NON-WOOD FOREST PRODUCTS MARKETS AND POTENTIAL DEGRADATION OF THE FOREST RESOURCE IN CENTRAL AFRICA:

Ousseynou Ndoye, Manuel Ruiz-Perez, Antoine Eyebe

Abstract

This paper investigates the importance of income generated from marketing selected NWFP in Central Africa. The focus is mainly on local markets in the Humid Forest Zone of Cameroon and markets at the borders with the Central African Republic (CAR), Equatorial Guinea and Gabon.

The main NWFP species considered are Irvingia spp. (bush mango nut), Cola acuminata (kola nut), Garcinia lucida (essock) and Garcinia kola (onie). Emphasis is on the marketing of the kernel and fruit of Irvingia spp. and Cola acuminata, and the bark for Garcinia lucida and Garcinia kola.

The study found that the value of the NWFP marketed was US$ 753 000 in the first half of 1995 and US$ 499 000 in the same period of 1996. Of these amounts, the value of barks marketed (for both Garcinia lucida and Garcinia kola) represented US$ 30 000 and US$

23 500 in 1995 and 1996, respectively. The decline in the significance of NWFP marketed, as a result of lower supply, shows the thinness of NWFP markets resulting from their dynamic and unpredictable nature, which changes the role of markets in assembling and distribution of NWFP from year to year.

The study also found that traders' weekly marketing margins from barks are comparable to and may even be higher than those from kernels (Irvingia spp.) and fruits (Cola acuminata). The harvesting of barks is not always carried out in a sustainable manner. However, their markets provide revenues for both traders and rural dwellers. This reveals the potential role of NWFP markets in forest resource degradation and underlines the difficulty in achieving a balance between improving the livelihoods of forest-dependent people and conserving the forest resource.

Future research should focus on determining the rate of bark harvest that would enable forest dwellers to sustainably derive part of their livelihoods from Garcinia lucida and Garcinia kola while preserving these trees.

1. Introduction

Many studies have shown the importance of Non-Wood Forest Products (NWFP) in the livelihood of forest dwellers (Chambers and Leach, 1987; de Beer and McDermott, 1989; Falconer, 1990; Scoones et al., 1992; Townson, 1995; Ndoye et al., 1997). Furthermore, NWFP are often cited as alternatives for sustainably managing and preserving the forests (Ndoye, 1994; Guedje, 1996). Despite their importance, NWFP are not always gathered or harvested in a sustainable way that will preserve the resource (see Hall and Bawa, 1993; Sayer, 1993; Peters, 1994; Laird, 1995; Freese, 1996). Ndoye (1995) found that 58 per cent of the palm trees (Elaeis guineensis) tapped by a sample of farmers in the Mbalmayo area, in the Humid Forest Zone of Cameroon, were killed. Furthermore, costs of resource degradation to society, as a result of killing each palm tree, varied between 70 000 to 129 000 CFA Francs (Ndoye, 1995). Champaud (1973) reported that the technique used by farmers in Mom, a village in the Centre province of Cameroon, to tap palm wine was to kill the palm tree (Elaeis guineensis). Oyono (1997) found that 1 000 palm trees (Raphia hookeri) were killed after tapping in Ekom, southeast Cameroon.

In the Southwest province of Cameroon (M'mouck village), farmers have embarked on planting Prunus africana because they had almost disappeared from some areas of the village, and farmers had to go deeper into the forest to harvest the bark (BDCPC, 1997). The market value of the bark of Prunus africana used for the treatment of Prostrate Gland Hypertrophy is estimated at US$150 million per year (Cunningham and Mbenkum, 1993, p. 7). Prunus africana bark is collected by 14 per cent of households around Mount Cameroon (BDCPC, 1997). Between 1986 and 1991, an average of 1 923 tons of Prunus africana bark were processed in Cameroon. This represented an average of 35 000 trees debarked per year, affecting at least 6 300 hectares of Afromontane forest each year (Cunningham and Mbenkum, 1993).

Due to the increased need for opportunities to generate cash income for people living in and around the forests, the potential role of NWFP markets in forest resource degradation and conservation should be studied on a continuous basis. The information generated could help researchers identify NWFP trees for domestication, to design improved and more sustainable harvesting methods (technologies) and to train forest gatherers to use these techniques. The data will also expand the range of appropriate decisions available to policy makers to improve the livelihood of forest-dependent people while preserving forest resources.

This paper attempts to clarify the potential role of NWFP markets in forest resource degradation and conservation by comparing the markets of two broad categories of NWFP: kernels/fruits and barks. In general, the gathering and marketing of kernels/fruits place less pressure on the resource than the harvesting of barks. Kernels and/or fruits are gathered and sold but the tree is preserved, whereas the harvesting and the marketing of barks may endanger the sustainability of forest resources because the survival of the tree could be at risk. Two particularly important NWFP in Central Africa included in the first category are the kernels of Irvingia spp. and the fruit of Cola acuminata. The barks of Garcinia lucida and Garcinia kola are part of the second category.

The demand of Irvingia gabonensis in southern Nigeria has been estimated at 80 000 tons per year (Leakey and Maghembe, undated). Nigerian traders import the fruit and the kernel from Cameroon (Scoones et al., 1992). In Cameroon, Nkongmeneck (1985) estimated the size of the market for kola nut (Cola acuminata) at 20 400 tons. In 1983, the income from kola nut in some rural households in southwest Cameroon was higher than that for coffee, contributing between 5 and 37 per cent of household cash revenues (Laird et al., 1996). Kola nut exports from Ghana were estimated at 6 million tons and 7.5 million tons in 1988 and 1989, with a value of US$1.03 million and US$1.48 million respectively (Laird et al., 1996).

In southern Cameroon, Ntamag (1997) reported that Garcinia lucida is found in little disturbed forests, normally common access areas. Diaw (1997) also found Garcinia lucida in these near virgin forests and that it is governed by an open access regime limited to the members of the community, i.e., the village. For Garcinia kola, the most important niches are cocoa plantations, virgin forests and fallow land (Ntamag, 1997). Van Dijk (1995) found

Garcinia kola occurred only in forest lands at distance from human settlements. She also noted that the species might be endangered by over-exploitation due to the use of the bark in production of palm wine. In an inventory of 11 complete transects in southern Cameroon, Van Dijk (1995) did not find a single Garcinia lucida tree and suggested that this could be due to high harvesting levels. Guedje (1996) also found that Garcinia lucida grows in dense populations with a preference for undisturbed forest habitats. This species cannot recover and will eventually die if the bark is removed over a large surface around the stem (Guedje 1996). She identified sites with more than 50 per cent of dead trees.

In January 1994, the devaluation of the CFA Franc severely affected Cameroon and other countries in Central Africa which are part of the CFA zone.⁷ The devaluation increased the price of beer and whisky and made palm wine more attractive in rural and urban areas (Ndoye, 1994). As already noted, the barks of Garcinia lucida and Garcinia kola are used in the production of palm wine. Garcinia lucida is also considered to be a social good because of its medicinal value and anti-poison properties (Diaw 1997). For these reasons pressure on both species may increase. Furthermore, because Garcinia lucida is governed by an open access regime, the trees could become more vulnerable to externalities created by the opportunistic behaviour of a few forest dwellers resulting from higher market demands for the bark. Opportunistic behaviour, as defined by Williamson (1985), is a way of acting in self-interest with guile. It brings immediate rewards to the individual and imposes high costs on society (Shaffer et al., 1987).

