Chapter 4

Potential for further development

The potential of wild plants in contributing to the objectives of farming systems development (FAO, 1990) involves making use of:

• species that are wild-harvested and for which there is an existing market, including extractivist species and both timber and non-timber forest products;
• species that are suitable for introduction into cultivation;
• species that are already cultivated or semi-domesticated on a small scale (underexploited species) and for which there is potential for more extensive or more productive cultivation.

Some species belong to more than one category; for example, some medicinal plants are both wild-harvested and cultivated. The contribution of the whole ecosystem is also important, as it provides "ecosystem services" such as soil stabilization, climate moderation and water supply.

MEDICINALS AND AROMATICS

Medicinal plants are big business. Trade in botanical medicine in the United States has been estimated at US$2 000 million, the German herbal product market was estimated in 1979 at US$1 700 million and the global trade is of the order of US$14 000 million. As mentioned in Chapter 3, many species that are used in the pharmaceutical industry are primarily wild-harvested. Others are both collected from the wild and cultivated; yet others are cultivated only. On the other hand, the greatest use of medicinal plants is made by local communities in their traditional health care systems; most of the species used are wild-collected or grown in home gardens or equivalent systems, and any surpluses are sold in local markets. There is potential for farming system development in both cases.

Wild-harvesting

Precise figures are difficult to obtain for the sources of medicinal plants, but according to Palevitch (1991) the main source of higher plants used as drugs in both modern and traditional medicines and as culinary herbs is the wild flora of developing countries. When there is reliance on wild sources production is often unpredictable, with rapid fluctuation between scarcity and oversupply as a result. The wild harvesting of medicinal and culinary herb plants has other disadvantages as summarized in Table 9. The situation is not good when demand is increasing so rapidly.

TABLE 9

Collection from the wild versus cultivation of medicinal plants

Source: Palevitch, 1991.

About 65 percent of world production of volatile oils such as lavender, peppermint, orange, jasmine and rose is derived from woody perennial plants (i.e. trees and bushes) that are either wild or cultivated (Verlet, 1993). The herbaceous plants that supply the remaining tonnage are mostly cultivated (32 percent of total production), while only about 1.4 percent of the total production is extracted from wild-collected plants.

Much of the world's production takes place in countries with a Mediterranean climate. Egypt is a major supplier of several culinary herbs such as mint and marjoram, as well as of medicinal herbs.

The use of wild-harvested species in traditional farm systems often requires considerable development effort to ensure self-sufficiency, especially in the face of the pressure on the resources caused by increased demand and diminishing plant populations.

Commercial trade in herbal medicines is usually not well developed and the local gatherers (usually farmers) have poor access to the market as there is often a long trade chain (Box 15). At the same time, with increasing urbanization and the continuing importance of traditional medicines to the growing number of immigrants from the countryside, there is often an increasing trade from rural source areas to urban markets and shops, as has happened in southern Africa.

Since uncontrolled wild-harvesting of wild medicinal plants will often lead to resource depletion, the development of this activity to contribute to farm household incomes should only be undertaken under controlled and monitored conditions. This will require the cooperation and usually the intervention of government and conservation authorities, since the local household gatherer will not have the necessary information. Farm households should be involved as far as possible in devising and implementing a sustainable strategy for the use of wild-collected medicinal plant material. Such a strategy might include:

• surveys to establish the amount of the resource available;
• listing of species whose populations are too small to sustain more than small-scale wild harvesting;
• prohibitions on the collecting of threatened species;
• issuing of permits for harvesting, indicating the volume of material that can be collected, how it should be collected and for which parts of the plant;
• consistent operation within current local, national and international legislation regarding the use and conservation of wild plants;
• use of appropriate harvesting practices;
• training of collectors;
• use of a rotation system in harvesting to allow populations to recover;
• liaison with local forestry or conservation agency staff;
• negotiation of joint management agreements (Box 16) with national park, biosphere reserve, other protected area or forestry authorities for access to and use of resources.

A participatory process is likely to be more successful and will allow the farm households to contribute their specialized knowledge of the plants, their availability and use. It is essential that household gatherers be made fully aware of any formal agreements negotiated or restrictions imposed on wild gathering of particular species or in particular areas. On the other hand, they may be unwilling to participate in any negotiations that restrict their access or limit their traditional harvesting rights.

