecological data on food resources in tropical rainforests is sufficient to establish this definitively. Nor do we have clear ideas about what enabled it or how hunter-gatherers could live independently of farmed agricultural products in tropical rain-forests. Further archeological records, ecological data of food resources, and behavioral data of foraging life in tropical rainforests are needed. And as Bailey and Headland (1991) noted, experimental studies of the costs and benefits of foraging are also necessary.
Since 1995, in order to examine the potentiality of tropical rainforests as a human habitat, we have surveyed the distribution and reserves of forest food resources, especially wild yam tubers, in the tropical rainforest of southeastern Cameroon. Our findings support Bahuchet’s view, showing that the forest of southeastern Cameroon could have enough wild yams and yam-like plants to sustain a pure foraging lifestyle in a tropical rainforest (Sato, 2001, 2006). Even if, however, yam and yam-like tubers were plentiful, additional practical and concrete data is needed on the foraging life in a tropical rainforest: how much forests can supply of the nutritional requirements of foragers, what cost must be paid for procuring them, how many people those starchy foods can feed, to what extent other food sources can contribute to foragers’ dietary needs, and so on. Furthermore pure foragers no longer exist anywhere at present (Bailey et al., 1989). In the African tropical rainforest we have only a few reports on temporary foraging trips of hunter-gatherer groups who usually subsist on self-farming or farmers’ assistance other than hunting and gathering (Kitanishi, 1995; Yasuoka, 2006). Therefore we designed observational surveys of controlled foraging trips to verify the ‘wild yam question’ and to reveal details of how they lived off the land. Gaining the cooperation of the Baka hunter-gatherers inhabiting southern Cameroon and northern Republic of Congo, we made two 20-day trips in different seasons, during which no agricultural or commercial food except salt and pepper could be used, in the forest of southeastern Cameroon.
In this paper we aim at testing the ‘wild yam question’ hypothesis, positive or negative, and at elucidating the critical factors determining whether the test is positive or negative, examining what food resources the Baka cooperators acquired and what cost they paid to get food, and, if negative, discussing whether the findings of this study can be applied to reconstruct a Stone Age foraging lifestyle in the African tropical rainforest.
All the Baka cooperators (Table 1) lived in a sedentary small settlement near a farmers’ village, Ndongo, in southeastern Cameroon. The Ndongo villagers consist of the majority Bakwele farmers and the minority of Bamileke and Hausa merchants from northern Cameroon while the Baka, about 300 people, are the largest ethnic group around Ndongo Village. Significant numbers of the Baka have now settled along a main road and keep banana fields and cacao fields, as neighboring farmers do (Sato, 1992; Kitanishi, 2003). In addition, almost all of them get some money, local wine and food by helping local farmers. Nevertheless, they customarily enter the forest to catch animals, chiefly using snares, and collect seed nuts such as wild mango seeds for one or two months a year, taking field crops, plantain bananas or cassava tubers. Thus the Baka people in this area maintain the knowledge and technology to live off the forest. The cooperators were no exception.
| Couple no. Aug. 2003 | Individual no. | Sex | Age estimated | Child. | Couple no. Oct. 2005 | Individual no. | Sex | Age estimated | Child. |
|---|---|---|---|---|---|---|---|---|---|
| DF1 | DF11 | M | 45–50 | 1 boy | RF1 | RF11 | M | 47–52 | 1 boy |
| DF12 | F | 45–50 | RF12 | F | 47–52 | ||||
| DF2 | DF21 | M | 25–30 | 1 girl | RF2 | RF21 | M | 27–32 | 1 girl |
| DF22 | F | 25–30 | 1 boy* | RF22 | F | 27–32 | 1 boy* | ||
| DF3 | DF31 | M | 35–40 | RF3 | RF31 | M | 37–42 | ||
| DF32 | F | 35–40 | RF32 | F | 37–42 | ||||
| DF4 | DF41 | M | 30–35 | 1 girl | RF4 | RF41 | M | 32–37 | |
| DF42 | F | 25–30 | RF42 | F | 27–32 | ||||
| DF5 | DF51 | M | 25–30 | RF5 | RF51 | M | 45–50 | ||
| DF52 | F | 20–25 | RF52 | F | 45–50 | ||||
| DF6 | DF61 | M | 35–40 | RF6 | RF61 | M | 25–30 | ||
| DF62 | F | 35–40 | RF62 | F | 35–40 | ||||
| RF7 | RF71 | M | 25–30 | 2 girls | |||||
| RF72 | F | 25–30 | |||||||
| RF8 | RF81 | M | 30–35 | 2 boys** | |||||
| RF82 | F | 25–30 | |||||||
DF1 and RF1, DF2 and RF2, DF3 and RF3, and DF4 and RF4 are the same families respectively.
The first trip was conducted in the dry season of August 2003 (Sato et al., 2006), and the second one in the rainy season of October 2005. The Baka cooperators spontaneously selected the same place (14°45′20″ E, and 2°12′30″ N) at the base of Mount Bek, about 600 meters above sea level, for their campsite in both trips (Figure 1). People rarely visited the forest around the campsite as it took two days on foot from Ndongo Village. The vegetation around Mount Bek is semi-deciduous forest including small marshlands with an annual rainfall of less than 1600 mm (Letouzey, 1985). In the survey area there are four seasons: a minor rainy season, April–May, a minor dry season, June–August, a major rainy season, September–November, and a major dry season, December–March. As shown in Figure 2, at Moloundou, the capital of this region, 50 kilometers east of Ndongo Village, the mean monthly rainfall in this area is 40 to <120 mm in the major dry season, about 150 mm in the minor rainy season, around 100 mm in the minor dry season, and 130 to >200 mm in the major rainy season. The mean monthly temperature is 25°C all year round. Mount Bek is rich in wild yam tubers (Sato, 2006) and the Leke River running beside the campsite was known as a treasure house of fish. Although this area was too far to be used for ordinary hunting or gathering trips, the Baka cooperators identified it as a bountiful forest and selected this place for the campsite.
Survey area.
Mean monthly rainfall in Moloundou, adapted from Sigaha-Nkamdiou (1993). Although rainfall was recorded at the station from 1933 to 1991, 48 years of data was used.
The cooperators in the first trip consisted on six married couples, including four children under 7 years old, and those in the second trip eight married couples, including seven children under 10 years old. Since the cooperators did not habitually count their age, we estimated their age except for the children whose ages we could identify. They had a wide age range from early 20s to early 50s but many of them were in the younger age group from 20s to 30s. Four among six couples in the first trip also participated in the second trip.
Controlled foraging trip and observation periodIn both trips it took three days to reach the campsite. On the day of arrival the cooperators spent a lot of time in building their traditional domical small huts called ‘mongulu’ in Baka language and providing for the foraging life. The following day the surveys started. From leaving the home settlement to the morning of the beginning day the cooperators ate agricultural food such as plantain bananas and commercial food, but from then to the morning of the final day they depended on forest food resources except for salt and pepper. During that time the research team members exclusively lived off agricultural and commercial foods. On the first trip the observation survey began on the morning of 16 August and ran through to the morning of 5 September 2003. On the second trip, six families were observed for 21 days from the morning of 2 October to the morning of 23 October 2005, while two other families were observed for 20 days from the mornings of 3 to the morning of 23 October because their participation was one day late.
Body weightWe checked the cooperators’ body weight using a digital weight scale (Tanita THD-652, Japan) every morning (6:00–6:30) before breakfast. The children’s body weight was also checked in the morning of the starting day and ending day during the first trip and every morning during the second trip.
Time spent in foraging activities and activity observationEveryday we timed every cooperator going out and returning to the camp. Coming back, all the cooperators were asked what they did in the forest. From these data we estimated the time spent in foraging activities by each cooperator. In order to know their activity pattern, the time allocation, the division of labor by sex, and so on, the direct activities of every cooperator were observed from morning (6:00) to evening (18:00) of one day, to which the observation day was randomly allotted in both trips. In this survey geographical information was also collected using a GPS receiver (Garmin eTrex, USA).
Weighing food and estimation of dietary intakeAll food brought back to the camp was identified and weighed using a digital hanging scale (Bonso model 393, HongKong), a digital kitchen scale (Tanita KD-174, Japan) and a spring scale. All the discarded rotten food during survey periods and the remnant food at the end of survey were also weighed. Animals were identified with the aid of Kingdon’s book (1997). As for the plants, we first recorded their vernacular names and then, referring to Letouzey (1976), Hamon et al. (1995), and Dumont et al. (1994), we cited their scientific names. Occasionally weighing the food before and after cooking and, in the case of game, the leftover bones after eating, we estimated the rate of edible portions for yam tubers, nuts, leaves, games’ meat, snails, and termites. Those for snakes, fish, and honey are derived from Kitanishi (1995) and those for any others are from Standard Tables of Food Composition in Japan (Ministry of Education, Culture, Sports, Science and Technology, Japan, 2005). The value of energy and protein contents for yam tubers (item number, hereafter IN: 260 and 264), nuts (IN: 446), game meat (IN: 1134 and 1148, both are African beef as there is no data on wild games.), snake (IN: 1184), fish (IN: 1263 and 1251), snails (IN: 1423), termites (IN: 1190) and honey (IN: 1060) are derived from African food composition tables (Leung, 1968), that for leaves from Mialoundama (1993), and those for birds (IN: 11240) and crabs (IN: 10335) from Standard Tables of Food Composition in Japan (Ministry of Education, Culture, Sports, Science and Technology, Japan, 2005). From above data we estimated the total energy intake of all participants in the camp and the per-capita daily energy intake (the total energy intake per consumption-day). The consumption-day was the adjusted number of participants × observation days. Participants were divided into two categories: adult persons and children (3–10 years old), one person of which was converted to 70% of an adult person based on estimated BMR (basal metabolic rate) from body weight by sex and age grade (FAO, 2001).
Energy expenditure and steps walkedA portable accelerometer (Kentz Lifecorder EX, Japan) was worn on the waist of every cooperator from morning (6:00) to evening (18:30) every day to measure the energy expenditure for foraging activities. Since the accelerometer was not made to catch the multidimensional actions of foraging activities in the forest, however, we have not used the data here. However, the accelerometer was equipped with a pedometer that counted the cooperator’s steps walked. We use the pedometer data as a method to measure the cost expended for foraging activities.
Outline of the controlled foraging tripIn both trips the cooperators used their usual tools such as machetes, spears, iron fittings fixed on the point of wooden digging sticks, which were instantly prepared at the time of digging yam tubers, wire-snares, axes, hooks and lines, fishing nets, carrying baskets, etc., for foraging activities.
During both survey periods the cooperators engaged in collecting wild yam tubers, wild honey, termites, edible fungi, nuts, etc.; trapping with wire-snares; and fishing. While honey collecting, trapping, net fishing, and hook and line fishing were done exclusively by men, and dam and bail fishing by women, collecting yam tubers, termites, nuts, and edible fungi were done by both men and women (Table 2).
| Activity | Target | Tool used | Worker | Brief discription | |
|---|---|---|---|---|---|
| male | female | ||||
| Gathering | yam | ‘ngoso,’* machete | ✓ | ✓ | Finding yam vines, they dig out yam tubers using an ‘ngoso’ or a machete. |
| nut** | machete | ✓ | ✓ | Collecting fallen seeds of Panda oleosa, they break them with a machete and remove kernels. | |
| honey | ax, machete | ✓ | Finding wild honeycomb on a tree, men cut it down with an ax and take the honey. | ||
| termite | machete | ✓ | Finding a termite hill, they dig out termites using a machete or a digging stick. | ||
| fungus | ✓ | ✓ | Fungi are normally picked during walking for other food getting activities. | ||
| Hunting | game | spear | ✓ | Whenever men walk in a forest, they carry a spear and use it on game encountered. | |
| game | wire snare | ✓ | All the snares set during research period were spring ones using wire for the loops. | ||
| Fishing | fish | hook, line, rod | ✓ | Targets of fishing using a small hook and a rod are small fishes in streams. | |
| fish | net | ✓ | Only a gill net was used during both research periods because it was expensive. | ||
| big fish | machete | ✓ | ✓ | Finding the entrance hole of fishes on the side of a stream bank, they dig into it and catch fish. | |
| fish et al. | pan, plant leaf*** | ✓ | Damming a stream with trees and clay, women bail water out of the downstream puddle and catch fish, crabs or shrimps. | ||
The common activity pattern was that each couple went out to collect wild yam tubers every two or three days and on the other days they engaged in collecting nuts, honey, termites, etc., or in fishing in the dry season. Most of the wire-snares were set around the campsite and men made the rounds of them every day before breakfast or on their way back from other foraging activities.
