Poverty of the rural resource-poor people in the HULWA region is the fundamental developmental problem addressed by this project.

Devaluation of the CFA franc, and effects of structural adjustment have, in some cases, even led to reverse migration, with urban poor returning to the forest areas. Poverty alleviation for these subsistence farmers must include the achievement of a regular and low-risk source of income. Many studies have highlighted the importance of trees (both on farm and in the forest) in providing resource-poor people with many essential subsistence products (medicines, fibres, etc.), food during "hungry" periods, and acting as forms of savings and security for emergencies. In areas such as the HULWA region, which are well endowed with widely used indigenous tree species, particularly fruit trees, there is a need to enable farmers to earn a more consistent and important income from them. The proposed research intends to take forward work on the role of tree management in farmer strategies, by focusing specifically on indigenous trees. In many cases these are already protected (as wildings), but farmers rarely take the next step of planting them for income generation.

The second developmental problem tackled by the project concerns the extension services (whether government or NGO), which are expected to advise farmers on how to improve production, on the basis of very little information. In the same conditions, extension services may be faced with many options, ranging from the promotion of better management of off-farm resources to the planting of known exotics, or of well-known but unimproved indigenous trees on farm.

Extension services rarely have sufficient understanding of the different options, particularly those involving the cultivation of indigenous tree species.

Foresters have traditionally sought technical fixes to problems, ignoring or unaware of the fact that they may also require interventions of a socio-economic kind.

Tree improvement work has been carried out on a very limited range of widely used species in the HULWA region. The project's key collaborators, IRAD and ICRAF, are among the main organisations involved in this tree improvement work. This has initially focused on developing successful vegetative propagation methods, but is now progressing to the stage of selecting individual trees for certain characteristics. These are primarily researcher-defined, although ICRAF has now started a process of involving farmers in selection. What is still missing is a differentiated analysis of how existing biological variability can be used to promote characteristics that respond to both farmer and market preferences.

Economic crisis and devaluation of the CFA franc have increased rural population pressure on NTFP resources. Certain widely used species are increasingly threatened, as habitat loss (partly due to expansion of agriculture and partly due to logging), destructive harvesting practices, and over-harvesting, reduce wild populations below recoverable levels. These species have not yet been satisfactorily domesticated, and the species as a whole may be vulnerable to the stresses placed on the wild populations. In Central Africa more generally, researchers feel that domestication is the only answer to the problem of providing people with a sufficient supply of desired non-timber forest products, while reducing over-exploitation.

Understanding of livelihood systems in forest margin communities improved and incorporated into land use strategies.

• Factors affecting farmers' interest and ability to grow and cultivate indigenous fruit trees or farm for income-generation analysed, in four communities in Cameroon and two in Nigeria.
• Market structure described and price trends analysed for key products.
• Potential for varietal improvement to enhance the role of priority tree species in farmer livelihoods investigated.
• Research results disseminated.