The tropical rain forests of West and Central Africa are endowed with abundant high value indigenous fruit and medicinal tree species. Many rural households of the region heavily depend on these resources for their fruit, medicinal, food, and construction needs. Moreover, some of these products, particularly medicinal plants, are traded internationally, significantly contributing to the economy of the countries of the region. Currently, the majority of these products are exploited exclusively from the wild. The accelerated rate of destruction of the forest of the Central Africa region is currently estimated at 60% per year (FAO, 1997). The clearance of forest for agriculture, increased urbanisation and other economic developments makes the need to domesticate many high value forest resources an imperative. It is hoped that this will enable small-scale farmers to cultivate these species in an environmentally-sustainable and economically-profitable system. In order to enhance the potential adoption of cultivation, the importance of identifying greater market opportunities through transformation to value added goods and developing a sound policy environment is also advocated. This paper presents ICRAF's tree domestication approach and preliminary results obtained so far in West and Central Africa.
Key words: Domestication, vegetative propagation, seed, marcotting
Throughout the humid tropics there are numerous perennial woody species that have provided indigenous people with many of their needs for millennia (Leakey, 1998; Okafor and Lamb, 1994; Abbiw, 1990). With the rapid population growth, these resources are being depleted owing to the increasing demand for productive land for agriculture, which is met by clearing more forest. Deforestation reduces species diversity and erodes the genetic base of tropical trees, including those vital for the very survival of the population of the region.
The rain forest environment, as rich and diverse as it is, is also known to be extremely fragile. As a result, the region's forest land cleared for crop cultivation (based on the traditional shifting cultivation) remains productive for less than three years. It requires a fallow or resting period of over 20 years before land can again be brought under cultivation. The practice of shifting cultivation destroys the vegetation and exposes the soil to harsh climatic factors (intense solar radiation and heavy raindrops). This, in turn, leads to disruption of the closed nutrient recycle and severe degradation of the natural resource base with its attendant consequences on the global environment (Ahn, 1974; Lal, et al., 1975; Jha et al., 1979). The practice of shifting cultivation is reported to account for 60% of forest losses each year (FAO, 1997).
In response to both environmental concerns and the need to ensure the sustenance of the livelihood of the population of the region, agroforestry is advocated as a potential solution (Leakey, 1998). The aim is to develop a more sustainable form of land use that will improve farm productivity while at same time improving the welfare of the community. Traditional agroforestry systems are part and parcel of the cropping systems of the humid tropics of Africa. Simple indigenous multi-strata systems such as the cacao (Theobroma cacao) cultivation system and compound gardens composed of several medicinal and fruit tree species already exist in the region (Okafor and Fernandes, 1987).
Elsewhere in the world, similar tree-based home gardens have been found to be capable of providing 44% of people's carbohydrate needs and 32% of their protein intake, while using only 7% of their time (Cooper et al., 1996). In addition to their economic and nutritional importance, these systems are also biologically diverse and environmentally resilient.
In West and Central Africa, the production objective of some of these simple agroforests, such as the cacao or coffee systems, are often targeted to a single commodity, such as cocoa or coffee. This exposes farmers to high risk at times of price fluctuation, policy changes or natural disasters such as prolonged and debilitating disease outbreaks. For example, when the world cocoa price fell drastically and remained depressed from the late 1980s through the early 1990s, both producers and nations suffered immense economic setbacks. Many farmers responded by either abandoning the farm or clearing the cacao field completely and replacing it with other crops (Duguma et al., 1998).
As indicated above, farmers grow different types of indigenous fruit and medicinal plants in almost all of the simple agroforests. However, none of these species have been systematically selected and bred for quality or quantity. The inter-cropping system informally developed by farmers has never benefited from modern science in order to optimise resource use efficiency for environmental quality as well as increased return on investment. Although different types of species are known to be inter-cropped, the simple agroforests of West and Central Africa are reported to be less diverse and under utilised compared to the highly complex agroforestry systems of Southeast Asia (ICRAF, 1987, Duguma et al., 1990; Duguma, 1994). This suggests there is considerable potential for the development of these systems in the West and Central African region.
In recent years however, there has been a growing interest in optimising the value and role of high value indigenous species in the tree-based cropping system of West and Central Africa. The International Centre for Research in Agroforestry (ICRAF) began such an initiative in 1994. The aim is to first identify priority (from the farmers' perspective) high value species, with a view to improving their genetic base and, secondly, to integrate these species through domestication into improved multi-strata systems developed through the application of agroforestry science and indigenous knowledge. The strategy of the initiative and preliminary results obtained are provided below.
