Commercial gathering of traditional medicines in large countries with small urban populations (e.g. Mozambique, Zaire and Zambia) is limited and cases of over-exploitation are rare. Harvesting by TMPs continues usually to be selective and on a small scale, and traditional conservation practices, where they exist, would be expected to be retained. In African countries with high rural population densities and small cities (e.g. Rwanda), gathering is also expected to be small scale, and where a species is popular and supplies are low due to habitat destruction and agricultural expansion, the tree will suffer a “death of a thousand cuts” rather than one-off ring-barking due to commercial harvesting (see Photo 3).

The emergence of commercial medicinal plant gatherers in response to urban demand for medicines and rural unemployment has resulted in indigenous medicinal plants being considered as an open access or common property resource instead of a resource only used by specialists. The resultant commercial, large-scale harvesting has been the most significant change, although seasonal and gender related restrictions have also altered. Rural traditional medical practitioners and the hereditary chiefs who traditionally regulate resource management practices admit that ring-barking and over-exploitation by commercial gatherers are bad practices that undermine the local resource base. In Natal (South Africa) it appears that restrictions placed by traditional community leaders and enforced by headmen and traditional community policemen have reduced commercial exploitation of local traditional medicinal plant resources. With cultural change, increased entry into the cash economy and rising unemployment however, these controls are breaking down.

Ring-barking or uprooting of plants is the commonest method of collection used by commercial gatherers (Photo 6). Where urban populations (and resultant commercial trade in traditional medicines) are relatively small, but high rural population densities and an agricultural economy have cleared most natural vegetation, tree species such as Erythrina abyssinica and Cassia abbreviata, which are popular and accessible, have small pieces of bark removed (Photos 3 and 4), rather than a one-off removal of trunk bark (Photos 5 and 6).

(3) Erythrina abyssinica (Fabaceae), Malawi (“death from a thousand cuts”) and		(4) Cassia abbreviata (Fabaceae), Zimbabwe,
(5) Large pieces of Warburgia salutaris (Canellaceae) bark from Namaacha on the Swaziland border commercially gathered for sale in Maputo, Mozambique,		(6) Curtisia dentata (Cornaceae) tree in Afro-montane forest, South Africa, debarked for sale in Durban, a city 100 km away.

In South Africa, where the taboo against gathering of traditional medicines by menstruating women was widespread in the past, urban herbalists now no longer place importance on this when buying plants from urban markets, or in some cases, treat the plants to magically “restore their power”. Strict seasonal restrictions are still placed on the gathering of Siphonochilus aethiopicus rhizomes in South Africa and Swaziland, but commercial collection of Alepidea amatymbica rhizomes now takes place “on misty days” in summer (although harvested material is stored away from the homestead for fear of lightning). Even where seasonal restrictions are still in place, demand can exceed supply. Siphonochilus natalensis for example, had disappeared from its only known locality in Natal before 1911 as a result of trade between Lesotho and Natal (South Africa) (Medley-Wood and Evans, 1898).

It is clear that medicinal plant species gathered for commercial purposes represent the most popular and often most effective (physiologically or psychosomatically) herbal remedies. From historical records (Gerstner, 1938, 1939; Medley-Wood, 1896) it is clear that the majority of species that were popular in the past are still popular today. Examples in southern Africa include Erythrophleum lasianthum, Cassine transvaalensis, Alepidia amatymbica and Warburgia salutaris. Commercially sold species thus represent a “short list” of the medicinal plants used nationally, since many species that are used to a limited extent in rural areas are not in demand in the urban areas. Also important from a resource management point of view, is that in virtually all African countries, it is not the limited, selective harvesting by specialist TMPs that represents the problem. In most cases, non-sustainable use of favoured species results from commercial harvesting to supply an urban demand for traditional medicines, after clearing for agricultural or urban associated development has already taken place. The widespread commercial harvesting and sale of the same genera and species throughout their distribution range is significant (e.g. Solanum fruits, Erythrophleum bark, Abrus precatorius seeds, Myrothamnus flabellifolius stems and leaves and Swartzia madagascariensis roots) (Appendix 1).