The authors believe that research on Garcinia lucida and Garcinia kola is wanting many key areas.

· There is no information on the sustainable rate of bark harvest that would preserve Garcinia lucida and Garcinia kola trees.

· Sustainable harvesting techniques that would guarantee the survival of Garcinia lucida and Garcinia kola trees have yet to be designed by researchers. This could help mitigate future pressure on these resources.

· Research on domestication of Garcinia lucida and Garcinia kola, and their subsequent cultivation by forest dwellers, is not currently included in the agenda for NWFP research in Central Africa.

This paper argues that while NWFP provide valuable alternative income-generating opportunities for rural dwellers, which increases the need to develop these markets, the harvesting practices are not always sustainable. This emphasises the difficult trade-offs between improving the livelihood of forest dependent people and forest conservation. For the four NWFP studied, the harvesting practices for Irvingia spp. and Cola acuminata are more sustainable than those used for Garcinia lucida and Garcinia kola bark. For the first group, the tree is preserved, whereas when harvesting barks, the survival of the tree is not always guaranteed. It is worth noting, however, that collection of fruits and kernels for consumption and sales may have a negative impact on the future population structure of the trees (Peters, 1994).

The second section of the paper is a brief description of the methodology of the study. Section three describes the distribution and utilisation of the four NWFP species analysed in this paper: Irvingia spp., Cola acuminata, Garcinia lucida and Garcinia cola. Next, the quantity of NWFP marketed and traders' marketing margins are discussed. Section five analyses the welfare effects of marketing Irvingia spp. and Cola acuminata compared to Garcinia lucida and Garcinia kola, while section six discusses the type of research needed to deal with the thinness of NWFP markets and (potential) forest resource degradation. The final section presents the conclusions and implications of the study.

2. Methodology of the study

Twenty-eight markets in the Humid Forest Zone (HFZ) of Cameroon were surveyed in 1995 and 1996. The HFZ comprises five (Centre, South, Littoral, East, South West) of the 10 provinces in Cameroon. Data from the 1987 census indicate that 45 per cent of the total population of Cameroon live in the HFZ. The zone is bordered on the south by the Republic of Gabon, Equatorial Guinea and Congo Brazzaville, on the east by the Central African Republic, and on the northwest by the Republic of Nigeria. The HFZ covers 270 162 square kilometres representing 58 per cent of the territory of Cameroon.

Irvingia spp. (wild mango nut), Cola acuminata (kola nut), Garcinia lucida and Garcinia kola were the NWFP selected for study in this paper. The marketed quantities of the kernels and the fruits are used for Irvingia spp and Cola acuminata, while the marketed quantities of the bark are considered for Garcinia lucida and Garcinia kola. The data reported for all these NWFP are based on 29 weeks of activity from January to July in both 1995 and 1996. Two hundred and sixty-seven (267) traders were interviewed in 1995 and 347 in 1996, representing 24 per cent and 27 per cent of the estimated total number of traders operating in the selected markets studied each year.

3. Distribution and utilisation of NWFP in Central Africa

Irvingia gabonensis

Harris (1993) identified two species of Irvingia used for food in Cameroon: Irvingia gabonensis and Irvingia wombolu. This section will only discuss the distribution and utilisation of the first. The Irvingia gabonensis tree is found throughout the whole Humid Forest Zone of Cameroon (i.e., evergreen, semi-deciduous and transition forests). The species is distributed from the south of Senegal to Zaire (Vivien and Faure 1985) in the low land warm humid forest of West and Central Africa. In West Africa, Irvingia gabonensis is found in Senegal, Guinea Conakry, Liberia, Sierra Leone, Côte d'Ivoire, Ghana, Togo, Benin and Nigeria. In Central Africa, it grows in Cameroon, Gabon, Central African Republic and Congo (Brazzaville).

Irvingia gabonensis commonly called "dika nut", belongs to the family of Irvingiaceae. The fruit is similar to a small "domesticated" mango, and is green when raw and yellow when ripe. The ripe fruit contains a lot of fibres. The pulp of the fresh fruit is eaten and the nut is cracked to obtain the Irvingia gabonensis kernel. This kernel can easily be separated into two parts.

The kernel is the most valued part of the Irvingia gabonensis tree. It is a delicious condiment for soup and an important source of cash income for farmers in the Humid Forest Zone of Cameroon. The kernel can be pounded and used as a substitute or complement for groundnut in the diet of forest dwellers in the Humid Forest Zone of Cameroon. The sauce made with Irvingia gabonensis kernel can be stored for three to four days without being refrigerated. The paste obtained from the crushed kernel is called "dika bread" in Gabon and "etima " in the Centre province of Cameroon and can be stored for more than a year after it is dried in the sun. The Irvingia gabonensis kernel can be stored for up to a year if completely dry.

Cola acuminata (kola nut)

Cola acuminata commonly called "abata cola", belongs to the family of Sterculiaceae and is found in the green forests of Cameroon. It is distributed from Nigeria to Congo (Vivien and Faure, 1985). The tree produces fruits that are consumed and sold because of the stimulant properties of the nut. Cola acuminata is the basis of an important trade within Cameroon and between Cameroon and neighbouring countries.

Garcinia kola

Garcinia kola belongs to the family of Clusiaceae (Gutiferae) and is found in the Humid Forest Zone of Cameroon. The tree is distributed from Sierra Leone to Gabon and Zaire (Democratic Republic of Congo) (Vivien and Faure, 1985). This encompasses part of the lowland warm humid tropic of West and Central Africa. In West Africa, Garcinia kola grows in Sierra Leone, Côte d'Ivoire, Ghana, Togo, Benin and Nigeria. In Central Africa, it is found in Cameroon, Gabon, Central African Republic and Equatorial Guinea. Garcinia kola is commonly called "bitter kola" by local people, which is the name of the aphrodisiac fruit that the tree produces. It is an important product in the Humid Forest Zone of West and Central Africa because it is eaten and also contributes to the revenues of rural households.

The bark of the Garcinia kola tree is used in palm wine to give the wine a bitter taste and to make it stronger. This practice is common in the Centre province of Cameroon and also in Gabon. The bark when soaked into water can be used as a treatment for intestinal worms and to cure stomach pain. Mixed with kola nut, the fruit of Garcinia kola aids digestion.

Garcinia lucida

Garcinia lucida belongs to the family of Clusiaceae (Gutiferae) and in Cameroon is found in the Humid Forest Zone. The bark is an anti-venom. It is mostly used by farmers who tap palm wine after harvesting the tree. According to farmers, the bark of Garcinia lucida gives a stronger taste to the palm wine.

4. NWFP markets and margins of traders

Volume and value of NWFP marketed

The total quantity of sales recorded in the sample for Irvingia spp., Cola acuminata, Garcinia kola and Garcinia lucida amounted to 138 tons in 1995 and 85 tons in 1996, with a value of 84 072 000 CFA Francs and 77 262 100 CFA Francs respectively (Table 1). A projection of the total number of traders operating in the 28 markets would yield an aggregate quantity of 677 tons in 1995 and 272 tons in 1996 with a total value of 361 558 000 CFA F in 1995 and 254 402 000 CFA F in 1996 (Table 2). (This aggregate figure is obtained by assuming that the same percentage of traders in the sample handling a particular NWFP in a given market can be generalised for the total number of traders operating in that market.) This is equivalent to US$ 753 000 in 1995 (480 CFA F= US$1) and US$ 499 000 in 1996 (510 CFA F= US$1). The projected sales of Garcinia kola and Garcinia lucida barks represented 57 tons in 1995 and 37 tons in 1996 with a total value of 14 331 000 CFA F (US$ 30 000) in 1995 and 11 977 000 CFA F (US$ 23 500) in 1996.