2. Species prioritisation
With only rare exceptions, the tree germplasm used or available for immediate use in agroforestry is wild and unimproved. The number of species in existing agroforestry technologies is enormous: 100-2 500 (Simons, 1996). Of course, a domestication programme cannot effectively handle so many species, nor would all species even warrant domestication at any one time.
For this reason, ICRAF has developed a decision-making framework to determine whether domestication of a particular species should proceed, and if so at what level of intensity and in what direction (Jaenicke et al., 1995; Franzel et al., 1996). This priority setting is important, as it identifies the 'top' species as farmers view them and which can best contribute to achieving research objectives, while focusing on key factors that minimise costs while increasing benefits. In the priority-setting exercise, farmers indicate which trees are most important for them and in what ways they would like them improved.
These guidelines for setting priority species constitute a great advance in procedural methodology. The process involves seven stages:
· Building a multi-disciplinary team (economists, foresters, agronomists, social scientists) and planning;
· Assessing client (farmers and users) needs;
· Assessing species currently used by clients;
· Ranking importance of tree products according to several criteria (food and nutritional security, market value and potential value);
· Identifying priority species;
· Evaluating and ranking priority species;
· The production of a final list of priority species for domestication.
The researchers then target their collections of germplasm to individual trees that farmers deem superior. This priority-setting methodology has now been used in a number of eco-regions: the semi-arid lowlands of West Africa, the humid lowlands of West Africa, the lowlands of the Peruvian Amazon, the Yucatan Peninsula in Mexico and the Miombo woodlands ecozone of Southern Africa (ICRAF, 1994).
The priority-setting exercise in the humid lowlands of West Africa (HULWA) identified the following species for domestication work (listed in order of importance):
· Irvingia gabonensis,
· Dacryodes edulis,
· Chrysophyllum albidum,
· Ricinodendron heudelotii,
· Garcinia kola.
In addition, on the basis of their value in existing and future international markets and as a potential source of income in the region, and because both trees are currently being over-exploited and harvested unsustainably in natural stands two medicinal trees, Prunus africana and Pausinystalia johimbe, were also added to ICRAF's domestication programme (Cunningham and Mbenkum, 1993; Dawson, 1997; Ndibi and Kay, 1997; Leakey, 1997; Sunderland et al., 1997).
Establishment of a genebank of priority species is a crucial first step in the process of tree domestication after the species prioritisation. First of all, it serves as a "bank" for continuous supply of diverse genetic material of the priority species for further selection. Of course this is possible only if the initial germplasm collection is undertaken in as wide and diverse a geographical spread as possible throughout the native range of the species. This enables researchers to obtain preliminary field information on growth performance, tree architecture, susceptibility or resistance to pests and diseases etc., which are crucial in the selection process. ICRAF's experience with Irvingia gabonensis, Irvingia wombolu, Prunus africana and Pausinystalia johimbe is reported below.
Farmer preference surveys in HULWA identified Irvingia gabonensis (bush mango) as a top priority species for the eco-region. The bush mango is marketed widely throughout the region (Ndoye, 1995), people eat the nutritious fruit and from the cotyledons, or kernels, they derive a thickening agent for stews. There is tremendous potential for improvement as there is great variation in the quality, taste and size of the fruit, the size of the kernel, the timing of fruit production and in the maturation process. Farmers cultivate the tree widely and tend to collect wildlings from the forest and plant these in their home gardens.
To launch the domestication programme, seeds were collected from tree farmers viewed as their 'best' or superior individuals in Gabon, Cameroon and Nigeria. This seed was used to establish three live genebanks: in Mbalmayo (Cameroon) and Ibadan and Onne in Nigeria. Preliminary results from these multi-site genebank trials revealed great variation between and among accessions. For the Nigerian genebanks, the difference in height development was significant.
There was also a statistically significant difference in the collar diameter between provenance in the Nigerian genebanks. The number of branches was another parameter recorded. In genebanks in Mbalmayo, there was also significant variation among and between accessions for height, stem diameter and number of branches. Twelve months after planting (MAP), one accession from Gabon (G1) and two from Nigeria performed well both in terms of diameter and height.