Medicinal plant gatherers are familiar with which species are becoming difficult to find, either because of limited geographical distribution, habitat destruction or over-exploitation. Their insights, coupled with botanical and ecological knowledge of the plant species involved, provide an essential source of information for a survey of this type. In this survey, it was not considered constructive to distinguish between plant species with symbolic uses and those with active ingredients. The important question here is whether the species are threatened or not, because:

(1) species that have a purely symbolic value are nevertheless important ingredients of traditional medicines for their psychosomatic value and are as effective as placebos are in urban-industrial society;

(2) the majority of traditional medicines have not been adequately screened for active ingredients and a number of species, for example Rapanea melanophloes in southern Africa, while being primarily used for symbolic purposes, also have active ingredients. Conservation efforts must therefore be directed at all species vulnerable to over-exploitation.

For any resource, a relationship exists between resource capital, resource population size and sustainable rate of harvest. Low stocks are likely to produce small sustainable yields, particularly if the target species is slow growing and slow reproducing. Large stocks of species with a high biomass production and short time to reproductive maturity could be expected to produce high sustainable yields, particularly if competitive interaction is reduced by “thinning”. The impact of gathering on the plant is also influenced by factors such as the part of the plant harvested and harvesting method.

Sustainable supplies of traditional medicines

Demand for fast growing species with a wide distribution, high natural population density and high percentage seed set can be met easily, particularly where leaves, seeds, flowers or fruits are used (Photo 7). The common sale and use of medicinal plant leaves as a source of medicine in Côte d’Ivoire and possibly other parts of west Africa (Photo 2) is therefore highly significant as it differs markedly from the high frequency of roots, bark or bulbs at markets in the southern African region (Photo 7). Throughout Lesotho, Malawi, Mozambique, Swaziland, Zambia, Zimbabwe, and particularly South Africa, herbal material that is dried (roots or bark), or has a long shelf-life (bulbs, seeds and fruits) dominates herbal medicine markets (see Appendix 1). In contrast, six sellers in Abidjan, Côte d’Ivoire, primarily sold leaf material (20-41 spp.), followed by roots (1-16 spp.), bark (0-8 spp) and whole plants (0-3 spp.). This situation was typical of the 111 traditional medicine sellers in Abidjan, apart from those bringing material from Burkina Faso and Mali, who sell more root and bark material. The situation with chewing stick sellers in Côte d’Ivoire and other parts of west Africa is somewhat different however, as stems and roots are the major plant parts used, with consequent higher impact on favoured species.

Photo 7. Medicinal plants for sale at a marketin Bulawayo, Zimbabwe, showing the dominance of bark and root material as a source of herbal medicines.

Despite limited information on the population biology of medicinal plants, it is possible to classify target plant species according to demand, plant life-form, part used, distribution and abundance (Cunningham, 1990). The large category of traditional medicinal plants which are under no threat at all are the cause of little concern to TMPs or to conservation biologists. For these species, demand easily meets supply. From a conservation viewpoint, on an Africa-wide scale, there are two categories of medicinal plants that are of concern:

(1) Slow growing species with a limited distribution which are the focus of commercial gathering where demand exceeds supply. Harvesting expands to areas progressively further afield, where rising prices for the target species are incentives to collect. This results in the species being endangered regionally and causes widespread depletion of the rural resource base of TMPs. Examples of this include Warburgia salutaris in east and southern Africa and Siphonochilus aethiopicus in Swaziland and South Africa. Endemic species with a very localized distribution are a particular problem, for example:

(a) Ledebouria hypoxidoides, which is endemic to the eastern Cape region (South Africa). Herbalists were observed removing the last bulbs from the locality near Grahamstown (F. Venter, pers. comm.).

(b) Mystacidium millari, also endemic to South Africa, which is threatened due to harvesting and commercial sale as a traditional medicine in the nearby city of Durban, South Africa (Cunningham, 1988a).

(2) Popular species which are not endangered because they have a wide distribution, but where habitat change through commercial harvesting is cause for concern. Trichilia emetica and Albizia adianthifolia for example, are not a high priority for conservation in southern Africa, although they are a popular source of traditional medicines. What is of concern however, is that ring-barking in “conserved” forests is causing canopy gaps and changing the forest structure, which can lead to an influx of invasive exotic species. This is important for local habitat conservation.

Both categories are of particular concern in protected area management, as core conservation areas will ultimately come under pressure from harvesting for favoured species if they are difficult to obtain elsewhere.