Irvingia spp. is the NWFP which is marketed by the highest percentage of traders, followed by Cola acuminata, Garcinia lucida bark and Garcinia kola bark. In 1995, Cola acuminata was the most important NWFP in terms of amount traded, followed by Irvingia spp., Garcinia lucida and Garcinia kola barks. In 1996, the order of the importance of quantity marketed changed slightly, with Irvingia spp. being the most important NWFP, then Cola acuminata and the two Garcinia. Value of the NWFP marketed displayed a different order of significance than the one related to volume. In both 1995 and 1996, the value of sales was greatest for Irvingia spp., followed by Cola acuminata, Garcinia lucida and Garcinia kola barks. The quantity and value of NWFP marketed reflect the typical characteristics of thin markets caused by a combination of environmental and economic factors. This will be discussed further in the next section

Table 1: Selected NWFP markets in the Humid Forest Zone, 1995 and 1996*

	% of Traders		Sales by Sample Interviewed
NWFP	Selling NWFP		Quantity (kg)		Value (CFA Francs)
	1995 (n=267)	1996 (n=347)	1995	1996	1995	1996
Irvingia spp.	45	47	36 390	34 400	43 000 000	47 534 600
Cola acuminata	19	24	80 960	31 650	35 884 000	23 660 300
Garcinia lucida (bark)	2	4	14 600	13 500	3 726 000	4 882 500
Garcinia kola (bark)	1	1.5	5 970	5 560	1 462 000	1 184 700
Total	67	76.5	137 920	85 110	84 072 000	77 262 100

* 29 weeks for all NWFP in 1995 and in 1996

Table 2: Sales projected over all NWFP markets in the Humid Forest Zone, 1995 and 1996*

	Projection of sales for all NWFP Markets
NWFP	Quantity (Kg)		Value CFA
	1995	1996	1995	1996
Irvingia spp.	111 000	107 100	125 237 000	147 769 000
Cola acuminata	509 000	127 400	221 990 000	94 656 000
Garcinia lucida (bark)	40 600	27 300	10 360 000	9 867 000
Garcinia kola (bark)	16 200	9 900	3 971 000	2 110 000
Total	676 800	271 700	361 558 000	254 402 000

* 29 weeks for all NWFP in 1995 and in 1996

From 1995 to 1996, the combined quantity of both Garcinia lucida and Garcinia kola bark marketed decreased by 7 per cent, while the actual value of sales increased relatively by 17 per cent (Table 1).

Thinness of NWFP markets

The quantity and value of the NWFP studied varied significantly between 1995 and 1996, especially for Cola acuminata (Table 1). This is characteristic of thin markets. Markets are defined as thin if a small change in production has a large effect on the quantity marketed. This changes the role of markets in assembling and distribution of NWFP from year to year. There are various ways to deal with thin markets, which will be discussed further in the conclusions.

From 1995 to 1996, the actual quantity marketed of Irvingia spp. declined by 5 per cent while the value of sales increased by 11 per cent (Table 1). In 1995, the markets of Ebolowa, Abang Minko (border between Cameroon and Gabon) and Kye-Ossi (border between Cameroon and Equatorial Guinea) accounted for 59 per cent of the total quantity of Irvingia spp. marketed; in 1996, these markets traded 41 per cent of total quantity marketed. The Mfoundi market (Yaounde) became more important, accounting for 17 per cent of quantity marketed in 1996 against 7 per cent in 1995.

The decline in the quantity of Irvingia spp. marketed can be explained by the fact that the production of Irvingia wombolu was less important in 1996 than in 1995. As noted by Ndoye et al., (1997) there are two species of Irvingia that are traded in the Humid Forest Zone of Cameroon and its borders: Irvingia gabonensis, which fruits from June to August; and Irvingia wombolu, which fruits from January to March. This significantly affected the quantity traded at Ebolowa (Table 3). Although the quantity of Irvingia spp. marketed increased in several markets, this was not sufficient to offset the decline in other markets, especially Ebolowa, one of the key regional markets for this product.

Figure 1 Response of price to quantity of main products

Table 3: Changes in the trade of Cola acuminata, Irvingia spp., Garcinia kola and Garcinia lucida barks in selected markets of the Humid Forest Zone and its borders

Markets	Cola acuminata		Irvingia spp.		Garcinia kola (Bark)		Garcinia lucida (Bark)
	Difference 96-95 (bags)	Relative Change %	Difference 96-95 (bags)	Relative Change %	Difference 96-95 (bags)	Relative Change %	Difference 96-95 (bags)	Relative Change %
Mbalmayo	+20	+147	+36	+259	-	-	+76	+40
Edea	+15	+157	-	-	-	-	-	-
Mfoundi (Yde)	-81	-58	+55	+117	-	-	-	-
Sangmelima	+6	+182	-42	-62	-	-	-	-
Mokolo (Yde)	-43	-60	-24	-72	-	-	-	-
Ebolowa	+32	+124	-137	-59	-	-	-324	-66
Abang Minko	+3	+35	-6	-4	-10	-7	+51	+69
Kribi	-	-	+28	+148	-	-	-	-
Bertoua	-2	-95	-8	-81	-	-	-	-
Abong Mbang	-	-	-10	-75	-	-	-	-
Bafia	-673	-99	+7	+612	-	-	-	-
Monatele	-	-	+0.4	+17	-	-	-	-
Zoatele	+5	+407	+1.2	+46	-	-	+80	+138
Akonolinga	-40	-92	-17	-84	-	-	-	-
Batouri	-	-	-2	-24	-	-	-	-
Kenzou	+27	+170	-	-	-	-	-	-
New-Bell	+83	+290	+15	+226	-	-	-	-
Kumba	-0.8	-12	+2.3	+72	-	-	-	-
Limbe	-	-	-12	-76	-	-	-	-
Saa	-	-	-1.1	-68	-	-	-	-
Ombessa	-209	-90	-	-	-	-	-	-
Kye-Ossi	-	-	+10	+98	-	-	-	-

The quantity marketed and the value of sales of Cola acuminata decreased from 1995 to 1996 by 61 per cent and 34 per cent respectively. In 1995, the markets of Mfoundi (Yaounde), Bafia and Ombessa accounted for 82 per cent of the total quantity of Cola acuminata, but only 18 per cent in 1996. The New Bell market (Douala) became more important, handling 22 per cent of quantity marketed in 1996 compared to 2 per cent in 1995. The most important markets for Cola acuminata in 1995 were Bafia and Ombessa, located further north of Yaounde. The following year, markets further south of Yaounde (Mbalmayo, Ebolowa), markets in the Littoral province (Edea, New-Bell) and the frontier market with Central African Republic (Kenzou) became more important in assembling and distribution of Cola acuminata (Table 3).

The decline in the quantity traded from Bafia and Ombessa can be explained by a drastic reduction in production, resulting from physical, biological and/or ecological factors. Further investigation is necessary, because of the NWFP studied, Cola acuminata is the most frequently planted by forest dwellers, especially in their cocoa plantations. Despite its widespread cultivation, it has had the greatest decline in the quantity marketed.