Wild medicinal plant material is increasingly collected by members of farm households for agents who have been contracted by pharmaceutical companies to supply them, often with large quantities that are unsustainable. The collectors are normally paid very low prices for this wild-harvested material, and these low prices may also stimulate overharvesting. The maintenance of artificially low prices by a small monopoly of wholesale distributors may push the wild-harvesters to overcollect so as to obtain a reasonable return for their efforts. Some of the quantities involved are astonishing and leave no doubt as to the damage caused to the wild populations of the plants concerned (Box 17).

A much-publicized case is that of Prunus africana (red stinkwood), used to treat prostatitis, which has been the subject of a special report by the German Federal Agency for Nature Conservation (Box 18).

The use of wild-harvested material varies considerably. It may be used in small quantities by pharmaceutical companies looking for specific active compounds that may eventually lead to the development of new drugs; or large quantities may be needed. When the harvested material comprises rare endemic species that only occur in small populations and are not cultivated, a serious problem arises, as in a recent case in the United States where the extraction of the anti-cancer drug taxol from the Pacific yew, Taxus brevifolia, threatened to destroy the forests where it grew. In some cases, even when it is possible to synthesize a drug, it may prove less costly to extract active ingredients from wild plants (WWF, 1993).

Few governments seem to be aware of or show any concern about the problem of overharvesting. The data are not collected, and often the consequences are not obvious until the damage has been caused and particular species have been threatened with imminent local extinction.

Bioprospecting, genetic resources and intellectual property rights

Ideally, when medicinal plant material is wild-harvested by local collectors for a commercial company, in addition to adopting the norms recommended above, it is desirable that the collectors and the company negotiate a suitable agreement, although this may be difficult with an agent acting as an intermediary.

A number of pharmaceutical companies are engaged in screening the flora and fauna of biodiversity-rich countries such as Costa Rica, Brazil and China for genetic or biochemical resources that could lead to improved crops and new medicines. The search for economically valuable genetic and biochemical resources from wild species, now termed "biodiversity prospecting", has led to pressure for introduction of appropriate legislation, mechanisms of benefit-sharing and appropriate guidelines to control these activities with a view to avoiding unethical, unfair and unsustainable mining of the resources. A number of formal agreements between industry, research institutions and governments have been negotiated.

The best-known agreement is that signed in 1991 between Costa Rica's National Biodiversity Institute (INBio), a private non-profit organization, and the United States-based pharmaceutical company Merck & Co., Inc. Under this agreement, INBio would provide Merck with chemical extracts from wild plants, insects and microorganisms from Costa Rica's protected areas for its drug screening programme in return for a two-year research and sampling budget plus US$1 135 000 and royalties on any resulting commercial products. Another example is an agreement signed by the United States National Cancer Institute (NCI) with the Government of Cameroon following the discovery of a potential anti-AIDS drug plant source, Ancistrocladus korupensis, in the Korup forest. The agreement arranges for the provision of plant samples from the forest without destroying the resource there. The payments received from NCI will be used for Korup community development projects.

The specific contractual arrangements for biodiversity prospecting (also called "bioprospecting") that are being drawn up between some developing countries and pharmaceutical, biochemical and cosmetic companies are part of the wider issue of access to genetic resources that has been exercising governments and institutions for some time. Access to and exchange of genetic resources have been at the centre of serious and often heated debates and disputes on international policy and law. The problems centre largely on the fact that most of the world's biodiversity - whether plant, animal or microorganism - is to be found in mainly tropical developing countries and it is the perception of many such countries that they have not benefited (or not sufficiently) from the exploitation of these resources. The resources have been collected from wild plant populations growing in developing countries and then transformed into new products by the private sector (mainly pharmaceutical and biochemical companies). The developing countries and those local communities and individuals that have supplied the resources or knowledge about them have not derived much benefit from them. The fact is that those communities and local people who have invested time and effort into the conservation and development of these resources seldom receive a fair and equitable share of the benefits that are obtained from them as a result of commercial use and development. If the countries and the individuals concerned do not receive benefits for access to and use of their resources, they have little incentive to conserve them.

The Convention on Biological Diversity created a new international legal framework to regulate access to genetic resources and to promote a fair and equitable share of the benefits. Most developing countries are now facing the formidable challenge of preparing and introducing national legislation and administrative measures that will allow them to implement the relevant articles and provisions of the convention.