Information on the quantities of traditional medicines being harvested or sold is sparse, whether for the local trade in traditional medicines, or for export and extraction of active ingredients. Apart from placing the quantities required from cultivation into perspective, the information available is of little relevance unless expressed in terms of impact on the species concerned. In South Africa, harvesting from wild populations of certain species is on a scale that gives cause for concern amongst conservation organizations and rural herbalists, and a listing of priority species is available (Cunningham, 1988a) (Box 2). The same applies to some chewing stick species, such as Garcinia afzelii in west Africa. The only quantitative data on the volume of plant material sold comes from Natal (South Africa), where medicinal plants are ordered by urban based herb traders in standard-size maize bags (Table 4).

Sustainability of chewing stick harvesting

Chewing sticks are obtained from wild populations of indigenous plants, apart from the infrequent sale of exotic species such as Azadirachta indica and Citrus sinensis (Appendix 1). Garcinia afzelii is considered to be threatened by this trade (Ake Assi, 1988b; Gautier-Beguin, pers. comm.). In Nigeria, Okafor (1989) reports that Randia acuminata chewing sticks are still collected from primary and secondary forest within 3 km of villages, but that the distance is increasing, which indicates that the resource is being depleted. At a single depot, for example, Okafor (1989) recorded that five commercial chewing stick collectors assembled 1144 bundles of chewing sticks, made up of seven or eight split stems one metre long per bundle. What is highly significant from a resource management viewpoint, and has not been taken into account previously, is that whilst peeled twigs are used as chewing sticks from most species, split stems and roots are the source of the commercially sold chewing sticks. Among the 27 species used in Ghana, for example, high impact harvesting of stem wood or root material from only seven species accounted for 88% of chewing sticks used. The low impact use of peeled twigs as chewing sticks accounted for the other 12 % of sticks used and for the remaining 20 species (Ake Assi, 1988b). Impact on those source species which are cut down or up-rooted to supply urban demand is therefore high.

Supplying international trade

Few data are available on the quantities of raw material harvested for the pharmaceutical trade, or the environmental impact of harvesting. It is clear however that large quantities of material are collected from the wild and that harvesting can be very destructive. The same can apply to plant material collected for screening purposes. Juma (1989) offers the example of Maytenus buchananii: 27.2 tons of plant material were collected by the American National Cancer Institute (NCI) from a conservation area in the Shimba Hills (Kenya), for screening purposes as a potential treatment for pancreatic cancer. When additional material was required four years after the first harvesting in 1972, regeneration was so poor that collectors struggled to obtain the additional material needed.

No studies are known to have been carried out on the social or environmental consequences of harvesting, for example:

(1) the 75-80 t of Griffonia simplicifolia seed exported each year to Germany from Ghana (Abbiw, 1990);

(2) the medicinal plant material exported from Cameroon to France (Voacanga africana seed (575 tons); Prunus africana bark (220 tonnes), Pausinystalia johimbe bark (15 t) (United Republic of Cameroon, 1989).

However, Ake-Assi (pers. comm.) reports that commercial gatherers in Côte d’Ivoire chop down Griffonia simplicifolia vines and Voacanga africana and Voacanga thouarsii trees in order to obtain the fruits. Concern has been expressed about a similar situation in Indonesia, where Rifai and Kartawinata (1991) point out that:

“Export of medicinal plants has been going on for many years, and the demand in the international market keeps increasing. One big Swiss pharmaceutical company, for example, has requested eight tons of seeds of Voacanga grandifolia and are willing to pay a high price. This species is rare and has light seeds. To satisfy the above request, all available seeds in the forest will perhaps have to be harvested, leaving nothing for regeneration. Similarly, five tons of rhizomes of a rare Curcuma (tema badur) has been sought by a West German pharmaceutical company, and 100 kg year-1 of pili cibotii (fine hairs of Cibotium barometz) by a French firm. It can be imagined how many plants of these species will have to be destroyed should such requests be satisfied.”

If the international companies involved in this trade are to operate in a responsible manner, then this situation needs to change to one of commercial cultivation and sustainable use.

The real price of trade

The categories of medicinal plant species that are most vulnerable to over-exploitation can be identified by combining the insights of herbal medicine sellers with knowledge on plant biology and distribution (Cunningham, 1990). However, due to the number of species involved and the limited information on biomass, primary production and demography of indigenous medicinal plants, no detailed assessment of sustainable off-take from natural populations is possible. Even if these data were available, their value would be questionable due to the intensive management inputs required for managing sustainable use of vulnerable species in cases where demand exceeds supply.