Between 1995 and 1996, the quantity of Garcinia lucida bark marketed declined by 8 per cent while the actual value of sales increased by 31 per cent. In 1995, the markets of Mbalmayo, Ebolowa, Abang Minko (border between Cameroon and Gabon) accounted for 93 per cent of the total quantity of Garcinia lucida bark marketed and 74 per cent in 1996. The market of Zoatele became more important and handled 18 per cent of quantity marketed in 1996 compared with 7 per cent in 1995. The trade of Garcinia lucida bark is dominated by markets south of Yaounde (Mbalmayo, Zoatele) and the frontier market with Gabon (Abang Minko) (Table 3). The overall decline in the quantity of bark marketed was due to the drastic decrease in amount traded at Ebolowa, which could not be offset by increases at Mbalmayo, Abang Minko and Zoatele. Furthermore, the reduction may be due to excessive pressure on the resource in the areas supplying the Ebolowa market.

For Garcinia kola bark, the quantity marketed and the value of sales decreased by 7 per cent and 19 per cent respectively from 1995 to 1996. All trade in both years was undertaken at Abang Minko. This bark is very important for Gabonese consumers who use it to improve the taste (bitter) of palm wine and to increase the level of alcohol.

The foregoing discussion demonstrates that NWFP markets are quite complex, highlighting the difficulties and the challenges associated with their study. The relationship between quantity marketed and prices received for the four products is related to the decline in volume of trade (see Figure 1). There were two different reasons for this decline in the amount of product changing hands in the markets.

1. For Irvingia spp., Cola acuminata and Garcinia lucida bark, reduced production of the product led to a decline in the quantity marketed, especially for Cola acuminata. This may have been caused by one of the following conditions or a combination:

· a reduction in both production and demand from traders and consumers for the NWFP;

· a linear relationship between production and quantity marketed by forest dwellers, meaning that a reduction in production translates to a reduction in quantity marketed due to poor long-term storage facilities in a year of a good harvest; and/or

· a reduction in production leading to a decline in quantity marketed by forest dwellers due to a fixed amount (threshold) being held by the household for home consumption.

Table 4: The three most important origins of Irvingia spp. traded in 1995 and 1996

MARKETS	1995			1996
	Origin 1	Origin 2	Origin 3	Origin 1	Origin 2	Origin 3
Mbalmayo	Mbalmayo	-	-	Mbalmayo	-	-
Edea	Mfoundi (Yde)	-	-	Edea	-	-
Mfoundi (Yde)	Mfoundi (Yde)	Mokolo (Yde)	Ebolowa	Mfoundi (Yde)	Mbalmayo	Ngomedzap
Sangmelima	Sangmelima	Ebolowa	Ezezang	Sangmelima	Elom	Ebolowa
Mokolo (Yde)	Mokolo (Yde)	Mfoundi	Mbalmayo	Mokolo (Yde)	-	-
Ebolowa	Ebolowa	Route Kribi	Nkok-Ekie	Ebolowa	Kribi	Mbalmayo
Abang Minko	Ebolowa	Essos (Yde)	Ngozip	Ebolowa	Kribi	Ambam
Kribi	Ebolowa	Biwong	Kribi	Kribi	Ebolowa	Likoube
Bertoua	Mfoundi (Yde)	Bertoua	-	Bertoua	-	-
Abong Mbang	Abong Mbang	-	-	Abong Mbang	Medjo	-
Bafia	Bafia	-	-	Bilig-Bidig	-	-
Monatele	Monatele	-	-	Monatele	-	-
Zoatele	Mfoundi (Yde)	Ebolowa	-	Ebolowa	Mfoundi (Yde)	-
Akonolinga	Akonolinga	Ebolowa	Mfoundi	Akonolinga	Mfoundi (Yde)	-
Batouri	Batouri	Kadey	Mfoundi	Batouri	Kadey	Mfoundi
Kenzou	-	-	-	-	-	-
New-Bell (Dla)	New-Bell	Saa	-	New-Bell	Saa	Mfoundi
Kumba	Kumba	-	-	Kumba	-	-
Limbe	Limbe	Muyuka	Mamfe	Limbe	Mamfe	-
Saa	Saa	-	-	Saa	-	-
Ombessa	-	-	-	-	-	-
Kye-Ossi	Ebolowa	-	-	Ebolowa	-	-

Table 5: The three most important origins of Cola acuminata traded in 1995 and 1996

MARKETS	1995			1996
	Origin 1	Origin 2	Origin 3	Origin 1	Origin 2	Origin 3
Mbalmayo	Mbalmayo	-	-	Mbalmayo	Mvog-Mbi	Mengueme
Edea	Edea	-	-	Edea	-	-
Mfoundi (Yde)	Mfoundi	Bokito	Mbangassina	Mfoundi	Bafia	Monatele
Sangmelima	Mfoundi	-	-	Sangmelima	Mbalmayo	Mfoundi
Mokolo (Yde)	Mokolo	Bokito	Ndom	Mokolo	Ndom	-
Ebolowa	Ebolowa	-	-	Ebolowa	Route Mvangan	Biboulman
Abang Minko	Mfoundi	Abang Minko	-	Mfoundi	-	-
Kribi	-	-	-	-	-	-
Bertoua	Obala	-	-	Bertoua	-	-
Abong Mbang	-	-	-	-	-	-
Bafia	Bafia	Bokito	Yabeng	Tiko (Bafia)	Banda	-
Monatele	-	-	-	-	-	-
Zoatele	Ebolowa	-	-	Ebolowa	Mfoundi	-
Akonolinga	Akonolinga	-	-	Akonolinga	-	-
Batouri	-	-	-	-	-	-
Kenzou	Mbanga	Loum	-	Mbanga	Batouri	-
New-Bell	Loum	-	-	Saa	New-Bell	Mbangassina
Kumba	Kumba	-	-	Kumba	-	-
Limbe	-	-	-	-	-	-
Saa	-	-	-	-	-	-
Ombessa	Bouraka	-	-	Melen	-	-
Kye-Ossi	-	-	-	-	-	-

Table 6: The three most important origins for Garcinia lucida bark traded in 1995 and 1996

MARKETS	1995			1996
	Origin 1	Origin 2	Origin 3	Origin 1	Origin 2	Origin 3
Mbalmayo	Mvog-Mbi	Ebolowa	-	Mbalmayo	Mvog-Mbi	-
Edea	-	-	-	-	-	-
Mfoundi (Yde)	-	-	-	-	-	-
Sangmelima	-	-	-	-	-	-
Mokolo (Yde)	-	-	-	-	-	-
Ebolowa	Ebolowa	Ngoul Essama	-	Ebolowa	Ebolowa-Assi	-
Abang Minko	Ebolowa	-	-	Ebolowa	-	-
Kribi	-	-	-	-
Bertoua	-	-	-	-
Abong Mbang	-	-	-	-
Bafia	-	-	-	-
Monatele	-	-	-	-
Zoatele	Ebolowa	-	-	Ebolowa	Mvog-Mbi	Lolodorf
Akonolinga	-	-	-	-	-	-
Batouri	-	-	-	-	-	-
Kenzou	-	-	-	-	-	-
New-Bell	-	-	-	-	-	-
Kumba	-	-	-	-
Limbe	-	-	-	-
Saa	-	-	-	-
Ombessa	-	-	-	-
Kye-Ossi	-	-	-	-