Usually adult men had meals together, which their wives cooked, at a meeting place called ‘mbanjo’ in the center of the campsite both in the morning and evening, whereas women and their children had meals at the front of their own huts. Normal meals in the morning were only boiled yam tubers; the evening meal consisted of yam tubers and soup of meat, fish, or fungi with nut paste. As husbands shared the food at the ‘mbanjo’ and wives often interchanged plates with each other, it was unlikely that any individuals lacked food.
Statistical analysisAll statistical analyses were done using statistical package JMP (version 8.0.2).
Research ethicBefore the two main trips, we made a preliminary 10-day trip with six married couples in the forest near their settlement in August 2001 to determine what conditions were needed to conduct each research trip safely. Based on this preliminary trip, we informed the cooperators on the details of our research and the necessary conditions: staying in the forest for about three weeks, subsisting only on forest food resources during the trip, being observed daily as to body weight, subsistence activities, meals, and so on, and being able to withdraw from the research whenever some problem occurred. We asked each of the cooperators to voluntarily participate in this study and gained their consent.
On both trips the body weight of almost all the cooperators varied little day to day. Comparing it between the first half and the last half of the survey period, most of them kept their weight consistently, or modestly increased their weight in the last half, with the exception of a few women in the dry season and one man in the rainy season who slightly lost weight in the last half (Table 3). Among them the weight loss of one woman (DF52) in the dry season and one man (RF81) in the rainy season appeared to be due to a feminine ailment and a digestive upset, respectively.
| Indiv. no. | Sex | Mean ± SD | MF | ML | BMR | t-test | Trend | Indiv. no. | Sex | Mean ± SD | MF | ML | BMR | t-test | Trend |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Aug. 2003 | |||||||||||||||
| DF11 | M | 36.0 ± 0.26 | 36.0 ± 0.32 | 36.1 ± 0.19 | 1286.1 | ns | DF12 | F | 36.4 ± 0.38 | 36.2 ± 0.45 | 36.6 ± 0.13 | 1141.4 | * | + | |
| DF21 | M | 51.6 ± 0.69 | 51.8 ± 0.82 | 51.4 ± 0.38 | 1469.1 | ns | DF22 | F | 47.7 ± 0.83 | 48.2 ± 0.72 | 47.1 ± 0.47 | 1193.4 | *** | – | |
| DF31 | M | 47.4 ± 0.35 | 47.4 ± 0.27 | 47.4 ± 0.44 | 1416.9 | ns | DF32 | F | 45.0 ± 0.59 | 45.1 ± 0.66 | 44.9 ± 0.55 | 1211.3 | ns | ||
| DF41 | M | 51.7 ± 0.54 | 51.5 ± 0.58 | 52.0 ± 0.31 | 1466.2 | * | + | DF42 | F | 39.7 ± 0.29 | 39.8 ± 0.19 | 39.5 ± 0.30 | 1074.9 | * | – |
| DF51 | M | 48.2 ± 0.51 | 48.2 ± 0.32 | 48.0 ± 0.60 | 1426.1 | ns | DF52 | F | 40.5 ± 1.10 | 41.4 ± 0.28 | 39.5 ± 0.58 | 1086.7 | *** | – | |
| DF61 | M | 47.2 ± 0.47 | 47.1 ± 0.28 | 47.3 ± 0.50 | 1414.6 | ns | DF62 | F | 35.9 ± 0.31 | 36.1 ± 0.30 | 35.8 ± 0.26 | 1137.3 | ns | ||
| Oct. ’05 | 1140.8 | ||||||||||||||
| RF11 | M | 34.1 ± 0.54 | 34.1 ± 0.72 | 34.0 ± 0.22 | 1264.3 | ns | RF12 | F | 36.5 ± 0.36 | 36.4 ± 0.45 | 36.6 ± 0.16 | 1142.2 | ns | ||
| RF21 | M | 52.7 ± 0.86 | 52.3 ± 0.92 | 53.2 ± 0.41 | 1500.8 | ** | + | RF22 | F | 47.8 ± 0.30 | 47.7 ± 0.32 | 47.9 ± 0.27 | 1194.9 | ns | |
| RF31 | M | 46.3 ± 0.48 | 46.0 ± 0.42 | 46.6 ± 0.39 | 1404.3 | ** | + | RF32 | F | 43.5 ± 0.43 | 43.3 ± 0.45 | 43.6 ± 0.39 | 1199.1 | ns | |
| RF41 | M | 51.6 ± 1.04 | 51.1 ± 1.02 | 52.3 ± 0.45 | 1476.5 | ** | + | RF42 | F | 39.2 ± 1.22 | 39.1 ± 1.69 | 39.4 ± 0.30 | 1164.1 | ns | |
| RF51 | M | 43.6 ± 0.67 | 43.1 ± 0.35 | 44.1 ± 0.52 | 1373.3 | *** | + | RF52 | F | 45.5 ± 0.64 | 45.2 ± 0.52 | 45.9 ± 0.62 | 1215.3 | * | + |
| RF61 | M | 44.6 ± 0.43 | 44.3 ± 0.42 | 44.8 ± 0.31 | 1363.7 | * | + | RF62 | F | 39.9 ± 0.79 | 39.4 ± 0.58 | 40.4 ± 0.67 | 1169.8 | ** | + |
| RF71 | M | 59.3 ± 0.81 | 58.8 ± 0.54 | 59.9 ± 0.62 | 1585.1 | *** | + | RF72 | F | 48.4 ± 0.32 | 48.3 ± 0.33 | 48.5 ± 0.29 | 1203.8 | ns | |
| RF81 | M | 48.7 ± 1.84 | 50.0 ± 1.49 | 47.2 ± 0.65 | 1431.8 | * | – | RF82 | F | 50.3 ± 0.46 | 50.4 ± 0.57 | 50.3 ± 0.29 | 1231.9 | ns | |
MF: mean weight and standard deviation during the first half of survey period.
ML: mean weight and standard deviation during the latter half of survey period.
BMR: basal metabolic rate estimated with the equations of FAO (2001: 37).
Trend: Plus indicates weight’s increasing and minus weight’s decreasing from the first half to the latter half of survey period.
Food brought back to the camp during both survey periods consisted of various food types: wild yam tubers (5 species), mammals (12 species), freshwater fish (18 vernacular names), termites (one vernacular name), honey (one vernacular name), nuts (2 species), and edible fungi (16 vernacular names). There were 22 species and 43 vernacular names of food in all (Table 4). Among these, the first ranking in weight was wild yam tubers, and then mammals followed by fish in the dry season and by nuts in the rainy season (Table 5). In both seasons yam tubers and mammals occupied more than 90% in weight composition although there were a few seasonal differences; the amount of yam tubers in the rainy season was larger than in the dry season, whereas the catches of mammals in the dry season were larger than in the rainy season. Catches of fish also were larger in the dry season. Although the lowered water level enabled large catches of fish in the dry season, the causes of large catches of mammals in the dry season were not identified but might have been attributable to the contingency in hunting. The large amount of yam tubers collected in the rainy season will be discussed later. Among the five species of wild yam tubers collected Dioscorea praehensilis (‘safa’), an annual plant, supplied the bulk of their food in both seasons. A large number of the mammals trapped were small-and-medium-sized forestry duikers weighing 5–15 kg. While a small number of large-sized catfish were found in their big catches, a great number of small fish constantly provisioned the diet of the camp, especially in the dry season. Almost all the nuts were Panda oleosa. Honey brought back to the camp represented only a portion of the total collected as the cooperators usually consumed part of the honey at the collecting place.
| Food | Aug. 2003 | Oct. 2005 | |||
|---|---|---|---|---|---|
| Total fresh | n | Total fresh | n | ||
| Scientific name or English name | Baka name | weight (kg) | weight (kg) | ||
| Wild yam | |||||
| Dioscorea praehensilis Benth. | safa | 636.6 | 1030.6 | ||
| D. semperflorens Uline | suma | 10.9 | 27.0 | ||
| D. mangenotiana Miége | ba | 7.7 | 22.4 | ||
| D. burkilliana Miége | keke | 4.8 | 12.3 | ||
| D. minutiflora Engl. | kuku | 0.7 | 0.6 | ||
| Mammal | |||||
| Cephalophus callipygus Peters | ngendi | 85.4 | 6 | 45.1 | 3 |
| C. dorsalis Gray | ngbomu | 74.3 | 4 | 132.4 | 7 |
| C. leucogaster Gray | mongala | 27.0 | 2 | ||
| C. monticola (Thünberg) | ndengbe | 5.5 | 1 | 26.8 | 6 |
| C. silvicultor (Afzelius) | bemba | 19.9 | 1 | ||
| Neotragus batesi De Winton | samba | 2.5 | 1 | ||
| Hyemoschus aquaticus (Ogilby) | akolo | 96.6 | 9 | 45.2 | 5 |
| Atherurus africanus (Gray) | boke | 3.2 | 1 | ||
| Potamochoerus porcus (L.) | famme | 45.5 | 1 | ||
| Bdeogale nigripes Pucheran | buse | 3.9 | 1 | 7.1 | 2 |
| Crossarchus obscurus F. Cuvier | ganda | 4.9 | 2 | 3.0 | |
| Felis aurata Temminck | ebbie | 5.4 | 1 | ||
| Reptile | |||||
| Bitis gabonica gabonica | buma | 3.6 | 1 | ||
| Bird | |||||
| ? | gbedi | 0.1 | 1 | ||
| ? | elúá | 0.1 | 1 | ||
| eggs of guinea hen | fafala-kanga | 0.3 | |||
| Fish | |||||
| electric catfish | gbibi | 12.8 | 8.4 | ||
| catfish | kannya | 18.1 | 3.4 | ||
| catfish | ngolo | 0.2 | |||
| carp | likamboka | 4.1 | 0.8 | ||
| characin | popo | 3.1 | |||
| characin | denge | 0.4 | 0.0 | ||
| characin | mayanga | 0.1 | |||
| characin | toshi | 0.1 | a | ||
| characin | jilelo | 1.1 | |||
| cichlid | sale | 0.3 | |||
| cichlid | toko | 0.7 | b | ||
| characin | jaseli | 4.9 | |||
| characin | monjangbe | 0.1 | |||
| elephant nose fish | mbose | 1.3 | |||
| African snakehead fish | monga | 1.4 | |||
| ? | mbongo | 0.5 | c | ||
| ? | mekpaso | d | |||
| ? | misango | 0.3 | |||
| Crustacea | |||||
| crab | kala | 0.3 | 0.1 | ||
| shrimp | kanji | 0.0 | |||
| Mollusc | |||||
| African snail | mbembe | 0.8 | 1.7 | ||
| Insect | |||||
| termite | bandi | 5.9 | 9.3 | ||
| Honey | |||||
| stingless bee | dandu | 8.0 | 3.1 | ||
| Plant seeds | |||||
| Panda oleosa Pierre | kanna | 22.9 | 44.3 | ||
| Irvingia excelsa Mildbr. | fayo | 0.1 | |||
| Plant leaves | |||||
| Gnetum spp. | koko | 0.0 | |||
| Fruit | |||||
| Anonidium manni (Oliv.) | ngbe | 12.0 | |||
| Gambeya lacourtiana (De Wild.) | bambu | 1.6 | |||
| Fungi | |||||
| ? | mondongola | 0.1 | |||
| ? | tulu-bela | 0.0 | |||
| ? | tulu-sakusa | 1.4 | |||
| ? | mawoluolu | 0.0 | |||
| ? | tulu-bongo | 2.0 | |||
| ? | tulu-kanga | 1.3 | |||
| ? | tulu-baala | 1.0 | |||
| ? | moselele | 2.0 | 0.3 | ||
| ? | yofiyo | 0.0 | |||
| ? | tulu-sakili | 0.7 | |||
| ? | deddele | 1.5 | |||
| ? | tulu-dengbe | 0.0 | |||
| ? | tulu-timi | 0.0 | |||
| ? | tulu-gbado | 0.1 | |||
| ? | moleseko | 0.4 | |||
| ? | asamoni | 1.0 | |||
a + b + c + d = 0.338 kg
| Food type | Aug. 2003 | Oct. 2005 | ||||
|---|---|---|---|---|---|---|
| Total fresh weight (kg) | % | Weight/CD (kg) | Total fresh weight (kg) | % | Weight/CD (kg) | |
| Yam tubers | 660.7 | 59.4 | 2.75 | 1092.9 | 74.7 | 3.29 |
| Seeds | 23.0 | 2.1 | 0.10 | 44.3 | 3.0 | 0.13 |
| Leaves | 0.0 | 0.0 | 0.00 | |||
| Fruit | 13.6 | 1.2 | 0.06 | |||
| Fungi | 3.0 | 0.3 | 0.01 | 8.7 | 0.6 | 0.03 |
| Mammal | 343.1 | 30.9 | 1.43 | 290.6 | 19.9 | 0.88 |
| Bird | 0.2 | 0.0 | 0.00 | |||
| Bird’s egg | 0.3 | 0.0 | 0.00 | |||
| Snake | 3.6 | 0.3 | 0.02 | |||
| Fish | 49.5 | 4.5 | 0.21 | 12.9 | 0.9 | 0.04 |
| Crustaceae | 0.3 | 0.0 | 0.00 | 0.1 | 0.0 | 0.00 |
| Snail | 0.8 | 0.1 | 0.00 | 1.7 | 0.1 | 0.01 |
| Termite | 5.9 | 0.5 | 0.02 | 9.3 | 0.6 | 0.03 |
| Honey | 8.0 | 0.7 | 0.03 | 3.1 | 0.2 | 0.01 |
| 1111.8 | 100.0 | 1463.8 | 100.0 | |||
Weight/CD: Weight per cooperator per day.