The rights of local communities and indigenous groups must be respected and acknowledged. The contractual approach to bioprospecting has its limitations. Those contracts that have been signed do not recognize the role of traditional knowledge as such, and its contribution to the development of biodiversity has been ignored.

It is important that all those concerned with the collection and use of wild plant resources, including notably indigenous communities, should be made aware of the issues involved, the Convention on Biological Diversity, the Convention on Trade in Endangered Species of Wild Fauna and Flora (CITES), national legislation and the various kinds of material transfer agreement (MTA) or bioprospecting contract that can be used (for detailed explanations, examples and suggested guidelines see ten Kate, 1995; Cunningham, 1993; Reid et al., 1993).

Cultivation

Medicinal and aromatic plants can provide a good source of farm development through systematic cultivation since the market demand has been increasing globally for a number of years. It appears that only a small proportion of medicinal plants that are required in large quantities are systematically cultivated. On the other hand, aromatic herbs not only have been traditionally collected since ancient times in many cultures but are now cultivated in many parts of the world, although usually on a small scale. They have considerable promise for agricultural development and as a source of income for farm households.

Many aromatic species are suitable for cultivation in arid or semi-arid zones or on marginal lands. Oregano (Origanum spp.), for example, is widely cultivated in Mediterranean Europe and North Africa. It is also grown in Mexico, where harvesting is done on a small scale and competes with farmers' work on chilies, maize and other garden crops. Wholesalers travel through the ejidos or peasant-owned lands and buy their stocks. In Argentina, in the mainly desert province of Mendoza in the foothills of the Andes, a wide array of aromatic and some medicinal plants are cultivated under irrigation by small farmers. Both native and introduced species are grown, mainly Euro-Mediterranean species, and this material not only supplies the home market but is exported to neighbouring countries. In Guatemala, a recent study has identified 15 native species of medicinal plants that have potential for industrial development. Likewise, the Agricultural and Rural Development Institute of the University of Fort Hare in Ciskei, South Africa has made a study of Ciskei aromatic plants that are suitable for the production of essential oils and hold promise for rural development.

There is often no incentive to cultivate wild medicinal and aromatic species because of the lack of a regular demand and market and the dependence on intermediaries who usually come to collect material on an irregular basis. It is important, therefore, if medicinal and aromatic plants are to be brought into cultivation, that a regular demand for the raw material be created and that value be added at site, for example through partial processing, so as to enhance the income obtained by the farmers.

The selection of species for cultivation must be made after local surveys of the availability and use of the plants in the area. The points in favour of cultivation of medicinal and aromatic plants include:

• relatively easy domestication;
• simple cultivation procedures;
• labour-intensive harvesting for some species (e.g. flowers, fruits);
• simple processing procedures;
• products that are stable (non-perishable) and can be stored for long periods;
• export potential as well as local use;
• employment creation;
• boosting of research and development efforts using local materials;
• encouragement of development of secondary processing industries to add value.

ORNAMENTAL PLANTS

Although often overlooked in considerations of the development potential of wild plants, species that are used as ornamentals have a considerable role in world trade and have great potential for development in farming systems in many parts of the world. Ornamental plants are important in most cultures and societies. In addition, the international ornamental plant trade has been conservatively estimated at US$1 000 million annually, not to mention the very substantial illegal trade.

The horticultural and flower trade is expanding rapidly around the world and new cut flower and pot plant introductions are in constant demand. In South Africa, indigenous ornamental flower species represent an untapped resource of inestimable proportions. Many species are commercially harvested from the wild and also cultivated. Cut flowers both cultivated (in the region and worldwide) and harvested from the wild include species of Leucadendron, Leucospermum and Protea (Proteaceae) and Brunia (Bruniaceae). Everlasting species of Helichrysum (Asteraceae), Aloe and Erica and thatching species of Chondropetalum and Thamnochortus (Restionaceae) are commercially harvested from the wild but rarely cultivated.

Bulbs, corms, tubers and rhizomes of wild-harvested monocotyledonous plants are exported in large numbers. In the vast majority of cases, when these and other ornamental plants are harvested from the wild they are harvested without a management plan or monitoring system. A major source country is Turkey, which is estimated to export up to 50 million wild bulbs to the Netherlands, the world's leading bulb-trading country. A list of the main geophytes exported from Turkey is given in Table 10.