Table 7: The three most important origins for Garcinia kola bark traded in 1995 and 1996

MARKETS	1995			1996
	Origin 1	Origin 2	Origin 3	Origin 1	Origin 2	Origin 3
Mbalmayo	-	-	-	-	-	-
Edea	-	-	-	-	-	-
Mfoundi (Yde)	-	-	-	-	-	-
Sangmelima	-	-	-	-	-	-
Mokolo (Yde)	-	-	-	-	-	-
Ebolowa	-	-	-	-	-	-
Abang Minko	Ebolowa	-	-	Ebolowa	Melanga	Akom
Kribi	-	-	-	-	-	-
Bertoua	-	-	-	-	-	-
Abong Mbang	-	-	-	-	-	-
Bafia	-	-	-	-	-	-
Monatele	-	-	-	-	-	-
Zoatele	-	-	-	-	-	-
Akonolinga	-	-	-	-	-	-
Batouri	-	-	-	-	-	-
Kenzou	-	-	-	-	-	-
New-Bell	-	-	-	-	-	-
Kumba	-	-	-	-	-	-
Limbe	-	-	-	-	-	-
Saa	-	-	-	-	-	-
Ombessa	-	-	-	-	-	-
Kye-Ossi	-	-	-	-	-	-

Table 8: Net margins (in CFA) for the marketing of NWFP in 1995 and 1996

	Total net margin (x 1000 CFA)		weekly margin per trader		margin as % of sales value
NWFP	1995	1996	1995	1996	1995	1996
Irvingia spp.	12 988	11 292	3 800	2 400	30%	24%
Cola acuminata	6 362	6 205	4 400	2 600	18%	26%
Garcinia lucida (bark)	972	1 627	5 600	4 000	26%	33%
Garcinia kola (bark)	544	289	6 200	2 000	37%	24%

Table 9: Traders' marketing margins in selected markets (CFA Francs)

Market		1995	1996
Cola acuminata
Mfoundi (Yde)		5 800	2 200
Bafia		13 900	2 600
Ombessa		8 000	1 500
New Bell (Dla)		3 100	6 800
Kenzou		3 700	5 800
Irvingia spp.
Mfoundi (Yde)		2 000	2 900
Ebolowa		5 500	1 800
Abang Minko		13 700	7 800
New Bell (Dla)		1 500	3 700
Limbe		1 800	900
Kye-Ossi		2 800	8 600
Garcinia lucida
Mbalmayo		6 000	4 800
Ebolowa		10 200	7 800
Abang Minko		1 600	2 300
Zoatele		4 000	5 000
Garcinia kola
Abang Minko	6 200		2 000

2. For Garcinia kola bark, demand was reduced because the distribution market for this product, Abang Minko, was performing poorly due to the breakdown of the ferry that affected the number of Gabonese crossing the border. This led to an overstocked market translating to lower prices. This further affected the quality of the bark, which was not the case for Irvingia spp. because of better storage potential in the short term.

Irvingia spp., Cola acuminata and Garcinia lucida all behaved as would be expected, displaying a similar price elasticity of supply. Prices increased as quantity marketed declined, and the greater the reduction in quantity marketed the higher the increase in price (Figure 1). The fourth NWFP, Garcinia kola bark, responded in a contrary fashion. The reduction in the quantity marketed was not the result of a decreased supply (as with the other three products) but because of a market glut in Abang Minko due to the ferry problem. When traders arrived to sell in the market, they found fewer buyers generating a lower demand and a consequent price reduction.

The role in assembling and distribution of several of the markets studied changed from 1995 to 1996. For Irvingia spp., Cola acuminata and Garcinia lucida bark, the increase in the quantity marketed in a few markets was not enough to offset the decline in other markets. This translated to an overall decline in the quantity marketed. Declining production of these NWFP may have been due to environmental factors like drought, excessive pressure on the resource resulting in reduced availability, and changes in weather patterns altering the period of flowering and fructification of the NWFP.

In 1995, there appeared generally to be stable relationships between selling markets and their immediate surrounding areas (as portrayed when the selling market is the same as the origin of the NWFP) as a source of supply for the four NWFP studied (Tables 4 to 7). For most of the selling markets, the origins of the NWFP changed from 1995 to 1996, which reinforces the argument of unstable or erratic supplies affecting (positively or negatively) the role individual markets play in assembling and distribution between the two years. Peters (1994) points out that "very few tropical forest species produce reliable fruits during a well-defined, predictable season each year". This implies the need for long-term monitoring of NWFP markets to better characterise them by linking the ecology of the NWFP (i.e., life cycle, type of output produced, abundance in the forest and the size-class distribution of natural population) (Peters, 1994), climatic factors, and other related environmental variables, as well as socio-economic and cultural factors.

Marketing margins of traders

In 1995 the average weekly net margins per trader were higher for Garcinia lucida and Garcinia kola barks than for Irvingia spp. and Cola acuminata (Table 8). The following year, the net margins were highest for Garcinia lucida bark and were comparable for Irvingia spp. and Cola acuminata. This is also reflected in the total marketing margins as a percentage of the value of sales, which is higher for barks. Clearly, traders will continue to demand these NWFP (barks), especially for their use in palm wine production.

The reasons for the differences in the marketing margins between fruits/kernels and barks are complex, and relate to different storage periods, different levels of perishability and to the fact that the quantity of fruits/kernels traded is significantly higher than the quantity of barks traded. As a result, the larger markets (Cola acuminata and Irvingia spp.) are subject to lower weekly net margins. From 1995 to 1996, the average weekly net marketing margins for each trader who sold Irvingia spp., Cola acuminata, Garcinia lucida bark and Garcinia kola bark declined by 37 per cent, 41 per cent, 29 per cent and 68 per cent respectively (Table 8). Garcinia kola bark, which is traded exclusively at Abang Minko, showed the greatest decline in margins.

In 1995, the markets of Abang Minko (border between Cameroon and Gabon) and Ebolowa had the highest weekly net marketing margins per trader for Irvingia spp., amounting to CFA F 13 700 and CFA F 5 500 respectively (Table 9). The next year, Kye-Ossi (border between Cameroon and Equatorial Guinea) and Abang Minko (border between Cameroon and Gabon) produced the highest margins of 8 600 CFA F and 7 800 CFA F respectively. The lower weekly marketing margins in 1996 at Abang Minko was caused by the breakdown of the ferry, which forced traders to sell at reduced margins. Although the average weekly marketing margins declined at Abang Minko in 1996, it remains the case that sales of Irvingia spp. are more lucrative at markets at the borders (Table 9).

Bafia and Ombessa had the highest weekly net marketing margins per trader for Cola acuminata in 1995 with 13 900 CFA F and 8 000 CFA F respectively. In 1996, the markets of New Bell (Douala) and Kenzou (border between Cameroon and Central African Republic) had the highest weekly net marketing margins totalling 6 800 CFA F and 5 800 CFA F. The increases over the two years were 119 percent at New Bell and 57 percent at Kenzou (Table 8).

For Garcinia lucida bark, the markets of Ebolowa and Mbalmayo had the highest weekly net marketing margins per trader in 1995, of 10 200 CFA F and 6 000 CFA F. In 1996, Ebolowa and Zoatele margins were highest weekly with 7 800 CFA F and 5 000 CFA F. The average weekly marketing margins per trader decreased by 24 per cent for Ebolowa and increased by 25 per cent for Zoatele from 1995 to 1996 (Table 8).