The cooperators usually collected wild yam tubers in units of married couples. Each couple left the camp with the intention to collect wild yam tubers on an average of 10 days at an interval of around a day during both surveys (Table 6). Henceforth we express the day when a couple collected wild yam tubers as the yam collecting day (YCD). The mean weight of yam tubers collected by each couple on YCD was 9–12 kg per day in the dry season and 10–16 kg per day in the rainy season (Table 7). Although the weight of yam tubers collected by each couple slightly varied from day to day, it did not correlate with the order of YCD with the exception of a couple that showed a negative correlation in the rainy season: the weight of yam tubers decreased by the day because of the husband’s sickness. Neither in the dry season nor in the rainy season was there a statistically significant difference in the weight of yam tubers collected by couple between the first half and the last half of the survey period (Table 8). This suggests that there was no evidence of gathering pressure on yam tubers, causing resource shortage, at least during the 20-day survey period. Although there was a slight seasonal difference and daily fluctuations, the cooperators consistently secured a considerable amount of yam tubers in both seasons. Other than yam tubers, there was great variability among couples in the weight of food brought back in both seasons (Table 9). It is especially noteworthy that the quantity of captured game, which was of importance in terms of food supply, ranged widely from couple to couple in both seasons. It is unlikely that the difference in the amount of acquired food among couples can be linked to the inequality of food intake, however, as not only game meat but also other foods such as boiled yam tubers were usually shared by all cooperators at meals. There was no statistically significant difference in the weight of game caught by couples between the first half and the last half of the survey period (Table 10); however, the weight of nuts in the dry season in the last half was larger than in the first half (Table 11). These things also suggest that there was no evidence of hunting and gathering pressure on game and nuts.
| Aug. 2003 | TYCD | Interval | Oct. 2005 | TYCD | Interval | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Couple no. | N | 0 | 1 | 2 | 3 | Couple no. | N | 0 | 1 | 2 | 3 |
| % | % | % | % | % | % | % | % | ||||
| DF1 | 12 | 36.4 | 63.6 | 9.1 | 0.0 | RF1 | 11 | 40.0 | 30.0 | 20.0 | 10.0 |
| DF2 | 10 | 22.2 | 55.6 | 22.2 | 0.0 | RF2 | 9 | 12.5 | 50.0 | 25.0 | 12.5 |
| DF3 | 10 | 22.2 | 55.6 | 22.2 | 0.0 | RF3 | 8 | 0.0 | 42.9 | 42.9 | 14.3 |
| DF4 | 11 | 30.0 | 60.0 | 10.0 | 0.0 | RF4 | 10 | 22.2 | 44.4 | 33.3 | 0.0 |
| DF5 | 9 | 12.5 | 50.0 | 37.5 | 0.0 | RF5 | 10 | 11.1 | 66.7 | 22.2 | 0.0 |
| DF6 | 12 | 36.4 | 63.6 | 0.0 | 0.0 | RF6 | 9 | 25.0 | 50.0 | 25.0 | 12.5 |
| RF7 | 10 | 33.3 | 33.3 | 22.2 | 11.1 | ||||||
| RF8 | 11 | 30.0 | 50.0 | 20.0 | 0.0 | ||||||
| Mean | 10.7 | 26.6 | 58.1 | 16.8 | 0.0 | 9.8 | 21.8 | 45.9 | 26.3 | 7.5 | |
TYCD: Total of the yam collecting day when a couple aimed at collecting wild yam tubers, excluding days when they collected by chance wild yam tubers on their way of other foraging activities.
Interval: interval of days between each contiguous yam collecting day.
| Couple No. | Order of the yam collecting day | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Aug. 2003 | I | II | III | IV | V | VI | VII | VIII | IX | X | XI | XII | S kg | M ± SD | S/T % | PC | Ss |
| DF1 | 10.2 | 11.2 | 6.3 | 12.3 | 6.4 | 12.3 | 11.5 | 8.6 | 4.6 | 10.5 | 4.6 | 7.9 | 106.4 | 8.9 ± 2.9 | 93.6 | −0.387 | ns |
| DF2 | 9.5 | 7.1 | 8.5 | 15.9 | 8.2 | 9.2 | 13.2 | 8.3 | 11.5 | 6.9 | 98.3 | 9.8 ± 2.9 | 99.6 | −0.005 | ns | ||
| DF3 | 7.3 | 5.7 | 11.9 | 8.6 | 13.1 | 9.3 | 11.5 | 8.7 | 2.5 | 11.3 | 89.9 | 9.0 ± 3.2 | 100.0 | 0.017 | ns | ||
| DF4 | 9.3 | 10.3 | 13.3 | 11.5 | 13.1 | 10.4 | 19.2 | 16.2 | 9.0 | 8.8 | 10.8 | 131.9 | 12.0 ± 3.3 | 99.0 | 0.041 | ns | |
| DF5 | 15.8 | 6.1 | 10.5 | 15.4 | 5.4 | 11.0 | 7.9 | 9.2 | 17.0 | 98.3 | 10.9 ± 4.3 | 97.7 | 0.063 | ns | |||
| DF6 | 9.1 | 15.5 | 15.9 | 7.5 | 8.3 | 13.9 | 9.5 | 3.5 | 15.4 | 11.5 | 3.0 | 11.5 | 124.6 | 10.4 ± 4.4 | 100.0 | −0.275 | ns |
| Oct. 2005 | |||||||||||||||||
| RF1 | 23.5 | 21.6 | 11.6 | 15.0 | 19.5 | 5.9 | 6.8 | 13.2 | 15.7 | 6.1 | 33.3 | 172.2 | 15.7 ± 8.4 | 99.7 | −0.061 | ns | |
| RF2 | 12.7 | 26.1 | 13.2 | 22.8 | 9.9 | 16.1 | 11.8 | 12.4 | 19.1 | 144.1 | 16.0 ± 5.5 | 98.5 | −0.248 | ns | |||
| RF3 | 10.1 | 7.7 | 16.6 | 6.2 | 12.8 | 8.2 | 10.9 | 18.0 | 90.5 | 11.3 ± 4.2 | 95.9 | 0.467 | ns | ||||
| RF4 | 15.7 | 7.9 | 25.0 | 6.3 | 11.3 | 16.5 | 11.2 | 10.1 | 13.4 | 11.7 | 129.1 | 12.9 ± 5.3 | 97.8 | −0.198 | ns | ||
| RF5 | 19.2 | 20.4 | 13.2 | 14.5 | 4.2 | 11.3 | 15.3 | 18.7 | 19.2 | 23.7 | 159.7 | 16.0 ± 5.6 | 99.3 | 0.248 | ns | ||
| RF6 | 11.3 | 18.0 | 7.3 | 13.8 | 5.5 | 9.7 | 4.6 | 10.8 | 9.2 | 90.2 | 10.0 ± 4.2 | 98.3 | −0.504 | ns | |||
| RF7 | 19.9 | 32.5 | 14.1 | 5.3 | 30.3 | 7.6 | 6.1 | 22.8 | 18.0 | 3.1 | 159.7 | 16.0 ± 10.5 | 92.5 | −0.432 | ns | ||
| RF8 | 21.7 | 16.6 | 11.1 | 13.9 | 16.5 | 9.1 | 6.1 | 5.5 | 15.2 | 9.4 | 8.3 | 133.4 | 12.1 ± 5.1 | 100.0 | −0.66 | * | |
S: Sum of the weight of yam tubers collected on the yam collecting days.
S/T: Rate of S to the amount of yam tubers collected during research period.
PC: Pearson’s product–moment correlation coefficient between the order of the yam collecting day and the weight of yam tubers collected on the yam collecting day.
Ss: Statistical significance (ns, not significant),
| Aug. 2003 | First | Last | Oct. 2005 | First* | Last |
|---|---|---|---|---|---|
| Couple no. | kg/day | kg/day | Couple no. | kg/day | kg/day |
| DF1 | 6.0 | 5.4 | RF1 | 8.3 | 8.1 |
| DF2 | 4.9 | 4.9 | RF2 | 7.7 | 6.2 |
| DF3 | 4.7 | 4.3 | RF3 | 3.8 | 5.3 |
| DF4 | 5.3 | 4.7 | RF4 | 7.5 | 4.9 |
| DF5 | 6.8 | 6.5 | RF5 | 6.6 | 8.8 |
| DF6 | 7.0 | 5.4 | RF6 | 5.1 | 4.1 |
| RF7 | 7.3 | 8.9 | |||
| RF8 | 8.1 | 4.5 | |||
| Wilcoxon’s test | ns | ns | |||
First: The weight per day of yam tubers collected in the first 10 days.