TABLE 10

Ornamental plants (mainly geophytes) collected in the wild or cultivated in situ that are exported from Turkey

Source: Baser, 1997.

In Australia, many species of value as ornamentals or cut flowers have been identified and are being promoted by the nursery industry. These include many rare and threatened species, whose conservation through cultivation is being encouraged. The challenge for farm households is to seek ways of cooperating with the nursery trade and becoming involved in the cultivation of such species. The cultivation of ornamentals and cut flowers will have much more potential in countries that do not have a well-developed nursery trade or a nursery trade that is interested in wild species.

Cultivation of threatened or endangered ornamentals is one way of reducing the pressure on wild populations. Recent efforts to cultivate Mexican cycads that are threatened with extinction are described in Box 19.

Ornamental species with potential for development are to be found in virtually all regions of the world. Even in harsh environments such as the Ecuadorian Andes, a number of potential new floricultural crops have been identified recently: several wild species of Rubus such as R. macrocarpus, R. roseus, R. adenothallus and R. glaucus are considered most promising for either greatly improved production or as potential new crops for the highland tropics.

PROCESSING AND AGRO-INDUSTRY

Industrial and energy crops

Biomass of wild species in the form of wood was the main source of energy and raw materials until the mid-nineteenth century. During the period of industrialization, with increasing reliance on fossil energy sources, the use of biomass for non-nutritional purposes was largely confined to rural communities in developing countries where alternative fuel sources such as coal, gas or electricity were unavailable or too expensive for household budgets. Today biomass contributes about 13 percent of global primary energy consumption. It is mainly used in developing countries, where it is usually the primary source of energy, accounting for about one-third of total energy production; in some countries, such as Rwanda, Nepal and the United Republic of Tanzania, it accounts for 90 percent of energy production (FAO, 1996b). Figure 4 shows the distribution of the different types of energy consumption, including energy from biomass, in the world and in developing and industrialized countries.

Although sustainable production and use of biomass is not competitive today in the developed world because of the current state of technology and the economic and subsidy structure, this situation could change with biotechnological advances in biomass conversion and the growing demand for greater fuel efficiency and more ecologically friendly forms of energy production. Increases in biomass-derived energy available per unit area could be achieved by:

• introducing naturally high-yielding species;
• enhancing the area of traditional crops through further development in plant breeding, genetics and biotechnology;
• improving the efficiency of engines and conversion technologies.

A recent FAO report on potential energy crops for Europe and the Mediterranean (FAO, 1996b) gives a catalogue of species, some of them wild and many already cultivated, together with information on their characteristics, cultivation methods and utilization for energy production. The report asserts that it is vital to increase the number of plant species that might be introduced into cultivation for this purpose. Many minor species that have traditionally been used as herbs, condiments and medicinals are now being looked at for possible industrial applications. A report prepared for the Commission of the European Communities (Smith et al., 1997) highlights the wide array of species and products that could be developed for industry and energy in Europe; some examples are given in Table 11.

TABLE 11

Native European species with industrial potential

Source: Smith et al., 1997.

Commercial development of plant products

In addition to their use for direct consumption and their contribution to household economies, many wild species have significant potential for the development of cottage industries. In the case of food plants, these industries could lead in turn to the further development and use of the plants through improvements in processing, preservation and food formulations.

An example can be taken from the research project Development and Utilization of Fast-Disappearing and Under-utilized Edible Woody Forest Species of Southeastern Nigeria, sponsored by the Biodiversity Support Programme, World Wildlife Programme, Washington, DC, United States (Okafor, Okolo and Ejiofor, 1996). The main objectives of the programme were to develop appropriate research and development efforts through:

• provision of basic data and information necessary for sustainable commercial exploitation of threatened species;
• development of suitable propagation and field management techniques, as well as provision of planting materials;
• inclusion of local participation in conservation programmes and generation of local employment.