Unsustainably high levels of exploitation are not a new problem, although the problem has escalated in regions with large urban areas and high levels of urbanization since the 1960s. Prior to 1898, local extermination of Mondia whitei had been recorded in the Durban area of South Africa due to collection of its roots “which found a ready sale in stores”. By 1900, Siphonochilus natalensis (an endemic species now considered synonymous with Siphonochilus aethiopicus; Gibbs-Russell et al., 1987) had disappeared from its only known localities in the Inanda and Umhloti valleys due to trade to Lesotho. This occurred despite a traditional seasonal restriction on harvesting this species. By 1938, all that could be found of Warburgia salutaris in Natal and Zululand was “poor coppices, every year cut right down to the bottom” (Gerstner, 1938). Most botanical and forestry records reflect the impact of commercial collection of Ocotea bullata bark due to the importance of this species for timber. Oatley (1979) for example, estimated that less than 1% of 450 trees examined in Afro-montane forest in South Africa were undamaged, and in the same region, Cooper (1979) estimated that 95% of all Ocotea bullata trees had been exploited for their bark, with 40% ring-barked and dying. The situation would appear to be similar in Kenya, where Kokwaro (1991) records that some of the largest Warburgia salutaris and Olea welwitschii trees have been completely ring-barked and have died. In Zimbabwe, due to the high demand and limited distribution of this species, the situation is worse, and all that remains of wild Warburgia salutaris populations are a few coppice shoots (S. Mavi pers. comm., 1990). In Côte d’Ivoire, Garcinia afzelii is considered threatened due to harvesting for the chewing stick trade (Ake Assi, 1988b). Destructive harvesting of Griffonia simplicifolia, Voacanga thuoarsii and Voacanga africana fruits for the international pharmaceutical market is also of concern (L. Ake-Assi, pers. comm., 1989). In Sapoba Forest Reserve, Nigeria, despite traditional restrictions on bark removal, Hardie (1963) observed how the trunk of a large Okoubaka aubrevillei tree (a very rare species in west Africa) “was much scarred where pieces of bark had been removed”. There appears to be nothing published on the current status of this species. Botanical records are also scanty for bulbous or herbaceous species, where little remains to indicate former occurrence after the plant has been removed. It would therefore be useful to carry out damage assessments for species such as:

(1) Okoubaka aubrevillei, Garcinia afzelii, G. epunctata, and G. kola in Côte d’Ivoire , Ghana, and Nigeria;

(2) Warburgia salutaris in Kenya, Tanzania and Zimbabwe;

(3) assessments of the impact of Prunus africana and Pausinystalia johimbe bark harvesting in Cameroon and Madagascar, and fruit harvesting of Griffonia simplicifolia, Voacanga thuoarsii and Voacanga africana for the international pharmaceutical market.

In South Africa, bark damage assessments using a 7-point scale (Figure 4) were carried out for key “indicator species” (medicinal plants chosen for their relatively slow growth rate - all were trees), popularity as a source of traditional medicines, their scarcity (all were forest species, and indigenous forest only covers 0.3% of South Africa), and where bark removal took place.

Figure 4. The seven-point scale used in field assessment of bark damage. All assessments represent the degree of bark removal below head height (2 m), which is marked by the dotted line and arrow in the figure (Cunningham, 1988a).

Bark damage assessments confirmed most of the observations of herbalists and herb traders (Tables 5 and 6), the exceptions being species that were scarce not because of over-exploitation, but due to limited geographical distribution in the region, such as Acacia xanthophloea and Synaptolepis kirkii. They also demonstrate the very different situation to customary subsistence use, and this fact needs to be taken into account in legislation covering protected area management where conservation of biotic diversity is a primary objective. Although the degree of bark damage varies, the level at all sites where commercial gathering is taking place is high and concentrates on large diameter size classes. What is significant is that extensive damage has taken place in State Forest, theoretically set aside for maintenance of habitat and species diversity (Figure 5). In the eMalowe State Forest, Transkei, South Africa, if coppice stems less than 2 cm diameter are excluded, then the level of damage to Curtisia dentata and Ocotea bullata trees encountered represents 51% and 57% of trees having more than half the trunk bark removed. All Warburgia salutaris trees found outside strict conservation areas in Natal were ring-barked, and many of those inside conserved areas had their bark removed as well.