Last: The weight per day of yam tubers collected in the last 10 days.
| Couple no. | Yam | Game | Fish | Nuts | Termite | Fungi | Family members |
|---|---|---|---|---|---|---|---|
| Aug. 2003 | kg | kg | kg | g | g | g | |
| DF1 | 5.7 | 2.4 | 1.1 | 78.3 | 94.6 | 0.0 | 3 |
| DF2 | 4.9 | 1.3 | 0.7 | 188.6 | 133.4 | 0.0 | 4 |
| DF3 | 4.5 | 1.3 | 0.2 | 288.6 | 0.0 | 85.0 | 2 |
| DF4 | 5.0 | 5.0 | 0.3 | 189.1 | 0.0 | 50.0 | 3 |
| DF5 | 6.7 | 5.0 | 0.1 | 174.8 | 41.5 | 17.7 | 2 |
| DF6 | 6.2 | 2.4 | 0.1 | 231.9 | 0.0 | 0.0 | 2 |
| Mean | 5.5 | 2.9 | 0.4 | 191.9 | 44.9 | 25.5 | |
| SD | 0.8 | 1.7 | 0.4 | 69.5 | 57.2 | 35.1 | |
| Oct. 2005 | |||||||
| RF1 | 8.2 | 0.0 | 0.0* | 97.4 | 82.3 | 102.1 | 3 |
| RF2 | 7.0 | 3.6 | 0.6 | 325.5 | 269.6 | 60.6 | 4 |
| RF3 | 4.5 | 0.9 | 0.0 | 351.5 | 24.7 | 34.4 | 2 |
| RF4 | 6.3 | 2.4 | 0.0 | 278.3 | 11.2 | 33.8 | 2 |
| RF5 | 7.7 | 3.6 | 0.0 | 234.6 | 0.0 | 13.1 | 2 |
| RF6 | 4.6 | 0.2 | 0.0* | 167.1 | 28.4 | 37.9 | 2 |
| RF7 | 8.1 | 1.9 | 0.0* | 374.5 | 27.5 | 105.3 | 4 |
| RF8 | 6.4 | 1.2 | 0.0 | 310.3 | 0.0 | 36.1 | 4 |
| Mean | 6.6 | 1.7 | 0.1 | 267.4 | 55.5 | 52.9 | |
| SD | 1.4 | 1.4 | 0.3 | 95.6 | 90.3 | 33.9 | |
| Aug. 2003 | First | Last | Oct. 2005 | First* | Last |
|---|---|---|---|---|---|
| Couple no. | kg/day | kg/day | Couple no. | kg/day | kg/day |
| DF1 | 1.2 | 3.2 | RF1 | 0.0 | 0.0 |
| DF2 | 1.1 | 1.5 | RF2 | 1.4 | 6.1 |
| DF3 | 2.0 | 0.6 | RF3 | 0.0 | 2.0 |
| DF4 | 1.8 | 8.2 | RF4 | 2.0 | 2.8 |
| DF5 | 3.1 | 7.0 | RF5 | 1.4 | 6.0 |
| DF6 | 1.9 | 2.8 | RF6 | 0.0 | 0.4 |
| RF7 | 3.3 | 0.6 | |||
| RF8 | 2.4 | 0.0 | |||
| Wilcoxon’s test | ns | ns | |||
First: The weight per day of games caught in the first 10 days.
Last: The weight per day of games caught in the last 10 days.
| Aug. 2003 | First | Last | Oct. 2005 | First* | Last |
|---|---|---|---|---|---|
| Couple no. | g/day | g/day | Couple no. | g/day | g/day |
| DF1 | 63.2 | 93.4 | RF1 | 109.4 | 84.2 |
| DF2 | 122.9 | 254.3 | RF2 | 295.3 | 358.8 |
| DF3 | 247.5 | 329.6 | RF3 | 381.8 | 318.2 |
| DF4 | 134.1 | 215.5 | RF4 | 293.7 | 261.3 |
| DF5 | 118.7 | 259.5 | RF5 | 145.1 | 333.1 |
| DF6 | 167.3 | 296.5 | RF6 | 162.4 | 171.7 |
| RF7 | 332.0 | 416.9 | |||
| RF8 | 284.5 | 338.7 | |||
| Wilcoxon’s test | P = 0.0313 | ns | |||
First: The weight per day of yam tubers collected in the first 10 days.
Last: The weight per day of yam tubers collected in the last 10 days.
We did not measure individual food intake. Instead, we derived the total food intake of cooperators and their children from all food acquired less any leftovers, the bulk of which was preserved dry meat taken back to their settlement, and the amount of food discarded during each survey period. The cooperators usually left the yam tubers brought back behind their huts, covered with leaves to prevent drying. According the cooperators, yam tubers start going off two or three days after being collected. On the second trip, about 5% of yam tubers collected were discarded because of rot. Since we did not weigh the discarded tubers in the first trip, in the estimation of food intake we assumed yam tubers would have been discarded at the same rate as the second trip. Some yam tubers were brought back to the home settlement and some used for packed lunches on the way back home from the camp after the surveys finished. Thus, ingested yam tubers were estimated at 1.4 kg per consumption-day in fresh weight in the dry season and 1.5 kg per consumption-day in the rainy season. Nuts, game, and fish were dried and preserved on their fires except for food that was cooked on the day of foraging. A portion of these dried foods was also brought back to their home. Above all, a large amount of game meat was brought back since it was of particular value in the settlement where the cooperators could not have it very often and could also sell it. Therefore, ingested meat was estimated at 0.4 kg per consumption-day in fresh weight in the dry season and 0.3 kg per consumption-day in the rainy season.
We estimated the daily total energy intake at 2528–2865 kcal per consumption-day in the dry season, and at 2479–2777 kcal per consumption-day in the rainy season (Table 12). Daily total energy intake per consumption-day in the dry season was slightly larger than that in the rainy season. The contribution of food type to total energy intake was similar in both seasons: yam tubers supplied more than 60% of the total energy intake, game meat 15–20%, and nuts around 10% and the sum of these three food types more than 90%. The daily total protein intake was estimated at 114.5–146.1 g per consumption-day in the dry season and 93.0–125.4 g per consumption-day in the rainy season. Although around 60% of total protein intake was from animal protein in both seasons, its contribution in the dry season was larger than in the rainy season due to larger catch of game and fish.
| Food type9 | Aug. 2003 (16 persons) | Oct. 2005 (23 persons) | 7Energy per 100 g | 7Protein per 100 g | 8Edible ratio | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| W1 kg | No2 | En5 | Protein | W1 kg | No2 | En5 | Protein | ||||||||||
| D3 | R4 | kcal | % | g | % | D3 | R4 | kcal | % | g | % | kcal | g | ||||
| Yam tubers | 660.7 | 30.66 | 18.9 | 1618.9–1792.3 | 62.6–64.0 | 21.7–46.3 | 18.9–31.7 | 1092.9 | 50.6 | 113.6 | 1712.8–1896.3 | 68.3–69.1 | 22.9–48.9 | 24.7–39.0 | 112–124 | 1.5–3.2 | 0.7 |
| Nuts | 23.0 | 4.8 | 275.6 | 9.6–10.9 | 8.5 | 5.8–7.4 | 44.3 | 14.4 | 315.2 | 11.4–12.7 | 9.7 | 7.7–10.4 | 498 | 15.3 | 0.9 | ||
| Fungi | 3.0 | 0.0 | 2.9 | 0.1 | 0.1 | 0.1 | 8.7 | 1.0 | 5.2 | 0.2 | 0.2 | 0.2–0.3 | 32 | 1.5 | 0.9 | ||
| Game meat | 343.1 | 172.5 | 427.2–560.4 | 16.9–19.6 | 65.8–74.5 | 51.0–57.5 | 290.6 | 95.8 | 339.7–445.6 | 13.7–16.0 | 52.3–59.2 | 47.2–56.3 | 109–143 | 16.8–19.0 | 0.68 | ||
| Bird | 0.2 | 0.0 | 0.2–0.3 | 0.0 | 0.1 | 0.0–0.1 | 105 | 22.5 | 0.5 | ||||||||
| Bird’s egg | 0.3 | 0.0 | 1.2 | 0.0 | 0.1 | 0.1 | 140 | 11.8 | 0.85 | ||||||||
| Snake | 3.6 | 0.0 | 8.0 | 0.3 | 1.2 | 0.8–1.1 | 94 | 14.4 | 0.7 | ||||||||
| Fish | 49.5 | 15.1 | 59.3–89.3 | 2.3–3.1 | 11.3–12.6 | 7.7–11.0 | 12.9 | 0.0 | 15.5–23.3 | 0.6–0.8 | 2.9–3.3 | 2.4–3.5 | 85–128 | 16.2–18.1 | 0.6 | ||
| Crustaceae | 0.3 | 0.0 | 0.2 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 63 | 13.9 | 0.3 | ||
| Snail | 0.8 | 0.0 | 1.2 | 0.0 | 0.0–0.2 | 0.0–0.2 | 1.7 | 0.0 | 1.7 | 0.1 | 0.3 | 0.2–0.3 | 107 | 17.7 | 0.4 | ||
| Termite | 5.9 | 0.0 | 71.0 | 2.5–2.8 | 4.1 | 2.8–3.6 | 9.3 | 0.7 | 72.0 | 2.6–2.9 | 4.1 | 3.3–4.4 | 356 | 20.4 | 1 | ||
| Honey | 8.0 | 0.0 | 63.0 | 2.2–2.5 | 0.1 | 0.1 | 3.1 | 0.0 | 17.0 | 0.6–0.7 | 0.0 | 0.0 | 311 | 0.4 | 0.75 | ||
| Total | 1111.8 | 2528.4–2865.1 | 100.0 | 114.5–146.1 | 100.0 | 1463.8 | 2479.4–2776.6 | 100.0 | 93.0–125.4 | 100.0 | |||||||
En = (W – (D + R)) × Energy value × Edible ratio/Consumption-day, Protein = (W – (D + R)) × Protein value × Edible ratio /Consumption-day
Consumption-day = Number of subjects × research days: 14.8 (six couples + their four children × 0.7) × 20 = 296 in Aug. 2003, and 15.5 (six couples and their five children × 0.7) × 20.6 + 5.4 (two couples and their two children × 0.7) × 19.6 = 425.1 in Oct. 2005.
A child was converted to 70% of an adult.
In the 2005’s survey one meal for a person in the evening on the beginning day of survey is calculated as 0.6 consumption-day, because the breakfast on the last day was not weighed.