In both 1995 and 1996, Abang Minko was the only market where Garcinia kola bark was marketed. The weekly net marketing margins per trader declined from 6 200 CFA F in 1995 to 2 000 CFA F in 1996. This drastic decline was due to the drop in the average sale prices of the bark over the two years (Table 1), and the additional transport and storage costs created by the non-functioning of the market. Traders were forced to return home without selling after incurring some costs of bringing their product to the market. The ferry of Abang Minko broke down in May-June 1995 and was still not repaired during the time of the survey in 1996. Furthermore, there were some problems on the border between Cameroon and Gabon, which made the market non-operational in several instances. Traders were not informed of these prior to their arrival at the market, which could have helped them to avoid the costs of needlessly transporting their product. This also lowered the quality of the bark.

5. Welfare effects of marketing different NWFP

This section investigates the welfare effects of selling Irvingia spp. and Cola acuminata compared to selling the barks of Garcinia lucida and Garcinia kola. Basic questions need to be asked:

· Are forest dwellers better off when they sell Irvingia spp. and Cola acuminata than when they sell Garcinia lucida and Garcinia kola barks?

· What is a possible solution to the trade-offs that are likely to occur?

Welfare effects for forest dwellers

Forest dwellers receive a certain percentage of the wholesale and retail prices when they sell their NWFP. Gatherers received a higher percentage of the retail prices when they sell Irvingia spp. than when they sell Garcinia lucida and Garcinia kola barks (Table 10). The difference in the percentage of the retail prices received by forest dwellers for sales within Cameroon is not significant. The differences when the NWFP is sold at the border with Gabon may be due to factors such as transportation costs and generally better margins for traders in frontier markets, especially for Irvingia spp. This implies that forest dwellers should be spending more time in the collection and sales of Irvingia spp. than the harvesting and sales of barks. Since forest dwellers are searching for income-generating enterprises, the above knowledge may not be an important factor in changing their behaviour.

Welfare effects for traders

As previously discussed, the weekly net marketing margins achieved by traders selling Garcinia lucida and Garcinia kola barks were higher or comparable to those of Irvingia spp. and Cola acuminata. This suggests that traders will continue to demand the barks of Garcinia lucida and Garcinia kola because of their importance in palm wine production, of which consumption has increased significantly since the devaluation of the CFA Franc.

Trade-offs are likely to be necessary when trying to maximise returns to forest dwellers and traders, as well as conserve the resource. This situation is discussed below as part of the need for research.

6. Future research needs

Reducing the thinness of NWFP markets

To reduce the thinness of NWFP markets, several conditions must be met. The most important is to increase aggregate production. This could be done through conservation or domestication.

· Conservation can be achieved by determining the harvesting rate that will preserve the NWFP tree, using sustainable harvesting techniques for bark extraction and educating forest dwellers to use these methods efficiently.

· Domestication and cultivation by forest dwellers will reduce the length of the production cycle of the NWFP while preserving the characteristics that are important to consumers.

Domestication may require longer-term research and more financial resources than conservation, which is of a shorter-term strategy. The current economic crisis, structural adjustment policies and the devaluation of the CFA F, have all increased poverty in rural and urban areas (World Bank, 1995), and resulted in return migration from urban to rural areas (Bela, 1994; Pokam, 1997). These pressures seem to reduce the possibility that conservation is a viable option. The domestication of NWFP and their cultivation in multi-strata agroforestry systems (like the current cocoa and coffee-based systems in Cameroon) could be a more viable alternative in reducing the pressure from the forests under present circumstances.

However, increasing the aggregate supply of NWFP is a necessary but not sufficient condition for reducing the thinness of NWFP markets. Other conditions that need to be met are improved access to market information and infrastructure, reduced transaction costs (information, contracting and enforcement costs), and improved institutional support (credit) for forest dwellers and traders. Without these, efforts to domesticate and plant particular NWFP may fail.

Potential resource degradation

Potential resource degradation from debarking

There are debatable figures in the various inventories carried out in the forest zone with regard to the available stock of Garcinia lucida from the forest. These discrepancies can be explained by the relative abundance of each species in natural conditions. These in turn could explain why the variation in the abundance and the distribution of certain NWFP can be high, even over a relatively small area (van Dijk, 1998).

In her inventory in southern Cameron, van Dijk (1995) did not find any Garcinia lucida trees. Ntamag's (1997) survey showed Garcinia lucida only in virgin forests at an average rate of 4 trees per hectare. Guedje (1997) recorded 58 live stems of Garcinia lucida trees per hectare in an inventory, which covered 8.45 hectares. She also found those trees with diameters equal to or greater than 10 centimetres were exploited (debarked) most - 32 live stems per hectare were noted in that category. All the above studies were carried out in the Tropenbos site, a relatively small area of 200 000 hectares.

The density of trees per hectare of Garcinia kola also appears to vary widely, according to the literature. Tchatchou (1997 cited by Doucet and Koufani 1997) found a density of 0.01 tree per hectare in eastern Cameroon. Doucet and Koufani (1997) themselves did not discover any Garcinia kola trees in their inventory carried out in eastern Cameroon. In southern Cameroon, van Dijk (1995) found a density of 0.4 tree per hectare for Garcinia kola.

There are three shortcomings in the current research of Garcinia lucida and Garcinia kola. There is presently no information about the sustainable rate of bark harvest per tree that would preserve it perennially. The literature reveals a high variability in the data about available stocks of Garcinia lucida and Garcinia kola. This may indicate that the potential (stocks) of the resource are not known precisely, and/or that there are problems in the methods used in the inventories to assess the availability of the resource. Peters (1994) cautioned that inventories should be carried out in different forest types. In addition to these problems, there is no information on average yield per tree.

The first two issues are important to reconcile livelihood and conservation objectives. Data on density and average yield per tree would allow calculation of the potential area of forest affected by debarking both Garcinia lucida and Garcinia kola.

Cunningham and Mbenkum (1993) assumed that the average yield of Prunus africana to be 55 kg per tree. We assumed similar yields for Garcinia lucida and Garcinia kola, because of the lack of information in the current research on those products. One factor that could justify this assumption is the similar height of the species; 25 and 30 metres for Prunus africana (BDCPC, 1997), 25 metres for Garcinia lucida (Guedje, 1996), and 35 metres for Garcinia kola (Vivien and Faure, 1996).

Table 10: Proportion of prices received by forest dwellers in 1996

Table 11: Potential number of trees debarked in 1995 and 1996*

	Garcinia lucida	Garcinia kola
1995	738 trees	295 trees
1996	496 trees	180 trees

*Based on an average yield of 55 kg per tree and the projection from Table 1.

Table 12: Potential number of hectares of forest affected (hectares)

_ - Based on Ntamag (1997)

_ - Based on Guedje (1996)

_ - Based on Tchatchou (1997) cited in Doucet and Koufani (1997)

_ - Based on van Dijk (1995)

Based on the above assumption about the average yields of the trees (55 kg per tree), the potential number of Garcinia lucida trees debarked was 738 in 1995 and 496 in 1996. For Garcinia kola trees, the number debarked would be 295 in 1995 and 180 in 1996. These figures represent a reduction in the number of Garcinia lucida trees debarked of 33 per cent and 39 per cent for Garcinia kola between 1995 and 1996 (Table 11). The possible area of forest affected by bark extraction (Table 12) can be calculated by using the different estimates of the density per hectare for Garcinia lucida and Garcinia kola from Ntamag (1997), Guedje (1996), Doucet and Koufani (1997) and van Dijk (1995). In 1995, the potential number of hectares of forest affected by the extraction of Garcinia lucida bark marketed varied between 13 and 185 hectares, with that of Garcinia kola varying between 738 and 29 500 hectares. In 1996, the estimates of area affected by Garcinia lucida bark extraction ranged between 9 and 124 hectares, while that of Garcinia kola was from 450 to 18 000 hectares (Table 12).