Figure 5. Damage to selected tree species in “protected” forest reserves where commercial bark exploitation is taking place:Ocotea bullata (Lauraceae) and Curtisia dentata (Cornaceae) in eMalowe Forest Reserve, South Africa. (Cunningham, 1988a). Note the selection for bark from large trees. DBH = stem diameter at breast height.

Even fewer data are available on the impact of harvesting bulbs, roots or whole plants although local depletion of Stangeria eriopus, Gnidia kraussiana and Alepidea amatymbica is known from Natal, South Africa. According to C. Stirton (pers. comm.) there has also been a marked reduction in numbers of the Afro-montane forest climber Dumasia villosa, which is sold in large quantities in herbal medicine shops (Cunningham, 1988a). C. Hines (pers. comm.) has similarly recorded exploitation of Protea gauguedi populations to the point that, with the possible exception of the eastern Caprivi, the species could be considered extinct in northern Namibia despite attempts by the conservation department to protect it. What is noteworthy in this case is that it has taken place in response to a local trade in an area where urban centres are small. Commercial harvesting of Harpagophytum procumbens tubers in Botswana removed up to 66 % of plants (Leloup, 1984). In Nambia, however, this species was not considered threatened as the 200 t exported each year only represented 2% of total stocks (de Bruine et al., 1977).

Increasing scarcity of popular species is followed by a increased price, which in turn results in greater incentives to harvest remaining stocks. The effects of this are firstly, decreased self-sufficiency of traditional medical practitioners as local sources of favoured species decline, and secondly, higher prices which people have to pay for those species. As demand is a one of the root causes of over-exploitation, the most popular and effective species are the most vulnerable.

The reasons for concern

In spite of increasing urbanization, a large proportion of the African population has retained their reliance upon this traditional approach to healthcare and continue to consult TMPs for medical treatment. Even where western medicine is available, it is unlikely that it will be adopted without first establishing a framework for national economic growth which would allow for socio-economic and cultural changes to take place, and give access to formal education. Good (1987) writes:

“Although most countries in Africa routinely allocate substantial portions of their budgets to health services and related infrastructure such as water supplies, sanitary works and roads, sustained improvements in community health status and increased accessibility to government and private health services have not materialized. Instead, health ministries find themselves preoccupied just with preventing the deterioration of existing “aspirin and bandage” services.”

In reality, most African countries are experiencing an unprecedented economic deterioration with per capita income having fallen by an average of 0.4% since 1986, and Africa’s debt being roughly three times greater than its export revenue. The heavy reliance upon traditional medicine therefore is unlikely to change. At the same time, there is significant evidence to show that the supply of plants for traditional medicine is failing to satisfy demand. This problem has been exacerbated by three main factors:

(1) A high rate of population growth and urban expansion, generating an informal and growing species-specific trade network which extends across international boundaries.

(2) The change from medicinal plant harvesting being a purely specialist activity of TMPs, to one involving an informal sector group of commercial plant harvesters whose prime motivation is profit. This is a response to increased population and greater demand and is resulting in a disregard for traditional conservation practice and a breakdown of taboos and customs in the opportunistic scramble for divining supplies. High unemployment means that labour is plentiful and cheap, keeping prices low and sales high. In the case of medicinal plants which are harvested and exported for the pharmaceutical industry, the price is kept artificially low through price agreements and does not reflect the resource replacement costs.

(3) A decline in the total area of natural vegetation as a source of supply for medicinal plants has occurred partly as a result of competition for the land for other uses such as forestry, agriculture, fuel supply, etc., and partly due to the commercial over-exploitation of the medicinal plant themselves. Examples where over-exploitation has occurred include Monanthotaxis capea which was formerly harvested for its aromatic leaves and traded from Côte d’Ivoire to Ghana. It is now extinct in the wild after its last remaining habitat in a forest reserve was declassified and cleared for agriculture. Pericopsis alata in Côte d’Ivoire and Pericopsis angolensis in Zambia and Malawi have both been affected by timber logging and Griffonia simplicifolia in west Africa has been affected by commercial harvesting for export for the production of western pharmaceuticals.