Five major approaches were used in the research:

• selection of 19 woody species that were fast disappearing and were identified as undeveloped potential sources of spices, fruits, nuts, seeds or leafy vegetables;
• nutritional and chemical evaluation;
• investigation of commercial potential through:
• extension of traditional utilization through improved processing, preservation, packaging and storage techniques,
• formulation and production of new products,
• recipe and culinary development involving local food items, leading to their popularization among Nigerian populations and expatriates;
• development of nursery practices such as seed collection, germination and vegetative propagation;
• integration of the species into the traditional farming system, with field establishment trials using agroforestry techniques.

The pathway for product development from seeds of Irvingia gabonensis is given as an example (Figure 5).

The project was multidisciplinary. The involvement of local farmers consisted of their establishing hedgerows with leguminous shrubs such as Cajanus cajan (pigeon pea) and Pterocarpus santalinoides for soil enrichment and planting the species under study in their home gardens and farms under intercropping and alley farming systems. The participating farmers also voluntarily formed a biodiversity conservation cooperative society, reflecting their interest in the project and the benefits and support obtained such as training and the supply of cash and other inputs during the period of the project.

The processing of palm products such as rattan has considerable potential for farming system development. Most of the major producers of rattan, such as Indonesia, Malaysia (except the states of Sabah and Sarawak), Thailand and the Philippines, have applied trade restrictions to encourage domestic downstream processing. In Indonesia, for example, since 1989 the exports of finished rattan equal the total value; previously most Indonesian rattan was sent to Hong Kong (now the Chinese Special Administrative Region of Hong Kong) and Singapore, where it was made into furniture or simply cleaned and re-exported. The Indonesian rattan industries currently employ about 15 000 people, of whom about 11 percent are involved in the handicraft sector (de Beer and McDermott, 1996). The impacts of these changes on farm households are difficult to assess; in principle they should enhance the income derived from harvesting the resource, but the risk of overharvesting may depress the price obtained.

Palm hearts are another product that is harvested by people who live in the region where the trees grow and processed locally, providing additional income and employment prospects. In the Choc�, on Colombia's Pacific coast, workers harvest the raw material (cogollos) of the species Euterpe cuatrecasana, known locally as naid�, and sell it to intermediaries or to the representatives of processing companies. The cogollos are then transported to the processing plant, where the remaining leaf stems are peeled and the palm hearts tinned. Tinning takes place within the region and palm heart production is important for the economic development of the region. Since it is an export product that can be processed where it is grown and harvested, some added value remains in the region (Broekhoven, 1996).

AGRICULTURAL PRODUCTION

Intensifying agriculture

As agriculture intensifies there is a tendency to focus on a smaller number of crops, primarily because, in contrast with less intensive systems of cultivation involving large numbers of species, the technologies of more intensive crop production are based on the creation of relatively homogeneous environments that tend to spread over large areas and replace locally grown species mixtures. The contribution of wild plants to enhancement of agricultural production to some extent counters this trend, mainly through their introduction as new crops, the increased use of underexploited crops or improvement in the yield and quality of existing crops obtained through the use of selected germplasm from wild species.

New crops

Although the introduction of economically viable new crops to large-scale agriculture is a long and complex process, there is increasing pressure today to seek alternative crops for small-scale agriculture, especially on marginal lands or in other areas that are not suitable for extensive monoculture. Interest is being shown, for example, in energy crops in view of the imminent shortage of non-renewable fuel supplies such as coal and oil, in fibre crops as a source of paper pulp, in medicinal and aromatic plants and in species that are adapted to arid-zone conditions.

Farmers trying to introduce new crops face many problems, such as a resistance on the part of growers, marketers and consumers to new crops, technical problems of domestication, the source of germplasm and problems of agronomy, marketing and transport. The following criteria for the selection of food-producing tree and shrub species in semi-arid regions (von Maydell, 1989) also apply to other groups and other areas:

• the species should meet demands;
• they should solve problems;
• they should be accepted by people;
• there should be no legal restrictions;
• they should have low risk;
• they should be free from negative properties or effects;
• they should be adapted to site conditions;
• they should be easy and safe to establish, with low inputs;
• they should have fast growth;
• they should produce high yields;
• the quality of products should be good;
• the crops should be compatible with other land use.

In developing a strategy for the introduction and development of a new or underutilized species, the International Council for the Development of Under-utilized Plants (ICDUP) recommended the following elements (Lazaroff, 1989):

• establishment of a network of scientists and agronomists to work on various aspects of the crop;
• establishment of a germplasm collection;
• a crop improvement programme;
• identification of factors causing environmental stress and devising of methods to reduce it;
• agronomy;
• processing and product development;
• marketing;
• harvesting and storage;
• meeting information needs.