We timed the periods the cooperators spent outside the campsite, except for urination and carrying water and fuel for their fires. This time (working time) could be regarded as the time spent in the cooperators’ food-getting activities. We also recorded with pedometers the number of steps walked by cooperators everyday. Most cooperators presented a strong positive correlation between their daily working time and number of steps walked except for a few persons in both seasons (Table 13). The mean working time and number of steps walked per person per day had a seasonal difference; those in the dry season were larger than in the rainy season (Figure 3). There was a large daily variance in the working time and number of steps walked by individuals. One of the reasons was the high intensity of activity on YCD (Table 14). The daily working time of couple on YCD, 800–1000 minutes, was 1.2–1.5 times longer than that on days other than YCD and the daily steps walked by couples on YCD, 20000–30000, were 1.4–3.0 times more than those on days other than YCD in both seasons. We recorded many cases where the steps walked by cooperators greatly decreased on the day after they took long walks. This is because the yam collecting area was concentrated around Mount Bek, about 2.5 km from the camp. As shown in Figure 4, any sites of foraging activities other than yam collecting were a shorter distance from the camp, although there was a difference that nut collecting sites were widely dispersed whereas most snaring sites were within a radius of 1 km from the camp. On YCD, cooperators had to walk 3–4 km along mountain roads back to the camp carrying heavy tubers of around 10 kg on their backs. Thus the cooperators would intentionally engage in lighter activities on the days other than YCD. However, neither the working time and steps walked of couples on YCD, nor the weight of yam tubers collected per unit of working time and per unit of steps walked by couples on YCD had a correlation with the order of YCD (Table 15). This suggests that during both survey periods there was no gathering pressure to cause the cooperators to increase the intensity of yam collecting activity, or to decrease its efficiency by the day. Nor was there a difference in the mean daily working time and number of steps walked by couples between the first half and the last half of research period (Table 16).
| Individual no. Aug. 2003 | Sex | Work time per day (min.) | Steps walked per day (n) | 1Correlation coefficient | 3Ss | Individual no. Oct. 2005 | Sex | Work time per day (min.) | Steps walked per day (n) | Correlation coefficient | Ss |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | ||||||||
| DF11 | M | 489 ± 88 | 9577 ± 3000 | 0.2349 | ns | RF11 | M | 382 ± 142 | 7489 ± 4055 | 0.8392 | *** |
| DF122 | F | 484 ± 87 | 26766 ± 6781 | 0.6958 | *** | RF12 | F | 378 ± 137 | 6915 ± 3967 | 0.8249 | *** |
| DF21 | M | 460 ± 82 | 10893 ± 4427 | 0.7740 | *** | RF21 | M | 409 ± 97 | 11654 ± 4423 | 0.6714 | *** |
| DF22 | F | 436 ± 78 | 9865 ± 6201 | 0.7478 | *** | RF22 | F | 373 ± 107 | 9175 ± 4741 | 0.6706 | *** |
| DF31 | M | 437 ± 77 | 9060 ± 4506 | 0.7288 | *** | RF31 | M | 356 ± 105 | 8101 ± 4412 | 0.5093 | * |
| DF32 | F | 435 ± 71 | 8032 ± 4906 | 0.7172 | *** | RF32 | F | 336 ± 98 | 5828 ± 3696 | 0.4328 | ns |
| DF41 | M | 459 ± 76 | 13457 ± 4080 | 0.5907 | ** | RF41 | M | 389 ± 108 | 9561 ± 4300 | 0.7411 | *** |
| DF42 | F | 436 ± 74 | 9998 ± 4901 | 0.5116 | * | RF42 | F | 331 ± 96 | 13490 ± 6483 | 0.6257 | ** |
| DF51 | M | 462 ± 69 | 15380 ± 4150 | 0.7374 | *** | RF51 | M | 437 ± 134 | 12770 ± 5092 | 0.7655 | *** |
| DF52 | F | 434 ± 76 | 9812 ± 4828 | 0.7624 | *** | RF52 | F | 389 ± 150 | 7147 ± 4510 | 0.7606 | *** |
| DF61 | M | 450 ± 93 | 12638 ± 4150 | 0.8044 | *** | RF61 | M | 381 ± 105 | 12862 ± 4564 | 0.7292 | *** |
| DF62 | F | 436 ± 93 | 13073 ± 6308 | 0.3841 | ns | RF62 | F | 310 ± 131 | 6727 ± 4234 | 0.7916 | *** |
| RF71 | M | 384 ± 116 | 8758 ± 3876 | 0.8377 | *** | ||||||
| RF72 | F | 342 ± 125 | 12364 ± 6184 | 0.8449 | *** | ||||||
| RF81 | M | 250 ± 166 | 5885 ± 5520 | 0.9131 | *** | ||||||
| RF82 | F | 309 ± 138 | 9276 ± 5219 | 0.9320 | *** | ||||||
Mean daily work time and number of steps walked by cooperators. The reason why one woman recorded the most steps was that she could not walk normally because of leg pain.
| Couple no. | NYCD | TYCD | St | SYCD | St | Couple no. | NYCD | TYCD | St | SYCD | St |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Aug. 2003 | NO-YCD | TOYCD min. | SOYCD | Oct. 2005 | NO-YCD | TOYCD min. | SOYCD | ||||
| DF1 | 12 | 1016 | ns | 41161 | *** | RF1 | 11 | 951 | *** | 19820 | *** |
| 8 | 907 | 29116 | 10 | 554 | 8446 | ||||||
| DF2 | 10 | 1004 | *** | 30689 | *** | RF2 | 9 | 913 | ** | 29888 | *** |
| 10 | 787 | 10827 | 12 | 683 | 14034 | ||||||
| DF3 | 10 | 961 | ** | 25664 | *** | RF3 | 8 | 800 | * | 22895 | *** |
| 10 | 782 | 8519 | 13 | 626 | 8411 | ||||||
| DF4 | 11 | 984 | *** | 29922 | *** | RF4 | 10 | 849 | ** | 30273 | *** |
| 9 | 788 | 17230 | 11 | 603 | 16486 | ||||||
| DF5 | 9 | 960 | ns | 30680 | *** | RF5 | 10 | 982 | ** | 27088 | *** |
| 11 | 841 | 17543 | 11 | 684 | 13399 | ||||||
| DF6 | 12 | 993 | *** | 30695 | *** | RF6 | 9 | 848 | ** | 26782 | *** |
| 8 | 724 | 18234 | 11 | 561 | 13704 | ||||||
| RF7 | 10 | 890 | *** | 28996 | *** | ||||||
| 10 | 562 | 13248 | |||||||||
| RF8 | 11 | 704 | ** | 21970 | *** | ||||||
| 10 | 400 | 7672 |
NYCD: the number of YCD; NOYCD: the number of days other than YCD.
TYCD: Mean combined daily work time of husband and wife on YCD.
TOYCD: Mean combined daily work time of husband and wife on days other than YCD.
SYCD: Mean combined daily steps walked by husband and wife on YCD.
SOYCD: Mean combined daily steps walked by husband and wife on days other than YCD.
St: Student’s t-test:
Plot of foraging activity sites recorded by GPS. This figure was made based on the daily activities of 12 cooperators (7 males and 5 females) recorded by GPS. A researcher recorded the geographical position of each new site by GPS every time a cooperator changed the site or the foraging activity. The number of sites plotted was 17 for yam collecting by 5 males and 2 females, 14 for panda nuts collecting by 3 males and 4 females, 5 for wire-snaring by 2 males, and 5 for other activities by 1 male and 3 females.
| Couple no. | Total of YCD | Order vs. Yam weight | Order vs. Work time | Order vs. Steps walked | Order vs. Y/WT | Order vs. Y/SW | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Aug. 2003 | T | Ss | T | Ss | T | Ss | T | Ss | T | Ss | |
| DF1 | 12 | −0.2462 | ns | 0.2727 | ns | 0.0606 | ns | −0.3939 | ns | −0.3333 | ns |
| DF2 | 10 | −0.0667 | ns | 0.2 | ns | −0.0222 | ns | 0.0667 | ns | 0.0222 | ns |
| DF3 | 10 | 0.0667 | ns | −0.1556 | ns | −0.1556 | ns | 0.1556 | ns | 0.0667 | ns |
| DF4 | 11 | 0.0556 | ns | 0 | ns | 0.3889 | ns | 0.0556 | ns | 0.0556 | ns |
| DF5 | 9 | −0.0182 | ns | 0.0545 | ns | −0.0182 | ns | −0.0909 | ns | −0.0909 | ns |
| DF6 | 12 | −0.1679 | ns | −0.1818 | ns | 0.1212 | ns | −0.1515 | ns | −0.3333 | ns |
| Oct. 2005 | T | ss | T | ss | T | ss | T | ss | T | ss | |
| RF1 | 11 | −0.1636 | ns | −0.0734 | ns | −0.3818 | ns | −0.2 | ns | −0.0909 | ns |
| RF2 | 9 | −0.1111 | ns | −0.3889 | ns | −0.1111 | ns | 0.1111 | ns | −0.0556 | ns |
| RF3 | 8 | 0.2857 | ns | −0.5455 | ns | −0.3571 | ns | −0.5 | ns | −0.2857 | ns |
| RF4 | 10 | −0.0222 | ns | −0.1556 | ns | 0.1111 | ns | −0.0667 | ns | −0.2 | ns |
| RF5 | 10 | 0.2247 | ns | −0.2889 | ns | −0.2889 | ns | 0.4667 | ns | 0.4222 | ns |
| RF6 | 9 | −0.3333 | ns | −0.1111 | ns | −0.3333 | ns | −0.2222 | ns | −0.2222 | ns |
| RF7 | 10 | −0.3333 | ns | −0.2444 | ns | −0.1556 | ns | −0.2889 | ns | −0.3778 | ns |
| RF8 | 11 | −0.5273 | * | −0.5273 | * | −0.4182 | ns | −0.0545 | ns | −0.0545 | ns |
YCD: yam collecting days.
Y/WT: The weight of yam tubers collected per 100 minutes of work time on YCD.
Y/SW: The weight of yam tubers collected per 1000 steps walked on YCD.
T: Kendall’s rank correlation coefficient.