In general, genebanks in Nigeria were in better shape than the one in Cameroon. Variation in height, stem diameter and phenological development between and within populations of I. gabonensis was highly significant at this stage of development (12 and 24 MAP for Cameroon and Nigeria respectively). Should the same trends be observed when the plants are mature enough to display traits that farmers view as desirable (size of the fruit, kernels, sell cracking property for I. wombulu) then the research focus will be on identifying appropriate propagation techniques, cloning, mass propagation and on-farm evaluation of superior lines.
Marcotting is a vegetative propagation technique undertaken on intact branches. It involves bark girdling and wrapping the girdled area in a damp medium to induce rooting. It is an approach that has been widely used in horticulture to multiply traditional fruit trees in temperate regions. Its application to indigenous tropical species is very recent and preliminary results are extremely encouraging indeed.
For marcotting or air layering of I. gabonensis, national teams went out to interview farmers and to set marcotts on trees farmers viewed as superior. In all, 1 200 farmers were involved in setting 2 000 marcotts in Ibadan, Onne (Nigeria), Ebolowa and Mamfe (Cameroon).
While the technique produced propagules, the success rate was low (30%) and the survival rate was even lower (10%). The rooting rate can probably be improved by the use of rooting hormones, and better management of the fragile rooted propagules should increase the survival rate. It is also assumed that the weaning of rooted propagules under non-mist propagators (described by Leakey et al., 1990), as for rooted leafy cuttings, could considerably improve the survival rate.
Despite the low success rate, marcotting is still preferable to rooting of cuttings, as it is notoriously difficult to get cuttings from mature trees to root. The only alternative is to fell the tree and to root cuttings from juvenile coppice shoots that develop subsequently on the stump. While this is feasible, it would result in a long delay before the propagules would again become sexually mature. Once selected phenotypes have been captured, techniques must be developed to root cuttings from the rooted marcotts, so they can be rapidly multiplied. Of particular interest in the selection for desirable traits, another species, I. wombulu, has individuals with nuts that are self-cracking, exposing the kernel and making it much easier to extract. Such a trait could have enormous economic and social spin-offs by saving considerably on the labour needed to crack open the hard seed coat with a machete to extract kernels from the nuts of unimproved trees.
Prunus africana (formerly Pygeum africanum) is an indigenous timber tree of tropical montane areas of West, Central, East Africa and Madagascar. The bark extracts are used in the treatment of benign prostatic hyperplasia and prostrate gland hypertrophy, ailments suffered by 60% of men in Europe and USA. The complexity of the extract apparently precludes the likely synthesis of this drug (Waterman, 1994). The compound is extracted almost exclusively from the raw bark of this tree, the trade of which is worth some $220 million a year (Cunningham et al., 1997). However, unsustainable methods of exploiting the bark have seriously eroded natural populations of this montane species in both Madagascar and Cameroon (Cunningham and Mbenkum, 1993). The international trade of Prunus africana is now controlled, in theory, by the placing of the species on Appendix II of the Convention for International Trade in Endangered Species (CITES).
ICRAF, in collaboration with UNESCO (United Nations Educational, Scientific and Cultural Organization), Limbe Botanic Garden, CDC (Cameroon Development Corporation) and IRAD (Institut de Recherches Agricole pour le Developpement), has developed a programme to investigate the possibility of domesticating the species. This programme is concerned with the selection of populations that produce the greatest quantity of bark, or alternatively, the best quality and quantity bark extracts for medicinal use. This work was initiated with a survey of the genetic variation of P. africana in the wild, followed by seed collection and the establishment of live genebanks.
Seeds for the genebanks were collected in three areas in Cameroon (Mendankwe, Kilum Mountain and Mount Cameroon). Seed collected from 80 trees were set in a genebank at Tole near Limbe, while leaf samples from 31 trees were also harvested for molecular genetic analysis. The genetic analysis is still underway and results are pending.
The preliminary results from genebanks in Limbe analysed by the Mount Cameroon Project showed that the survival rate of all provenance's varied from 60% to 100%. There was statistically significant variation in early plant growth among the various accessions. There was a variation in mean height of 5-month-old plants of less than 40 cm, to over 100cm. The variation observed in survival rate and early height development does not seem to relate directly to the provenance of the seeds. Regardless of the source of the seed, however, the existence of such variation is a good indication that there is great potential for genetic improvement of this species through careful selection.