Several groups have focused on the development of new or underutilized crops, such as the International Centre for Underutilized Crops (ICUC) in Southampton, United Kingdom, whose goal is to improve the food security, nutrition and economic welfare of human beings through assessment, development and utilization of the biological diversity of underutilized crops and species for sustainable and economic production of food and industrial raw materials. ICUC has published a number of books on genetic resources of underutilized crops (e.g. Anthony, Haq and Cilliers, 1995; de Groot and Haq, 1995; Smartt and Haq, 1997). New crops are also the focus of a series of conferences organized by Purdue University in the United States (Janick and Simon, 1990). Various networks have been created, such as the Network on Underutilized Tropical Fruits for Asia (UTFANET), the Southern and Eastern Network for African Underutilized crops (SEANUC) and those established under the International Plant Genetic Resources Institute (IPGRI) Underutilized Mediterranean Species (UMS) project, e.g. the Rocket Genetic Resources Network concerning Eruca and Diplotaxis species (Padulosi, 1995). IPGRI also publishes the monograph series Promoting the Conservation and Use of Underutilized and Neglected Crops, which gives information on cultivation, agronomy, production, prospects and related topics (Box 20).

One of the most ambitious projects dealing with underdeveloped crops is the series on promising plant species of the countries of the Andr�s Bello Convention (Bolivia, Colombia, Chile, Ecuador, Spain, Panama, Peru and Venezuela), Especies vegetales promisorias de los pa�ses del Convenio Andr�s Bello, published by the Executive Secretariat of the convention (SECAB) in Bogot�, Colombia. More than 1 000 species have been identified and more than ten volumes published containing a mass of valuable information on taxonomy, geography, ecology, properties, uses, phytochemistry, economic importance, agronomy and industrialization. Promising species are defined as those:

• that are fundamentally native to the Andean subregion (in a few cases including naturalized species);
• that have not been extensively domesticated;
• that are underutilized or little known but have economic potential in the short, medium and long term;
• for which basic scientific knowledge is available to validate their status as promising species.

A review of neglected crops of the New World has been published by FAO (FAO, 1994).

Wild palms

As already mentioned, wild palms make an important contribution to many rural households, providing not only nutrition but a wide range of other products (see Box 9 in Chapter 3). They contain many species that have either a management potential or a domestication potential. An FAO publication (FAO, 1997) gives a global list of candidate palms for domestication and management and also discusses the problems involved in domestication such as selection, breeding, propagation and commercialization.

Role of germplasm of wild plants in crop improvement

Although it has little direct relevance to the needs of farm households (see Chapters 5 and 6), increasing importance has been given in recent years to the importance of wild germplasm, in particular wild relatives of crops, fodder and pasture species and trees, as a source of material that might enhance productivity, disease resistance, winter hardiness, drought survival and similar features. The number of wild relatives is difficult to assess. A recent survey of the wild relatives of cultivated plants that are native to Europe identified some 260 species and subspecies growing on the continent (Heywood and Zohary, 1995); this is a remarkably high number for a continent with only some 12 000 species. No comparable surveys have been made for other regions. High-priority species in need of conservation are listed in Table 12.

TABLE 12

Priority list of wild relatives of crop species in need of conservation

Source: Hoyt, 1988.

A need for greater genetic diversity in many tropical crops is frequently observed. In yams - for example, the white Guinea yam (Dioscorea rotundata), which is West African in origin - a great amount of diversity is being lost throughout Africa and the future of the crop is at risk. Germplasm of wild relatives is proving an important source of new traits that will help restore diversity to the crop.

Another example is Prosopis africana, an important tree species in the Sahel which provides fodder from leaves and pods, a food condiment from fermented seeds, and fuel, charcoal and carving material from the wood, as well as having great potential for improving agroforestry systems and enhancing crop yields through improvement of soil fertility. This species is at risk of local extinction through land clearance and overexploitation. To remedy this situation, seed germplasm is being collected by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), IPGRI and national forestry programmes.