Fewer Garcinia lucida and Garcinia kola trees were debarked in 1996 than in 1995. This could have occurred because of a reduction in the number of mature trees resulting from the scarcity of tree resources. Demand for the products in markets at the borders, especially for Garcinia kola bark, may also have been lower. A combination of these factors may also have contributed to the reduction in trees debarked.

Role of research in finding a balance between welfare improvement and forest conservation

Improving welfare from forests and forest conservation is always a difficult challenge for researchers, policy makers and the international community. As the study showed, NWFP contribute to the livelihoods of forest-dependent people. For Garcinia lucida and Garcinia kola, it is to the advantage of traders to continue to provide market outlets for forest dwellers to sell NWFP, since weekly net marketing margins from barks were higher or comparable to those of fruits/kernels (Table 3). Furthermore, traders will seek these products because of international demand from Gabon and other neighbouring countries.

Garcinia lucida and Garcinia kola have been used for many decades by forest dwellers in a sustainable way. Beside their medicinal properties, these products have traditionally been used by forest dwellers to improve the taste of palm wine. Garcinia lucida has other social attributes because it is an antidote for poison that is highly valued in rural areas.⁸ However, when other factors contribute to the general economic situation, trade-offs are necessary between welfare and forest conservation. The decline in international markets for cocoa and coffee in the 1980s, the economic crisis that has affected Cameroon since 1986, structural adjustment programs which are accompanied by the lay-off of people, and the devaluation of the CFA F in January 1994, are all likely to increase pressure on the forests to provide alternative income-generating opportunities. In particular, the devaluation of the CFA Franc increased the price of beer and whisky, and these were substituted by palm wine and local whisky (odontol), which is made after distilling the fermented palm wine. The resulting boost in palm wine production will necessarily increase the pressure on palm trees (Elaeis guineensis, Raphia hookeri) and on Garcinia lucida and Garcinia kola. In eastern Cameroon, Oyono (1997) reported that young Raphia trees are now exploited, which was not the case ten years ago. In southern Cameroon, Guedje (1996) found sites where more than 50 per cent of Garcinia lucida trees were dead. According to information from the market interviews carried out, Garcinia kola trees that still bear fruit are sold by forest dwellers to traders who extract the bark. This practice was not common in the past. Before the devaluation of the CFA F, the price of an adult Garcinia kola tree varied between 10 000 and 15 000 CFA F; after devaluation the price of a tree was between 20 000 and 25 000 CFA F.

There could be a danger of a conflict between the social value of Garcinia lucida (and Garcinia kola) and the financial profitability of these NWFP as a result of emerging local and international markets. In a situation of economic crisis and an active search for alternative income-generating opportunities, private gains can override the hitherto sustainable social benefits that were derived from these NWFP, especially Garcinia lucida because it is governed by an open access regime. The displacement of social benefits by private gains from individual opportunistic behaviour can deplete forest resources and place a heavy cost on the larger community of forest dwellers who depend on these products.

Inventories show that there is a wide variation in the stocks available of Garcinia lucida and Garcinia kola, and there is no information on the sustainable harvest rate that would preserve these species. Furthermore, these NWFP are not on the research agenda of priority species to domesticate. Should researchers wait until the resources are depleted to be to effective action?

Reconciling improvement in welfare and forest conservation requires that researchers undertake both leading edge or strategic as well as problem-solving research. That is, scientists need to develop solutions to problems which may arise in the future, rather than merely responding after it is widely known that a given resource, important to forest dwellers, is depleted or under threat of disappearing.

Improved technologies are needed to enable forest-dependent people to continue to derive part of their livelihoods from NWFP while at the same time preserving the forest. Other required conditions are the access of forest dwellers to those technologies. This could be very important in reducing any future pressure on forests resulting from growing poverty and the search for alternative income-generating opportunities, as well as increased return migration from urban to rural areas. The best way to achieve the desired impact of meeting both livelihood and forest conservation goals is through multidisciplinary research where biological and social scientists work in partnership.

7. Conclusions and implications

NWFP are very important to the well-being of forest dwellers and for that reason they need to be produced on a sustained basis. This paper has shown the importance of NWFP markets by comparing the marketing of Irvingia spp. and Cola acuminata on the one hand, and Garcinia lucida bark and Garcinia kola bark on the other.

The estimated value of the four NWFP in the markets surveyed during the first half of 1995 and 1996 amounted to USD 753 000 and USD 499 000 respectively. Markets appear to be thin and thus change their role in assembling and distribution of NWFP from one year to the next. Related to this changing function, the aggregated quantities marketed of the four NWFP declined from 1995 to 1996. Over the two years, the quantity marketed increased in few markets, but this was not sufficient to offset the decline in other markets.

The decrease in the quantity of NWFP marketed resulted from a reduction in aggregate production, caused by several factors, including drought, excessive pressure on the resource, and changes in weather patterns affecting the period of flowering and fructification of NWFP. This highlights the complexity of NWFP markets and the challenges facing their study, and the need for long-term monitoring in order to better characterise them.

In general, markets in major urban centres, Yaounde, Douala and border markets are more profitable than the other markets.

Although the exploitation of NWFP always have an ecological impact, the gathering and the marketing of Irvingia spp. and Cola acuminata are more sustainable because the tree is preserved after gathering the fruit or the kernels. The collection of Garcinia lucida and Garcinia kola barks usually kills or weakens the tree that is exploited.

With the devaluation of the CFA Franc and the continuous search for alternative income-generating opportunities by forest dwellers, the barks of Garcinia lucida and Garcinia kola will continue to be demanded by traders and consumers of palm wine, thereby generating revenues for forest dwellers and traders. In particular, Garcinia lucida, which is governed by an open access regime is more vulnerable to individual opportunistic behaviour with a danger of increased pressure on the resource. This shows the difficult trade-offs between improving the livelihoods of forest-dependent people and forest conservation. There is a high level of variability among different inventories of Garcinia lucida and Garcinia kola that could be due to their relative abundance in natural conditions. Research has much to contribute in finding an appropriate balance for achieving livelihood improvement through NWFP and forest conservation goals.

Research could make a significant contribution by:

· providing information on the sustainable rate of harvest that would preserve Garcinia lucida and Garcinia kola trees;

· designing sustainable harvesting techniques that could guarantee the survival of Garcinia lucida and Garcinia kola trees, thereby reducing future pressure on these resources;

· domesticating Garcinia lucida and Garcinia kola and promoting their cultivation by forest dwellers;

· enabling forest dwellers to access to improved technologies; and

· evaluating the impact of these technologies on the livelihoods of forest dwellers and forest conservation.

Single-disciplinary research alone is unlikely to solve the problem. We believe that only multidisciplinary studies, or at least improved communication between biological and social scientists, can help solve the livelihood and forest conservation dilemma. The welfare of hundreds of millions of people is at stake, as well as the maintenance of environmental services from the forests

References

BDCPC (Bioresources Development and Conservation Programme Cameroon) (1997) Newsletter, October/November.