Vegetative propagation techniques are the quickest and most efficient means to select, capture and multiply traits that farmers have stated are desirable in these priority species in anticipation of their incorporation into different agroforestry systems.
Vegetative propagation offers the tree geneticist the ability to replicate traits of particular interest and also to overcome the problem of lack of seeds. The latter is important because most of the rain forest species are irregular in their fruiting pattern. Vegetative propagation is also a relatively straightforward technique to multiply, test, select and utilise the full range of genetic diversity present in most tree species that can easily be practised by farmers. In this way, selected and highly productive but unrelated clones can be used commercially, for reforestation and different agroforestry technologies. ICRAF's programme uses vegetative propagation techniques such as rooting, grafting and air layering to capture desirable traits for high-value species.
Rooting of juvenile cuttings has been shown to be an effective tool in vegetative propagation (Tchoundjeu and Leakey, 1996; Tchigio and Duguma, 1998). Advances in the development of low cost non-mist propagators are a key to the success of this technique (Tchoundjeu, 1997; Leaky et al., 1990).
The objective of the vegetative propagation research programme of ICRAF in HULWA is to determine the main factors that affect the rooting of juvenile leafy cuttings of P. africana. For this reason, we tested a variety of factors, including type of substrate and leaf area effects.
Substrates tested included sawdust, sand and a mixture of the two. Rooting of cuttings was significantly better in sawdust and the sand-sawdust mixture than it was in sand alone. A similar trend was noted regarding the highly significant difference between percentages of dead cuttings in each treatment. The greatest number of dead cuttings was observed in the sand-sawdust mixture, followed by sawdust and sand substrates, respectively. Despite the low rate of rooting success in sand, paradoxically this medium also produced a low rate of mortality of cuttings. Sand seems to be an appropriate rooting substrate for conserving fragile cuttings. This is valid in the humid environment of a tropical forest. Similar results would be unlikely in the Sahel where atmospheric humidity is lower.
Leaf area was also an important factor and accounted for significantly different rooting rates of P. africana. As observed in many similar experiments with tropical trees, the first cuttings rooted only two weeks after the trial was set up. The presence of a leaf is crucial to the rooting process; no leafless cuttings rooted in this experiment. Furthermore, rooting success appeared to improve with increased leaf size. Cuttings with leaves of 20 and 25 cm² rooted significantly better than those with areas of only 5 and 10 cm². Although previous experience with other tropical species such as Lovoa trichilioides has shown that larger leaf areas can in fact impede rooting (Tchoundjeu, 1998), this was not the case in this trial with P. africana where 80% of the cuttings with leaf areas of 20-25 cm² rooted. Follow-up experiments to test a range of larger leaf areas should help clarify this issue.
When dead cuttings were examined, it appeared that the mortality rate was inversely proportional to the leaf areas. No leafless cutting survived at the end of the experiment, and leaves and leaf area are both requisites to successful rooting of P. africana.
Contrary to the inverse relation between the mortality rate and the size of the leaf area, the mean number of roots per rooted cutting was proportional to the leaf areas. Cuttings with leaf areas of 20 and 25 cm² produced the highest mean number of roots, although the difference was not significant.
These two experiments with P. africana clearly indicated that this species can be multiplied by rooting. However, more studies on the effects of different types and concentrations of hormones are needed to shed more light on the process of rooting of this important species. This could help pave the way for domestication of P. africana, a species that produces recalcitrant seeds which lose their ability to germinate only a few days after collection. This has proved to be a particular constraint to cultivating the tree in Madagascar where farmers have a long tradition of tree propagation and planting.
The work to develop vegetative propagation techniques for this species will enable researchers and their farmer partners to produce a stable domesticated population of the species. This will serve to guarantee income to the rural population using this improved germplasm, while decreasing pressure on a valuable natural resource that is currently harvested unsustainably throughout its natural range.
P. johimbe is a late secondary forest tree species, the bark of which is commercially exploited. It forms the main ingredient in many pharmaceutical and herbal remedies to alleviate the symptoms of organic impotence (Sunderland et al., 1997). On a local level, the bark is used as an aphrodisiac and as a mild stimulant to prevent drowsiness. This species is thus important for the supply of raw material for well-developed local and international markets.