Ss: Statistical significance:
| Individual no. | Sex | Work time per day (min.) | St | Steps walked per day | St | ||||
|---|---|---|---|---|---|---|---|---|---|
| Entire | First half | Last half | Entire | First half | Last half | ||||
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | ||||
| Aug. 2003 | |||||||||
| DF11 | M | 489 ± 88 | 461 ± 112 | 517 ± 39 | ns | 9577 ± 3000 | 10710 ± 3718 | 8443 ± 1517 | ns |
| DF21 | M | 460 ± 82 | 438 ± 98 | 481 ± 56 | ns | 10893 ± 4427 | 10250 ± 5060 | 11536 ± 3852 | ns |
| DF31 | M | 437 ± 77 | 429 ± 84 | 444 ± 70 | ns | 9060 ± 4506 | 8695 ± 5114 | 9424 ± 4048 | ns |
| DF41 | M | 459 ± 76 | 429 ± 80 | 488 ± 60 | ns | 13457 ± 4080 | 13084 ± 4656 | 13830 ± 3625 | ns |
| DF51 | M | 462 ± 69 | 456 ± 85 | 468 ± 49 | ns | 15380 ± 4150 | 15089 ± 4952 | 15671 ± 3409 | ns |
| DF61 | M | 450 ± 93 | 437 ± 110 | 463 ± 73 | ns | 12638 ± 4150 | 13164 ± 4957 | 12112 ± 3821 | ns |
| DF121 | F | 484 ± 87 | 451 ± 108 | 516 ± 39 | ns | 26766 ± 6781 | 24846 ± 8108 | 28687 ± 4809 | ns |
| DF22 | F | 436 ± 78 | 418 ± 88 | 454 ± 65 | ns | 9865 ± 6201 | 9841 ± 6720 | 9889 ± 5999 | ns |
| DF32 | F | 435 ± 71 | 432 ± 69 | 438 ± 74 | ns | 8032 ± 4906 | 7773 ± 5525 | 8292 ± 4486 | ns |
| DF42 | F | 436 ± 74 | 422 ± 84 | 450 ± 61 | ns | 9998 ± 4901 | 10025 ± 5291 | 9971 ± 4763 | ns |
| DF52 | F | 434 ± 76 | 423 ± 87 | 445 ± 63 | ns | 9812 ± 4828 | 8878 ± 5008 | 10559 ± 4807 | ns |
| DF62 | F | 436 ± 93 | 427 ± 106 | 445 ± 80 | ns | 13073 ± 6308 | 10192 ± 4017 | 15954 ± 7028 | * |
| Oct. 2005 | |||||||||
| RF11 | M | 382 ± 142 | 353 ± 180 | 413 ± 83 | ns | 7489 ± 4055 | 7382 ± 5158 | 7606 ± 2635 | ns |
| RF21 | M | 409 ± 97 | 415 ± 121 | 402 ± 66 | ns | 11654 ± 4423 | 11894 ± 4835 | 11390 ± 4166 | ns |
| RF31 | M | 356 ± 105 | 382 ± 120 | 327 ± 81 | ns | 8101 ± 4412 | 7977 ± 5197 | 8237 ± 3634 | ns |
| RF41 | M | 389 ± 108 | 370 ± 131 | 410 ± 78 | ns | 9561 ± 4300 | 9306 ± 4782 | 9842 ± 3940 | ns |
| RF51 | M | 437 ± 134 | 436 ± 166 | 438 ± 94 | ns | 12770 ± 5092 | 12509 ± 5433 | 13057 ± 4964 | ns |
| RF61 | M | 381 ± 105 | 362 ± 134 | 399 ± 67 | ns | 12862 ± 4564 | 12093 ± 5352 | 13631 ± 3744 | ns |
| RF71 | M | 384 ± 116 | 364 ± 121 | 404 ± 114 | ns | 8758 ± 3876 | 8608 ± 4510 | 8908 ± 3366 | ns |
| RF81 | M | 250 ± 166 | 314 ± 162 | 180 ± 147 | ns | 5885 ± 5520 | 7937 ± 6263 | 3628 ± 3651 | ns |
| RF12 | F | 378 ± 137 | 358 ± 170 | 404 ± 90 | ns | 6915 ± 3967 | 6653 ± 4849 | 7203 ± 2946 | ns |
| RF22 | F | 373 ± 107 | 373 ± 125 | 372 ± 89 | ns | 9175 ± 4741 | 8820 ± 5045 | 9565 ± 4620 | ns |
| RF32 | F | 336 ± 98 | 362 ± 117 | 307 ± 66 | ns | 5828 ± 3696 | 5824 ± 4133 | 5833 ± 3373 | ns |
| RF42 | F | 331 ± 96 | 315 ± 122 | 348 ± 58 | ns | 13490 ± 6483 | 10471 ± 6123 | 16811 ± 5325 | * |
| RF52 | F | 389 ± 150 | 385 ± 194 | 394 ± 91 | ns | 7147 ± 4510 | 7471 ± 5055 | 6791 ± 4067 | ns |
| RF62 | F | 310 ± 131 | 324 ± 129 | 295 ± 138 | ns | 6727 ± 4234 | 6768 ± 4817 | 6686 ± 3825 | ns |
| RF72 | F | 342 ± 125 | 317 ± 146 | 367 ± 102 | ns | 12364 ± 6184 | 10343 ± 6049 | 14385 ± 5922 | ns |
| RF82 | F | 309 ± 138 | 328 ± 163 | 289 ± 109 | ns | 9276 ± 5219 | 10105 ± 6032 | 8365 ± 4283 | ns |
St: Student’s t-test:
Although the surveys suffered from certain restrictions, such as a small population, short term, only two seasons, and a narrow forest area, there was no evidence against the possibility of a hunting-gathering lifestyle in a tropical rainforest. Recently, a long-term foraging trip (‘molongo’ in Baka language) of a different Baka group in a forest close to this survey area was reported (Yasuoka, 2006). According to Yasuoka, about 100 Baka people stayed there and lived off forest food resources for about two months, from February to April 2002. Among diverse food items acquired, wild yam tubers were the most important as an energy source during the trip. In northeastern Republic of Congo, six groups of Aka hunter-gatherers, consisting of 15–75 persons, engaged in hunting trips for 6–27 days, during which they were almost independent of agricultural products, which supplied only 11% of their total energy intake (Kitanishi, 1995). Here also wild yam tubers were one of their major energy sources. In addition, the Late Pleistocene sites excavated by Mercader and Martí (2003) were slightly distant from the region of these surveys. These findings suggest a high likelihood of a hunting-gathering lifestyle at least in the forest at the northwest margin of the Congo Basin. However, some questions still remain to be solved.
Carrying capacity of the African tropical rainforestFirst we will discuss how many people can live off forest food resources over many years, i.e. the carrying capacity of this forest. The tubers of yams and yam-like plants are the most important foods for a foraging life in the forest and a key food involved in the ‘wild yam question.’ Referring to the little available data on yams and yam-like tubers in the forest of the Lobaye Region, Central African Republic (Bahuchet et al., 1991) and in the southern Cameroon forest (Hladik and Dounias, 1993), we will examine this question based on the research of standing stem density and tuber productivity of yams and yam-like plants conducted in the forest from Ndongo Village to Moloundou town using the line-transect method (Sato, 2001). As shown in Table 17, the tuber biomass of yams and yam-like plants in six sites was estimated at 5.3–17.0 kg per hectare. The vegetation at three sites (total length of line-transects: 4 m × 6 km) showing higher values (15.3–17.0 kg/ha) was in the secondary forests including fallows, whereas that of the other three sites (total length of line-transects: 4 m × 7 km), the productivity of which was 5.3–8.7 kg/ha, had few anthropogenic disturbances. Here we will adopt the value of 5.0 kg/ha as the tuber biomass of yams and yam-like plants in this region since that in a forest site, almost all of which had a micro-vegetation type, ‘manja,’ was 5.3 kg/ha. ‘Manja’ is a forest with a closed canopy and occupied a large part of our survey area. Next we will calculate what weight of tubers of yams and yam-like plants an adult person requires per year based on our survey. The fresh weight of yam tubers intake was estimated at 611.2 kg in August 2003 and 928.7 kg in October 2005, which were converted to 2.06 and 2.18 kg per consumption-day, respectively. Adopting the value in the rainy season, we can calculate the annual requirements of yam tubers at about 800 kg per person. We consider the utilization ratio of yam tubers biomass as one-fifth, referring to the data of Bahuchet et al. (1991) from the Central African Republic. Thus we can estimate the forest area that will reserve 4000 kg of annual tuber requirements per person at 8 square kilometers. Carrying capacity in the survey area was 0.125 person/km2, which is similar to the 0.14 person/km2 in Yasuoka’s ‘molongo’ area (Yasuoka, 2006) and 0.10 person/km2 in the Lobaye Region (Bahuchet et al., 1991). It is noteworthy that this value could indeed be modest. We referred to the value of tuber biomass, 5.3–8.7 kg/ha in three sites that had few anthropogenic disturbances in the calculation of carrying capacity. More than 70% of tuber biomass in these three sites was from perennial yam plants such as Dioscorea minutiflora, D. smilacifolia, and D. burkiliana, whereas the tubers of annual D. praehensilis hardly contribute to the biomass. As was observed in all six sites, these three species of perennial yam plants must be distributed evenly over any forest. This means that these three perennial yam plants have the potential to sustain a carrying capacity comparable to 0.08 person/km2. As shown in our survey and Yasuoka’s ‘molongo,’ however, the tubers of annual D. praehensilis were indeed necessary for a pure foraging lifestyle in this region.We should knowhow much productivity this plant has, but it is difficult to determine the tuber biomass because the plants clump and the clumps are scattered over the broad forest. As the yam collecting sites were concentrated in Mount Bek, there was a large clump of D. praehensilis on the slope and top of the mountain, the biomass of which had been estimated at 118 kg/ha by Sato (2006). According to the cooperators, there were three clumps of D. praehensilis around the camp other than at Mount Bek and they mentioned their names, Mboto, Mokinda, and Mokondo, all of which were the names of hills. Yasuoka (2006) also reported several clumps of D. praehensilis which were used during the long-term expedition. Although Hart and Hart (1986) mentioned that the Mbuti in the Ituri Forest did not have clumped food such as tuberous plants in the Hadza and the mongongo nut in the Kalahari San, the Baka in the northwestern forest of Congo Basin could have many clumps of D. praehensilis. As it is difficult to determine clearly the tuber productivity of this plant, whose clumps are scattered about, with ordinal methods such as line-transects, we could not obtain useful data for it immediately. However, the facts that the clump at Mount Bek had a huge biomass of tubers and actually supplied a large amount of tubers for consumption, and there were at least several clumps around the survey area, suggest that taking into account the tuber productivity of D. praehensilis, we can expect a much larger carrying capacity from this forest than that mentioned above. In addition, although no tubers of Dioscoreophyllum cumminsii and only a few tubers of D. semperflorens were used in this survey, those tubers also could contribute to a foraging life in some areas as these yams and yam-like plants, especially D. semperflorens, played an important role as energy sources in both Yasuoka’s ‘molongo’ (Yasuoka, 2006) and hunting trips of the Aka hunter-gatherers (Kitanishi, 1995).
| Species | Site | S1 | S2 | S3 | S4 | S5 | S6 |
|---|---|---|---|---|---|---|---|
| g/ha | g/ha | g/ha | g/ha | g/ha | g/ha | ||
| Dioscoreophyllum cumminsii | 2793 | 6584 | 6052 | 1197 | 1696 | 150 | |
| Dioscorea praehensilis | 1110 | 278 | |||||
| D. mangenotiana | 12 | 12 | 6 | 12 | 3 | ||
| D. minutiflora | 370 | 398 | 974 | 500 | 412 | 19 | |
| D. burkilliana | 13575 | 8824 | 6788 | 6788 | 3054 | 7466 | |
| D. smilacifolia | 206 | 213 | 353 | 248 | 107 | 131 | |
| “njakaka” | ? | ||||||
| Total kg/ha | 17.0 | 16.0 | 15.3 | 8.7 | 5.3* | 8.0 | |
As mentioned above, it is unlikely that the hypothesis of the ‘wild yam question’ applies at least to this region, the southeastern part of Cameroon. We have insufficient data on the species of Dioscoreaceae to extend our conclusion to any other African tropical rainforest, whereas we can expand on their distribution. Although the annual species of Dioscoreaceae are commonly distributed in dry forest and woody savanna, D. praehensilis and D. semperflorens are particularly found in tropical rainforests (Hamon et al., 1995). However, these two species cannot be seen in the two botanical reports of importance in the Ituri Forest (Hart and Hart, 1986; Tanno, 1981), presumably because its vegetation type favors evergreen rainforests, where less productivity of D. praehensilis can be expected (Hladik et al., 1984). As Headland (1987) pointed out, the ‘wild yam question’ might be correct in evergreen tropical rainforests. According to White (1983: 79), however, the type of drier peripheral semi-evergreen rainforest, a synonym for semi-deciduous forest, which covers the entire survey area, occurs in the form of two bands running transversely across Africa, to the north and south of the moister forests. Large tuber productivity of D. praehensilis and D. semperflorens to support a foraging lifestyle could be expected within the broad bands of this vegetation type. Botanical studies of D. praehensilis and D. semperflorens in this vegetation type must therefore be undertaken.
Finally, we will comment on the vegetation of Mount Bek and its high productivity of yam tubers. As Bailey and Headland (1991) drew attention to the anthropogenic disturbance in tropical rainforests, e.g. the clearing of forests for farming, which was advantageous for light-demanding plants such as Dioscoreaceae, we also did not believe that the forest of Mount Bek has been left untouched. In fact, we found pottery fragments around Mount Bek. However, as we cannot identify who the users were, farmers or hunter-gatherers, when, and how long they lived there, neither we can determine the extent to which the inhabitants altered the vegetation of Mount Bek, nor to whether the altered vegetation contributed to increasing yam tuber productivity. There were several abandoned farmers villages, which were on the old map issued in 1910, around many yam patches which were used by the Baka people in Yasuoka’s ‘molongo’ (Yasuoka, 2009). Yasuoka said that the activities of forest people, including the Baka, might have influenced the distribution of yam patches or the productivity of yam plants, especially annual yams, but not directly; we agree with Yasuoka’s view. Archeological and historical surveys around Mount Bek need to be done. Here we will mention another possible factor linked to the high productivity of yam tubers. It is that the survey area was a hilly area. Much more sunlight can reach the surface of the top and slopes of hills than on the surface of flat forest. Such areas are suited to the genus Dioscorea of light-demanding plants, especially annual species. All the places of D. praehensilis clumps that the Baka cooperators named were hills. Extensive geomorphological surveys can bring in new findings for the tuber biomass and the distribution of yam plants.