Bela, Laurent Manga (1997). "Migrants de Retour et Développement Rural: Le Cas de Yémesoa." Les Cahiers d'Oscisca, No. 10, Septembre. ORSTOM, MINREST.

Chambers, R. & Leach, M. (1987). Trees to meet contingencies: saving and security for the rural poor. Discussion Paper 228, Institute of Development Studies, University of Sussex, Brighton.

Champaud, Jacques (1973) "MOM, Terroir Bassa (Cameroun)." Atlas des Structures Agraires au Sud du Sahara, no. 9, ORSTOM, Paris.

Cunningham, A. B. & F.T. Mbenkum (1993). "Sustainability of Harvesting Prunus africana Bark in Cameroon. A Medicinal Plant in International Trade." People and Plants Working Paper, May

de Beer, J. & McDermott, M. 1989. The economic value of non-timber forest products in Southeast Asia. Netherlands Committee for IUCN, Amsterdam.

Diaw, Mariteuw Chimère (1997). "Si, Nda Bot and Ayong: Shifting Cultivation, Land Use and Property Rights in Southern Cameroon." Rural Development Forestry Network, Network Paper 21e, ODI, London.

Doucet, J.L. & Koufani A. (1997). "Etudes des Produits Secondaires Végétaux de la Forêt de Kompia (Cameroun). Utilisation, Inventaires, Régénération, Commercialisation et Gestion Durable. Unité de Sylviculture, Faculté Universitaire des Sciences Agronomiques de Gembloux (Belgique)/ Herbier National du Cameroun.

Falconer, J. (1990) The major significance of minor forest products. FAO Community Forestry Note, 6. FAO, Rome.

Freese, C. 1996. The commercial, consumptive use of wild species: managing it for the benefit of biodiversity. WWF Discussion Paper. WWF, Washington.

Guedje, Nicole Marie (1996). "Evaluation Ecologique de Quelques Produits Forestiers Non-Ligneux de la Region de Bipindi-Akom II: Abondance, Distribution et Impact des Récoltes sur les Peuplements. Programme Tropenbos Cameroun.

Hall, Pamela & Kamaljit Bawa (1993). "Methods to Assess the Impact of Extraction of Non-Timber Tropical Forest Products on Plant Populations. Economic Botany, Vol. 47 (3): 234-247.

Harris, D. J. (1993). "A Taxonomic Revision and an Ethnobotanical Survey of the Irvingiaceae in Africa." PhD dissertation, University of Oxford, Oxford.

Laird, Sarah (1995). The Natural Management of Tropical Forests for Timber and Non-Timber Products. OFI Occasional Paper No. 49, Oxford Forestry Institute, Department of Plant Sciences, University of Oxford, Oxford.

Laird, Sarah & 17 others (1996). Medicinal Plants of the Limbe Botanic Garden. Mount Cameroon Project. Edited by Sylvia Howe.

Leakey, R. R. B. & J. A. Maghembe (undated). "Domestication of High Value Trees for Agroforestry: An Alternative to Slash and Burn Agriculture." ICRAF Position Paper 1.

Ndoye, Ousseynou (1994). "Research on Non-Timber Forest Products Utilization and Marketing at the International Institute of Tropical Agriculture." Paper Presented at the Workshop on Extractivism and Multiple Use Forest Reserves in Africa, Naro Moro (Kenya), 8-13 May.

Ndoye, Ousseynou (1995). "Commercialization and Diversification Opportunities for Farmers in the Humid Forest Zone of Cameroon: The Case of Non-Timber Forest Products." Consultancy Report for the Alternatives to Slash and Burn (ASB) Project, March.

Ndoye, Ousseynou, Manuel Ruiz-Perez & Antoine Eyebe (1997). "The Markets of Non-Timber Forest Products in the Humid Forest Zone of Cameroon." Rural Development Forestry Network, Network Paper 22c, ODI, London.

Nkongmeneck, B. (1985). "Le Genre Cola au Cameroun." Revues Sciences et Techniques (Série de la Science Agronomique, Yaoundé, Cameroun), 1 (3):57-70.

Ntamag, Cécile Ndjebet (1997). "Spatial Distribution of Non-Timber Forest Product Collection. A Case Study of South Cameroon." Unpublished MSC Tropical Forestry Thesis, Wageningen Agricultural University, Department of Forestry.

Oyono, Phil René (1997). "Ruptures Socio-Economiques et Surexploitation du Palmier Raphia par les Populations Forestières de Lomié (Sud-Est Cameroun). Bulletin Arbres, Forêts et Communautés Rurales n. 11: 27-33, Septembre.

Peters, Charles M. (1994). Sustainable Harvest of Non-Timber Plant Resources in Tropical Moist Forest: An Ecological Primer. Biodiversity Support Program, Washington, DC.

Pokam Wadja Kemajou Jacques (1997). "Les Migrations dans la Zone de Forêt Humide du Cameroon." Document de Travail No. 2. CIFOR Cameroun, Septembre.

Sayer, Jeffrey A. (1993). "Using Non-Timber Products to Support Forest Conservation Programmes." Status and Potential of Non-Timber Products in the Sustainable Development of Tropical Forests. ITTO Technical Series No. 11.

Scoones, Ian, Mary Melnyk & Jules N. Pretty (1992). "The Hidden Harvest: Wild Foods and Agricultural Systems. A Literature Review and Annotated Bibliography." International Institute for Environment and Development, London.

Shaffer, James D., with Michael Weber, Harold Riley & John Staatz. (1987). "Influencing the Design of Marketing Systems to Promote Development in Third World Countries." MSU International Development Papers, Reprint no. 2. Department of Agricultural Economics, Michigan State University.

Townson, I.M. (1995) Forest products and household incomes. A review and annotated bibliography. OFI, Oxford, CIFOR, Bogor.

Van Dijk, Han (1995) "Assessment of the Abundance and Distribution of Non-Timber Forest Product Species. Intermediate report of the project `The Economic and Ecological Assessment of Non -timber Forest Products in the Bipindi-Akom II region, South Cameroon'. The Tropenbos Cameroon Programme.

Van Dijk J.F.W. (1998). "Assessment of Non -timber Forest Products Resources in View of the Development of Sustainable Commercial Extraction." Paper Presented at the International Expert Workshop on Non-Wood Forest Products (NWFP) for Central Africa, Limbe Botanic Garden, 10-15 May.

Vivien, J. & J.J. Faure (1985). Arbres des Forêts Denses d'Afrique Centrale. Espèces du Cameroun. République Française, Ministère des Relations Extérieures Coopération et Développement. Agence de Coopération Culturelle et Technique.

Vivien, J. & J.J. Faure (1996). Fruitiers Sauvages d'Afrique. Espèces du Cameroun. Ministère Français de la Coopération. Centre Technique de Coopération Agricole et Rurale (CTA).

Williamson, Oliver E. (1985). The Economic Institutions of Capitalism. Free Press, New York.

The World Bank (1995). "Cameroon: Diversity, Growth, and Poverty Reduction." Africa Regional Office.

7 CFA=Communauté Financière Africaine. Cameroon, Gabon, Equatorial Guinea, Central African Republic, Congo (Brazzaville), Chad are the countries in Central Africa that are part of the CFA zone.

8 According to Diaw (1997), "Garcinia lucida is vital to social interaction in societies where the sharing of drinks and the fear of poisoning are both widespread".