Because of its many uses, its value on the international market and the destructive methods used to harvest the bark, P. johimbe is now seriously threatened, especially in South Cameroon. Entire trees are felled and then cut into logs, from which bark is stripped. To date, bark has been taken entirely from trees in natural stands and there has been no effort made to develop strategies of sustainable exploitation of the wild resource nor, until recently, had any attempts been made to domesticate the species (see Sunderland et al., this volume).
ICRAF together with Boehringer Ingelheim have initiated a programme to assess the domestication potential of this species, using vegetative propagation techniques. The ICRAF collaborative team have tested the effectiveness of different types of substrates for rooting leafy stem cuttings under non-mist propagators in the nursery at Mbalmayo, Cameroon. Material used for rooting in this experiment was harvested from the stumps where exploiters have carried out destructive felling. These trials clearly indicate that the rooting success of P. johimbe differs significantly depending on the substrate used. Cuttings set in sawdust alone and in a sand/sawdust mixture rooted significantly better than did cuttings in a medium of pure sand. Although the percentage of cuttings that rooted in the sawdust in the first six weeks was quite low, the rooting rate increased rapidly from week seven, to attain a maximum rooting rate of 74% by week nine, when the experiment ended.
Observations of the mortality rates of the cuttings in the various media showed that no cuttings died in the sand substrate. The number of dead cuttings in the sand/sawdust medium was significantly higher than in sawdust alone. Sand is a porous medium, where cuttings could easily absorb humidity from the water table of non-mist propagator by capillarity. This could explain the fact that non-rooted cuttings were still alive after six weeks in the propagator. Moreover it is also a good indication of the resilience of P. johimbe cuttings. Therefore the collection of cuttings under appropriate conditions could be extended over many days. This points to the possibility of collecting germplasm from this threatened species from neighbouring countries, such as Equatorial Guinea, where extensive natural stands still exist.
The mean number of roots per rooted cutting also varied according to the rooting medium. Sand produced cuttings with the highest mean number of roots, compared with fewer roots in the sand/sawdust mixture and in sawdust alone. The evidence from this trial suggests that rooting P. johimbe cuttings using non-mist propagators is entirely feasible.
Further research is being carried out on the germination of seeds of this important species. The seeds are extremely small and so far attempts to germinate this species has proven difficult, with extremely high mortality of the young seedlings.
The domestication research programme in the humid lowlands of West Africa is still in its very early stages. So far, work has begun on only a few priority species (Irvingia gabonensis, Prunus africana, Pausinystalia johimbe, Dacryodes edulis, Ricinodendron heudelotii, Garcinia cola and Cola nitida). As the programme advances and improved germplasm for species that farmers view as potentially lucrative or vital to their household welfare is made available, these agroforestry trees will be incorporated into the planting culture in farming systems throughout the region. Once incorporated and developed to complex agroforests, they have the potential to become major income producers in themselves because of the range of highly marketable products produced in such systems. Such agroforestry strategies also reduce risks; should any one product lose its market value, the diversity of trees that farmers will be cultivating will reduce the impact of this potential loss of revenue. Not only are such systems economically viable, they are also biologically diverse, creating a veritable marriage between environmentally sustainable production and profitability; the ultimate aim of all ICRAF's research programs.
Abbiw, D. 1990. Useful plants of Ghana: West African uses of wild and cultivated plants. Intermediate Technology Publications, Royal Botanic Gardens, Kew.
Ahn, P.M. 1974. Some observations on basic and applied research in shifting cultivation. FAO Soils Bulletin (24) 123-154.
Cooper, P.J.M., R.R.B. Leakey, M.R. Rao & L. Reynolds. 1996. Agroforestry and the mitigation of land degradation in the humid and sub-humid tropics of Africa, Experimental Agriculture. 32: 235-290.
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 No. 2, UNESCO, Paris.
Dawson, I. 1997. Prunus africana: How agroforestry can help save an endangered medicinal tree. Agroforestry Today. Vol. 9 (2): 15-17.
Duguma, B. 1994. Environmental characteristics and land use systems of the humid lowlands of west and central Africa. In Adeola, I.O.O and D. Ladipo (eds). Proceedings of the regional workshop on multipurpose tree prioritisation. 24-26 June 1993, IITA, Ibadan, Nigeria.
Duguma B., J. Gockowski, & J. Bakala. 1998. Smallholder cocoa (Theobroma cacao) cultivation in agroforestry systems of West and Central Africa: Challenges and opportunities. A paper presented at international workshop on sustainable cocoa, organised by Smithsonian Tropical Research Institute, Panama City, Panama, 29 March to 3 April 1998.