The seasonality of D. praehensilisThe next question is the seasonality of D. praehensilis, which renews both stem and tuber annually (Hladik and Dounias, 1993). According to the cooperators, the amount of tubers they can collect will be slightly smaller in April–May, the minor rainy season, because some tubers of D. praehensilis sprout. Yasuoka (2011) described that the ‘molongo’ finished in the middle of April because fibrous tubers or tubers with a bad taste became more common due to germination. The minor rainy season, April–May, can be said to be the severest season for harvesting tubers of D. praehensilis. In April 2010, we conducted a similar two-week survey with the Baka cooperators, eight couples and their four children. The previous survey area could not be used because of new zoning laws for that forest area in Cameroon. Therefore the cooperators established the campsite in a forest area about two hours on foot from their settlement. The forest around the campsite, a semi-deciduous forest, was used daily for snaring or for short hunting trips by the Baka people. Concerned that this area was under high hunting-gathering pressure, and with the restricted use of wire snares by the cooperators to catch prey during the foraging trip, we decided to shorten the survey period. Here we will focus on the collecting activity of wild yam tubers, publishing the details of this survey elsewhere. As shown in Table 18, 576 kg of fresh tubers of yams and yam-like plants were collected during the 14-day survey period. The tubers of D. praehensilis accounted for more than 90% of collected tubers, as in our previous surveys. We also examined the weight of collected fresh tubers per consumption-day to compare the tuber productivity of yams and yam-like plants between four seasons; the minor dry, major rainy, and minor rainy season in our survey and the major dry season in Yasuoka’s survey. Calculating the consumption-day, we adopted the conversion factors in Yasuoka’s study, i.e. boy and girl 12 or over, and adult = 1, boy and girl between 2 and 12 = 0.5, and infant under 2 = 0. The value of D. praehensilis in the minor rainy season (1.97) is slightly less than that in the minor dry season (2.27) and much less than that in the major rainy season (2.66), but considerably larger than that in the major dry season (1.36). These facts are different from the accepted knowledge on the seasonality of D. praehensilis. According to Mckey et al. (1998), its biological cycle consists of a stem growth phase in April–May, a development phase in June–August, and a resting phase in September–March. Focusing on the tuber development, Dounias described its biological cycle as follows: D. praehensilis, which renews its aerial stem and tuber reserve every year, uses its tuber as an energy source for the sprouting and growth of new stems from April to July, stores tuber reserves from August to November, and then keeps its tuber reserves at maximum for the next sprouting after March (Dounias, 2001). He also pointed out that November–March is the best period for harvesting tubers. Generally agreeing, Yasuoka revised it to the period from November to April from his observations of ‘molongo’ (Yasuoka, 2006). In short, Dounias and Yasuoka say that April and August are not good seasons for harvesting the tubers of D. praehensilis due to expending tuber reserves for renewing stems and the early phase of storing new tuber reserves, respectively. Our three surveys seems to support their view: the weight of collected fresh tubers per consumption-day in October (the major rainy season) was the largest, followed by that in August (the minor dry season) and that in April (the minor rainy season) was the least. However the lowest value in April (1.97) was much larger than that in the major dry season from February to April in Yasuoka’s survey (1.36). Although we were unable to directly compare these values, as the circumstances of the survey were not identical, the difference is too large to ignore.
| Study season | Our studies | Yasuoka’s study (2006) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Aug.–Sep. 2003 (CD = 280) | October 2005 (CD = 388) | April 2010 (CD = 266) | Feb.–Apr. 2002 (CD = 3325) | |||||||||
| Tubers of yams and yam-like plants | Weight | W/CD | Weight | W/CD | Weight | W/CD | Weight | W/CD | ||||
| kg | % | kg | % | kg | % | kg | % | |||||
| Dioscorea praehensilis | 636.6 | 96.4 | 2.27 | 1030.6 | 94.3 | 2.66 | 525.1 | 91.1 | 1.97 | 4519.8 | 83.0 | 1.36 |
| D. semperflorens | 10.9 | 1.6 | 0.04 | 27.0 | 2.5 | 0.07 | 2.3 | 0.4 | 0.01 | 654.6 | 12.0 | 0.20 |
| D. mangenotiana | 7.7 | 1.2 | 0.03 | 22.4 | 2.0 | 0.06 | 31.0 | 5.4 | 0.12 | 61.8 | 1.1 | 0.02 |
| D. burkilliana | 4.8 | 0.7 | 0.02 | 12.3 | 1.1 | 0.03 | 16.5 | 2.9 | 0.06 | 184.7 | 3.4 | 0.06 |
| D. minutiflora | 0.7 | 0.1 | 0.00 | 0.6 | 0.1 | 0.00 | 0.8 | 0.1 | 0.00 | 22.7 | 0.4 | 0.01 |
| Dioscoreophyllum cumminsii | 0.7 | 0.1 | 0.00 | 3.1 | 0.1 | 0.00 | ||||||
| Total | 660.7 | 100.0 | 2.36 | 1092.9 | 100.0 | 2.82 | 576.4 | 100.0 | 2.16 | 5446.7 | 100.0 | 1.65 |
Weight: The fresh weight of tubers of wild yams and yam-like plants brought into the camp.
CD: consumption-day.
W/CD: Weight per consumption-day.
The conversion factors of consumption-day in Yasuoka’s study were adopted to compare our studies with Yasuoka’s study.
Conversion factors in Yasuoka’s study: boy and girl 12 or over and adult = 1; boy and girl between 2 and 12 = 0.5; infant under 2 = 0.
How could the much larger amount of D. praehensilis tubers collected in April and August than in the major dry season, December–March, or that a sufficient amount of D. praehensilis tubers to enable a foraging life for 2–3 weeks was collected in April and August, be accounted for? Although the length of survey periods and the number of cooperators is problematic, here we will lay out a likely hypothesis. As shown in Figure 2, the graph of rainfall in this region has bimodal peaks, which we can also see in the rainfall graph at Moloundou in Dounias’s paper (Dounias, 2001: 138). In fact, this region, including the survey area, has two distinct rainy seasons, April–May and September–November, and two dry seasons, June–August and December–March. Thus, it is possible that the beginning of the major rainy season in September is also another starting point for the biological cycle of D. praehensilis, although Dounias considered the beginning of the minor rainy season, April–May, as its starting point (Dounias, 2001). If it is correct, the best period for harvesting the tubers of this type could be April–August and it is likely that the sympatric presence of two types of D. praehensilis, which have different biological cycles, account for the relatively large amount of collected tubers in August or for the lack of seasonal disadvantage in the minor rainy season. Endicott and Bellwood (1991) mentioned the reason why wild yam tubers in the Malay Peninsula could be obtained at any time of year was that the dry season was so slight that it did not coordinate the tubering of different yam plants. Experimental research on how two dry seasons a year can result in tubering of yam plants or annual tubers productivity is needed.
2. The implications of this survey to reconstruct the Paleolithic foraging lifestyle in the African tropical rainforest Essential food for the Paleolithic foraging life in the African tropical rainforestConsidering the high contribution of wild yam tubers to the food supply in our survey and Yasuoka’s ‘molongo,’ they were the most dependable food for a purely foraging life in this region. Although the cooperators obtained various food items during the survey period, we could not find other reliable foods that had the potential to supply over 50% of dietary energy requirements. Among wild yam tubers, the supply of the annual plant D. praehensilis was particularly notable. D. praehensilis is likely to have been an essential food for the Paleolithic foraging life in the African tropical rainforest. We cannot know its productivity for certain as yet, but many clumps of this plant are present in this region. As a clump could supply a large amount of tubers to about 20 persons for 3 weeks in this survey area, and several clumps about 100 persons for more than 1 month in the northern part of this region (Yasuoka, 2006), these clumps could have sustained a Paleolithic foraging life. In addition, perennial wild yam plants also should have contributed to it because they had considerable biomass and were ubiquitously distributed in any forest (Sato, 2001); in fact, they supplied a fair amount of tubers during the nomadic period of ‘molongo’ (Yasuoka, 2006).
The results of this survey and Yasuoka’s ‘molongo’ indicated that various kinds of food other than wild yam tubers also could be important key foods that supported a Paleolithic foraging life. Among others, wild game, nuts, fish, honey, and insects must have played complementary roles as energy and protein sources. Below we will mention the significance of each food type and their issues with foraging activity.
(i) GameAlthough the hunting gear used in this survey comprised wire snares and iron spears, all game were caught with wire snares and almost all of them were forest duikers. Since Paleolithic hunters might have used only stone spears, it is unlikely that the amount of catch per person or per unit of time spent could exceed that of this survey where efficient wire snaring was conducted. According to Yasuoka (2011), the Baka iron spear hunters caught red river hogs or dwarf forest buffalos in six of ten cases of collective hunting forays including more than four hunters. Although Yasuoka noted that spear hunting was not suited for forest duikers, if a hunting method such as the driving method which Mbuti archers (Harako, 1976) or net hunters (Tanno, 1976; Ichikawa, 1983) in the Ituri Forest adopted, i.e. driving the duikers hidden in the bush out into the open and shooting them or into nets, was used, Paleolithic spear hunters may well have hunted forest duikers. The problem is that Paleolithic hunters who were devoted to hunting would have had no time for collecting wild yam tubers. Game meat must have played an important role for Paleolithic people, not only in a nutritional aspect but also in an emotional aspect as pleasure food.
(ii) NutsAlmost all nuts collected in this survey were P. oleosa. Since this fruit has a hard-shelled seed, it takes eight months to three years for this fruit to germinate after falling (Vivien and Faure, 1996). Therefore many black seeds, the size of table tennis balls, usually can be seen on the surface around this tree. Furthermore because this tree is common in the survey area, it is not difficult to collect seeds anytime and anywhere. Beside P. oleosa, several species of the genus Irvingia are also common in the survey area. Although few irvingia seeds were collected in this survey, it is quite usual for the Baka people to enter the forest to gather these seed nuts in the minor dry season, from June to August. Supplying a fair amount of dietary energy and enhancing the flavor of food as a source of dietary oil, nuts such as panda nuts and irvingia nuts must have been of substantial importance for Paleolithic hunter-gatherers. A particular problem is that Paleolithic gatherers would have spent a lot of time and effort shelling nuts after collecting seeds, and did so without the use of iron tools. In this survey couples usually collected seeds and shelled nuts in pairs using machetes. Judging from our observation, there would have been few differences in working efficiency between iron tools and stone tools. Rather, it needs to be considered whether Paleolithic women gatherers engaged by themselves in nut-collecting activity. Having been considered light work, this must have placed a considerable burden on Paleolithic woman gatherers.
(iii) FishIn this survey a small but indispensable amount of fish was procured in the dry season. Fish are indeed a valuable animal food source in this area, where there are two dry seasons. The present Baka cooperators knew three fishing methods—net fishing, hook and line fishing, and dam and bail fishing—whereas Paleolithic hunter-gatherers would have conducted only dam and bail fishing. Nevertheless, this last method is a reliable and effective means of catching fish. Engaging in it for one hour in a neighboring stream, even a solitary woman can obtain a sufficient catch for her family. Furthermore this fishing method does not need iron tools. In this survey, normal dam and bail fishing was a woman’s task, whereas men with spears joined in a special type of dam and bail fishing targeting big catfish hiding in holes on the bank. Considering the ease of fishing activities and the stability of fish procurement, fish must have been a contributor to the typical Paleolithic foraging life beyond the numbers in this survey.
(iv) Honey and insectsAlthough significant amounts of honey were not collected in this survey, it is well known that honey is a most attractive food for all pygmy groups (Ichikawa, 1981; Bahuchet, 1985; Kitanishi, 1995; Yasuoka, 2006). The Baka people, both men and women, pay special attention to seeking out honeycomb whenever they enter the forest. As Ichikawa (1981) reported, the Mbuti depend on honey for the bulk of their food in the rainy season. Honey must also have been a most valuable food for Paleolithic hunter-gatherers. The present Baka men usually cut down a tree with an iron ax to get honey, whereas Paleolithic hunters would have had no choice but to climb to get the honey. Climbing would have been much riskier than cutting, no matter which was the more effective.