Focus of management effort

The need in Africa for institution building, and better staffing and funding of herbaria, particularly in high conservation priority areas is well known (Davis et al., 1986; Hedberg and Hedberg, 1968; Kingdon, 1990; Leloup, 1984). There is a great need for international co-operation to conserve large regions of high biotic diversity with small human populations such as Guineo-Congolian forest of the Zaire Basin. However, medicinal plant resource management problems exist not in these areas but in densely populated and rapidly urbanizing regions, and it is here that reaching a balance between human needs and medicinal plant resources is most urgent. This involves:

(1) identification of habitat with a high density of endemic families, genera and species with medicinal properties;

(2) pro-active management effort around core conservation areas through interaction with resource users and provision of alternatives to wild populations of threatened species; this would include species that are a high conservation priority on a national scale. In the areas visited, a preliminary listing is given in Box 2.

Sites that are a conservation priority from a more general species conservation viewpoint may therefore not be a priority with regard to conservation of traditional medicinal plants. From surveys of medicinal plant markets in selected African countries for example, it is clear that while the Cape Floral region (which is of high conservation priority due to the large number and proportion of endemic species) is not under threat from the herbal medicine trade, but from habitat destruction.

From human demography data, we know that the west and southern African regions have highest rates of urbanization. We also know that the bigger the urban settlement, the larger the traditional medicine markets (Table 2). At a macro-scale, priority areas for resource management action can be determined through mapping overlay of the main African phytochoria (Figure 6), where information is known on numbers of endemic plants, birds and mammals (Table 7) with the major urban growth points (Figure 7). As discussed previously, (Cunningham, 1990), demand is most likely to exceed supply for slow growing, slow reproducing species with specific habitat requirements (primarily forest trees). Forests also contain a high number of medicinal plant species, yet represent a small (and declining) proportion of total land area in the eastern section of Africa and trees are often subject to removal of bark or roots rather than leaves for medicinal preparations (e.g. Kenya, where forest reserves cover 2.7%; Tanzania, 1-2%; South Africa, 0.3%; (Cooper, 1985; Davis et al., 1986; Kokwaro, 1991). The urgent challenge is therefore to meet the increasing demand from rapidly growing urban areas, restore the self-sufficiency of TMPs affected by this trade, and provide acceptable alternative resources outside increasingly fragmented core conservation areas to stop over-exploitation of favoured species inside them.

Figure 6. The main African phytochoria (after White, 1983) showing one high conservation priority area (dark grey) and focal priority areas for action on medicinal plant conservation (black). I. Guineo-Congolian regional centre of endemism. II. Zambezian regional centre of endemism. III. Sudanian regional centre of endemism. IV. Somalia-Masai regional centre of endemism. V. Cape regional centre of endemism. VI. Karoo-Namib regional centre of endemism. VII. Mediterranean regional centre of endemism. VIII. Afromontane archipelago-like centre of endemism (including IX, Afroalpine archipelago-like region of extreme floristic impoverishment, not shown separately). X. Guinea-Congolia/Zambezia regional transition zone. XI. Guinea-Congolia/Sudania regional transition zone. XII. Lake Victoria regional mosaic. XIII. Zanzibar-Inhambane regional mosaic. XIV. Kalahari-Highveld regional transition zone. XV. Tongaland-Pondoland regional mosaic XVI. Sahel regional transition zone. XVII. Sahara regional transition zone. XVIII. Mediterranean/Sahara regional transition zone.

Figure 7.The relative size and distribution of major urban centres in sub-Saharan Africa(after Udo, 1982).

Conditions for cultivation as an alternative source of supply

Commercial gatherers of medicinal plant material, whether for national or international trade, are poor people whose main aim is not resource management but earning money.Cultivation as an alternative to over-exploitation of scarce traditional medicinal plants was suggested over 50 years ago in South Africa for scarce and effective species such as Alepidea amatymbica (Gerstner, 1938) and Warburgia salutaris (Gerstner, 1946). Until two years ago, no large scale cultivation had taken place. There are two main reasons for this, and both apply elsewhere in Africa:

(1) lack of institutional support for production and dissemination of key species for cultivation;

(2) the low prices paid for traditional medicinal plants by herbal medicine traders and urban herbalists.