Duguma B., Tonye J. & D. Depommier. 1990. Diagnostic survey on local multipurpose tree/shrubs fallow systems, and livestock in southern Cameroon. ICRAF working paper No 60. ICRAF, Nairobi, Kenya.
FAO, 1997. State of the World's forest. Worlds End Publication, Oxford, UK.
Franzel, S., H. Jaenicke & W. Janssen. 1996. Choosing the right trees: Setting priorities for multipurpose tree improvement. ISNAR Research Report No. 8. The Hague: International Service for National Agricultural Research.
ICRAF, 1987. Annual Report 1987. International Centre for Research in Agroforestry, Nairobi, Kenya.
ICRAF, 1994. Annual Report 1994. International Centre for Research in Agroforestry, Nairobi, Kenya.
Jaenicke, H., S. Franzel & D.J. Boland. 1995. Towards a method to set priorities amongst species for tree improvement research - a case study from West Africa. Journal of Tropical Forestry Science 7: 490-506.
Jha, M.N., P. Pande & T.C. Pathak. 1979. Studies on the changes in the physico-chemical properties of tripura soils as a result of jhuming. Indian Forester, 105: 436-441.
Lal, R., B.T. Kang, F.R. Moormau, A.S.R. Juo & J.C. Moomaw. 1975. Soil management problems and possible solutions in western Nigeria. In Bornemisza, E. and A. Alvarado (eds): Soil management in tropical America, pp. 372-408. Raleigh: North Carolina State University. (quoted in Nair, P.K.R., 1984).
Leakey, R.R.B., J.F. Mesen, Z. Tchoundjeu, K.A. Longman, J. McP. Dick, A. Newton, A. Matin, J. Grace, R.C. Munro & P.N. Muthoka. 1990. Low-technology techniques for the vegetative propagation of tropical trees. Commonwealth Forestry Review 69: 247-257.
Leakey, R.R.B. & A. Newtown. 1994. Domestication of "Cinderella" species as the start of a woody-plant revolution. In Leakey, R.R.B & A. Newton, (Eds.) Tropical trees: Potential for domestication, rebuilding genetic resources. HMSO, London.
Leakey, R.R.B. 1997. Domestication potential of Prunus africana (Pygeum) in Sub-Sahara Africa. In Kinyua, A.M., W.M. Kofi-Tsekpo & L.B. Dangana (eds). Conservation and utilization of indigenous medicinal plants and wild relatives of food crops.
Leakey, R.R.B. 1998. Agroforestry in the humid lowlands of West Africa. Some reflections on future directions in research. In Duguma, B. (ed). Agroforestry Systems (Special Issue) 40(3).
Ndibi, B.P. & E.J. Kay. 1997. The regulatory framework for the exploitation of medicinal plants in Cameroon: the case of Prunus africana on Mount Cameroon. Biodiversity and Conservation 6, 1409-1412.
Ndoye, O. 1995. The markets for non-timber forest products in the humid forest zone of Cameroon and its borders structure, conduct, performance and policy implications. Report to CIFOR. Bogor, Indonesia. 86 p.
Okafor, J.C. & E.C.M. Fernandes. 1987. Compound farms of south eastern Nigeria: A predominant agroforestry homegarden system with crops and small livestock, Agroforestry Systems, 5: 153-168.
Okafor, J.C. & A. Lamb. 1994. Fruit trees: Diversity and conservation strategies. In Leakey, R.R.B & A. Newton, (Eds.) Tropical trees: potential for domestication, rebuilding genetic resources. HMSO, London.
Sunderland, T.C.H., A. Njiamnshi, A. Koufani, M.L. Ngo- Mpeck & C. Obama. 1997. The ethnobotany ecology and natural distribution of yohimbe (Pausinystalia johimbe K. Schum.) and evaluation of the sustainability of current bark harvesting practices and recommendations for domestication and management. A report prepared for ICRAF.
Tchoundjeu, Z. Vegetative propagation of African mahogany: effects of provenance, substrates, auxins and leaf areas. (in press).
Waterman, P.G. 1994. Trees for drugs and other extractives: future prospects. In Leakey, R.R.B & A. Newton. (Eds.) Tropical trees: potential for domestication, rebuilding genetic resources. HMSO, London