Although there were various edible insects in this area, only termites were collected in this survey. Finding plump termites gave the cooperators great pleasure. Many kinds of edible insects, such as beetles, lepidopteran larvae, and termites, which can be gathered with simple tools or even without tools, must have been considered precious food as animal protein sources for Paleolithic hunter-gatherers in some seasons or in some areas.
Beside these foods, other diverse food items were obtained during both survey periods. The longer the survey period, the more the variety of food items increases. Although Hart and Hart (1986) and Bailey and Headland (1991) pointed out the high cost in accessing food resources in small, widely dispersed patches as one of challenges facing foragers in tropical rainforests, we will cite the presence of large, densely distributed patches of annual wild yam plants and the tremendous resource diversity regardless of the sparse distribution of each food resource as factors that enhance the possibility of a Paleolithic hunting-gathering lifestyle in this area.
The Paleolithic foraging lifestyle in the African tropical rainforest was not easyThe cooperators went out to collect yam tubers every 2 days in this survey. Such an activity pattern must have resulted from the facts that (1) yam tubers spoil in a couple of days after being dug out, and (2) the yam collecting places were not so close that they could visit them every day. Whereas the gathering activities in hunting-gathering societies heretofore have been described as women’s work, in this survey husband and wife commonly engaged together in yam collecting activity. It must have been because wire-snaring, men’s work, was very economical in terms of work. Setting 10–20 wire-snares in a couple of days, men made the round of snares every morning and/or on the way back from other activities for half an hour at most. Wire-snaring is not only economical but effective. In this survey wire-snares were set in the forest where people rarely hunted and were so strong that they could provide a large amount of catch: 0.7–1.2 kg per consumption-day. However, Paleolithic hunters without wire-snares and other iron hunting gear could not have afforded the time to join in the yam collecting activity but were compelled to dedicate themselves solely to hunting activity. If so, a woman must have had to seek, dig out, and carry back yam tubers of about 10 kg for her family every 2 days by herself. This would have been fairly hard work for women. To solve this problem, Paleolithic hunter-gatherers must have built their camp as close to the clump of annual yam plants as possible. However, they could not always have done so because they had to secure other kinds of food and water, all of which might not have been close to the yam clump. The reason why in this survey the cooperators did not build their camp close to the yam clump was certainly that they could not have done so. Since there was no watering place along the mountain roads between the camp and the yam clump, they set their camp on the bank of the Leke River where they could easily get both water and fish. Furthermore, digging tubers would have been harder for Paleolithic gatherers since they did not have iron points for their wooden digging sticks. The Baka people can make and use two types of wooden digging stick, ‘ngbafa’ and ‘bodunga.’ ‘Ngbafa’ is a simple digging stick, the end of which is sharply cut, and ‘bodunga’ is a special wooden stick, the end of which is cut into four parts which are chamfered and fixed with vines like a cone, to dig out tubers of D. praehensilis and D. semperflorens deep in the ground (see Dounias, 1993: 628). The present Baka people rarely use these wooden sticks now because they have iron points. Wooden sticks are suited for soft soil but not for soil and gravel, whereas a wooden stick with an iron point attached can be used for both soil types. Furthermore, making a ‘bodunga’ is a time-consuming task, but the attachment of an iron point to the end of a wooden stick is fast and easy. According to Hurtado and Hill (1989), Machiguenga women using wooden tools spent two to three times as many minutes in digging and peeling manioc tubers as women using machetes and knives.
An important consideration of a foraging lifestyle in tropical rainforests is the large energy expenditure required. We estimated the daily energy intake at 2528–2865 kcal per person in the dry season and 2479–2777 kcal per person in the rainy season. Since there was no large fluctuation in the body weight of cooperators during both survey periods, the daily energy intake, i.e. total energy intake (TEI), could be considered as the daily energy expenditure, i.e. total energy expenditure (TEE). We can estimate the mean BMR from the body weight of the cooperators at 1413 kcal/day in males and 1141 kcal/day in females in the dry season and at 1425 kcal/day in males and 1190 kcal/day in females in the rainy season (FAO, 2001). If the physical activity level (PAL: TEE/BMR) were the same for both sexes, we can calculate approximate PAL value at 1.98–2.24 in the dry season and 1.90–2.12 in the rainy season substituting total energy intake for total energy expenditure. According to FAO (2001), a lifestyle with a PAL value of 1.70–1.99 is classified as an active or moderately active lifestyle, and that of 2.00–2.40 as a vigorous or vigorously active lifestyle. Leonard and Robertson (1992), who studied the bioenergetics of primates from the evolutionary viewpoint, described the PAL values of two hunter-gatherer groups as 1.71 for males and 1.51 for females of the !Kung and 2.15 for males and 1.88 for females of the Ache. The PAL value of the Baka is higher than that of the !Kung and seems to be comparable to that of the Ache. It is noteworthy that both higher groups are tropical rainforest foragers. In this sense, as per Bailey et al., such costs as searching and traveling for resources in small, widely dispersed patches are high in the tropical rain-forests, particularly for centrally placed foragers like the cooperators in this survey staying in camps for a certain period (Bailey and Headland, 1991). Yamauchi et al. (2000) reported on the daily physical activity level of African populations, including the Baka leading a village life in their sedentary settlement. Among more than ten populations, a farmer group in Gambia had the highest PAL value in both sexes: 2.02 for males and 1.97 for females, which are equivalent to that in this survey. In contrast, the Baka leading a village life had a PAL value of 1.41 for males and 1.56 for females, corresponding to sedentary or light activity lifestyle in the classification of FAO (2001). The hunting-gathering life examined in this survey suggests a most active level of daily life for human survival. Paleolithic hunter-gatherers without iron tools in tropical rainforests may have led a much more severe foraging life.
Who were Paleolithic hunter-gatherers in the African tropical rainforest?Finally we will address the question of who such Paleolithic hunter-gatherers in the tropical rainforest of the Congo Basin would have been. We believe they would most likely have been the ancestors of present pygmy groups for two reasons: one is that their physical features were unlikely to be established without close association with their environment in tropical rainforests, and the other is recent genetic evidence.
Hiernaux argued that the physical characteristics of the Mbuti pygmy, shorter stature, low and wide nose and lighter skin color, represents their biological adaptation to the hot and humid equatorial forest (Hiernaux, 1975a). Agreeing with Hiernaux, Cavalli-Sforza developed a convincing argument that a selective factor must have been whether the heat produced inside the body could be effectively dissipated in a hot and humid environment and the pygmy’s short stature and low weight would have resulted from the adaptation for heat dissipation in the hot and humid tropical rainforest (Cavalli-Sforza, 1986). That is, the adaptation to a hot and humid forest environment, where convection, radiation, and evaporation cannot be expected to effectively dissipate body heat, would have been accomplished by ‘miniaturizing’ the body, short stature and low weight, resulting in decreasing heat production. Both are consistent in that the pygmy’s physical features are genetic and the consequences of adaptation to tropical rainforests. Although we basically agree, we still question Cavalli-Sforza’s argument. Should tropical rainforests be regarded as so hot and humid? We know that the maximum temperature reaches most 30°C during the day in the tropical rainforest of the Congo Basin and before dawn the temperature drops to 20°C; those who lived in the Congo Basin may have felt rather chilly in the early morning. In general the humidity reaches 80–90% on chilly mornings whereas it drops to less than 60% as the temperature rises during the afternoon, and it is rather cooler on rainy days. Could such climatic conditions have been a selective factor for Paleolithic hunter-gatherers?
Both Hiernaux (1975a, b) and Cavalli-Sforza (1986) concluded that nutrition would not have been solely responsible for pygmies’ physical features. Nor do we think so, but, as mentioned already, the hardness of foraging life in the tropical rainforest, i.e. the high cost paid to secure food requirements, may have been a contributing factor to their short stature and low body weight. For example, if a 25-year-old man weighing 60 kg and his twin weighing 50 kg lead their lives at a PAL (TEE/BMR) value of 2.0, the former’s BMR is 1595.6 kcal and TEE 3191 kcal and the latter’s BMR 1445.1 kcal and TEE 2890.1 kcal (see FAO, 2001: 37). The former needs 300 kcal of energy more than the latter. This is equivalent to about 300 g in wild yam tubers. In tropical rainforests, where the cost of exploiting resources is high, smaller food requirements could be of great advantage. Thus the small body weight of the pygmy would have been a likely evolutionary development.
Hiernaux (1975a: 117) thought that the pygmy’s physical features were genetic in orgin and the period of 20000 years indicated by prehistory would be long enough for selection to establish the present pygmy’s gene pool. Cavalli-Sforza calculated the separation between Khoisan, Mbuti, and other African groups to be about 10000–20000 years based on 20 loci (Cavalli-Sforza, 1986: 414). By considering new genetic information, he revised the separation time to be 18000 years BP (Cavalli-Sforza et al., 1994). Recently it has been reported that the separation between pygmy groups and Bantu groups would have taken place less than 70000 years ago based on the analysis of mitochondrial (mt) DNA (Quintan-Murci et al., 2008). Furthermore resequencing 24 independent noncoding regions across the genome,Patin et al. (2009) inferred the separation time between pygmy hunter-gatherers and African farmers to be 60000 years BP and the separation between the western and the eastern groups of pygmy as 20000 years BP. As for the latter, elsewhere, the separation time has been estimated at 18000 years BP based on mt-DNA analysis (Destro-Bisol et al., 2004). This genetic information indicates that the divergence between pygmy groups and other African groups was 70000–60000 years BP and the divergence of two Pygmy groups 20000–18000 years BP. If this is correct, it is likely that Paleolithic hunter-gatherers in the African tropical rainforest were the ancestors of present pygmy groups considering their physical features.
This study, the observations on two 20-day controlled foraging trips in the minor dry season of August 2003 and the major rainy season of October 2005 in southeastern Cameroon, found no evidence that it was impossible for human beings to live independently of farmed agricultural products in a tropical rainforest environment. In both seasons a sufficient amount of food was constantly acquired, without any food shortages, allowing the Baka cooperators to maintain their body weight during the survey period. Among more than 60 kinds of foods procured during the survey periods, wild yam tubers provided more than 60% of their estimated energy intake in both seasons, followed by game meat and wild nuts. These three major types of food supplied 90% of their dietary energy. Wild yam tubers, especially the annual plant D. praehensilis were found to be an essential food to enable a foraging life in the southeastern Cameroon rainforests. The productivity of D. praehensilis in this survey area was sufficient for six to eight families of the Baka cooperators to maintain a foraging lifestyle for 20 continuous days, with each marriage couple engaged in collecting yam tubers every 2 days. We also considered the implications of this study for the possibility of a Paleolithic foraging lifestyle existing in the African tropical rainforest.
In conclusion, we proved that the hypothesis of the ‘wild yam question’ could not be applied to all tropical rainforests. In the northwestern Congo basin, rich in wild yam tubers, a foraging lifestyle is possible although at a high energy cost to secure sufficient food resources to maintain it. This implies that a Paleolithic foraging lifestyle in the African tropical rainforest may well have been possible, although not easy, and that Paleolithic foragers may have been the ancestors of the present ‘pygmy’ hunter-gatherers.
We are extremely grateful to Dr Ngima of the University of Yaoundé, Cameroon, for supporting our field surveys in Cameroon. We also express our gratitude to Mr Takanori Ohishi, Kyoto University for giving us useful documents. Special thanks go to our Baka friends in the survey area, especially Mr G. Menyata, for their hospitality, kindness, and cooperation over the long period of this research. This study was supported by Grant-in-Aid for Scientific Research (Nos. 06041046, 15405016, 20405014) from the Japanese Society for the Promotion of Science.