If cultivation is to be a success as an alternative supply to improve the self-sufficiency of TMPs and take harvesting pressure off wild stocks, then plants have to be produced cheaply and in large quantity. Any cultivation for urban demand will be competing with material harvested from the wild that is supplied onto the market by commercial gatherers who have no input costs for cultivation. Prices therefore increase with scarcity due to transport costs, search time and the long-distance trade.

At present, low prices (whether for local or international pharmaceutical trade) ensure that few species can be marketed at a high enough price to make cultivation profitable. Even fewer of the potentially profitable species are in the category most threatened by over-exploitation.At present, cultivation of herbs and medicinal plants is chiefly restricted to temperate areas (Staritsky, 1980) and with the exception of India (Kempanna, 1974) and Nepal (Malla, 1982), few tropical countries have investigated the potential of cultivating medicinal plants on a commercial scale. Cultivation of herbs and medicinal plants is widespread in eastern Europe, but even where cultivation is well developed, such as in the Russian Federation, about half of the supplies are gathered from wild populations (Staritsky, 1980). In all cases where cultivation has taken place, whether in Europe, Asia or Africa, plants have been grown for profit or a high level of resource returns (e.g. multiple use species for fruits, shade and medicinal properties) and are either fast growing species, or plants where a sustainable harvest is possible (e.g. resins (Bosweilia), leaves (Catha edulis).

With few exceptions, prices paid to gatherers are very low, taking no account of annual sustainable off-take. In many cases, medicinal plants are also an open access, rather than a limited access or private resource. To make a living, commercial medicinal plant gatherers therefore “mine” rather than manage these resources. If cultivation of tree species is to be a viable proposition as an income generating activity then either:

(1) the flood of cheap bark/roots “mined” from wild stocks is reduced through better protection of conserved forests in order to bring prices to a realistic level; or,

(2) wild populations will have to decline further before cultivation is a viable option.

Cultivation for profit is therefore restricted to a small number of high priced and/or fast growing species (Box 3).

Although some of these species are threatened in the wild (e.g. Garcinia afzelii and Warburgia salutaris), low prices ensure that few slow growing species are cultivated. With the declining economic state of many African countries, it is unlikely that subsidized production of these species is likely to occur, and collection of seed or cuttings for establishment of field-gene banks (for recalcitrant fruiting species) and seed banks must therefore be seen as an urgent priority.

Strong support and commitment are necessary if cultivation is to succeed as a means of meeting the requirements of processing plants for pharmaceuticals (whether for local consumption or export) or urban demand for chewing sticks and traditional medicinal plants. If cultivation does not take place on a large enough scale to meet demand, it merely becomes a convenient bit of “window dressing”, masking the continued exploitation of wild populations. The regional demand for wild Scilla natalensis (Liliaceae) in Natal, South Africa is 300 000 bulbs yr-1, all at least 8-10 years old. On a 6-year rotation under cultivation at the same planting densities as Gentry et al., (1987) used for Urginea maritima, 70 ha would be required (Cunningham, 1988a). Due to their slow growth rates, the rotational area required for tree species would be far greater, with total area dependent on demand.

The success of cultivation also depends on the attitude of TMPs to cultivated material, and this varies from place to place. In Botswana, TMPs said that cultivated material was unacceptable, as cultivated plants did not have the power of material collected from the wild (F. Horenburg, pers. comm.). Discussions with some 400 TMPs in South Africa over a two year period showed general acceptance of cultivated material as an alternative. Similarly, TMPs in the Malolotja area of Swaziland accepted cultivation as a viable alternative. In both countries there is a tradition of growing succulent plant species near to homesteads to ward off lightning. Similarly, in Ghana, plants of spiritual significance such as Datura metel, Pergularia daemia, Leptadenia hastata and Scoparia dulcis are tended around villages. Therefore, although little is known about attitudes to cultivation of medicinal plants in west Africa, it is possible that TMPs would be in favour of cultivation of alternative supply sources.

An interesting model is provided in Thailand where a project for cultivation of medicinal plant of known efficacy has been initiated in about 1000 villages and traditional household remedies, with improved formulae, are produced as compressed tablets packed in foil and distributed to “drug co-operatives” set up through a Drug and Medical Project Fund in more than 45 000 villages as well as in community hospitals (Desawadi, 1991). Wondergem et al. 1989; WHO, 1977) have already drawn on the Thailand experience in making recommendations regarding primary healthcare in Ghana.E