Arid and semi-arid zones

In arid and semi-arid lands, the use of native rather than introduced plants in agriculture helps to maintain soil fertility and stability and the ecological health of the fragile agro-ecosystems, thereby reducing the effects of desertification. The choice of suitable species for agricultural development is therefore critical.

Box 21 gives a list of forage and browse crops for arid and semi-arid zones of Africa, compiled from the database of the Survey of Economic Plants for Arid and Semi-Arid Lands (SEPASAL) at the Royal Botanic Gardens, Kew, United Kingdom (IBPGR, 1984). The list is just a selection of the range of plants that are available.

In arid and semi-arid lands, sensitivity of plants to salinity is an important consideration. An FAO document gives a range of species according to their salt tolerance (FAO, 1971).

Extractivism (non-timber forest products)

Although extractivism may contribute a substantial if not the major part of the yearly income of forest-dwelling families, it is beginning to show a decrease in importance from a macro-economic viewpoint in some areas. In the Amazon, plant extractivism and extractive reserves of natural forests have been considered as a paradigm for agricultural development following the assassination of the trade union leader Chico Mendes in 1988, but a recent review suggests that extractivism constitutes rather a weak base for development and that as an economy it is dying out (Homma, 1994).

In terms of potential development, extractivism in the Amazon should be seen as part of an economic cycle with four phases (Figure 6):

• expansion;
• stabilization;
• decline;
• cultivated plantation.

In the expansion phase, clear expansion of extraction may be observed, as in the case of palm hearts. The stabilization phase represents a balance between supply and demand. Decline is caused by the reduction of resources and increases in the cost of extraction, leading to a gradual fall-off in extraction; with the same effort as before, unitary costs increase, as in the case of rosewood (Aniba rosaeodora). Cultivation begins during the stabilization phase, when the conditions for cultivation are set by the availability of appropriate technologies, the lack of substitutes (natural or synthetic) and favourable prices.

Extractive reserves. Extractive reserves are territorial spaces protected by the State which are designated for the sustainable use and conservation of renewable natural resources by populations with a tradition of extractive resource use. They are regulated by concessionary use contracts backed by a utilization plan approved by the country's responsible environmental policy body. They are defined by the Brazilian Government as "forest areas inhabited by extractivist populations granted long-term usufruct rights to forest resources which they collectively manage" (Allegretti, 1994).

Production systems. Most types of extractive activities are associated with other types of production such as agriculture, fruit-tree cultivation, fishing, hunting or even small-scale animal husbandry. The potential for agricultural development involving extractivism has to be viewed, therefore, in the context of different types of production system. These in turn will depend on the local socio-economic and ecological conditions.

In the Amazon, the main types of production system are as follows, three of them involving extractivism (Lescure, Pinton and Emperaire, 1994).

• Purely extractive system. This traditional system, based solely on extractivist activities, is typified by the rubber tappers during the boom period and involves only a small number of people who do not work in agriculture. The workers are in effect a bonded labour force working for patrons. By its very nature, this system does not lend itself to agricultural development.
• Mixed agriculture-extractivism system. This system may have developed in response to the decreasing market for the products involved and the patrons' withdrawal of capital from extractivism for investment in other activities. In this system the producer combines extractivism and agriculture in varying proportions. The choice of species to be collected depends on local availability and accessibility, their market value and the compatibility of their collection with the demands of the agricultural calendar. Development of this system can be improved by setting up a small multispecies orchard around the house to provide for the consumption needs of the household.
• Three-part system: agriculture, agroforestry, extractivism. In this system, fruits, mosly from palms, form the largest part of the resources exploited. Extractivist activities are restricted to the forest near the house. Various kinds of agroforestry are practised. Cultivation is speculative and is only profitable if production balances the non-productive days needed.
• Mixed agriculture-agroforestry system. This type of system is exemplified by the small farming settlements of the v�rzea. The orchard around the house comprises some 20 to 26 species and close to the house a small 1- to 2-ha plantation of cacao is often grown under the shade of rubber trees.
• Pure agriculture system. This system, in the neighbourhood of urban centres, is essentially reduced to its agricultural component only. A small agroforestry area of orchard is maintained and fruit cultivation is focused on single-species plantations (e.g. pawpaw, passion fruit).

Potential for improvement. Improvement of the extractivist system may be obtained by:

• setting up extractive reserves;
• adding value;
• improving marketing channels;
• protecting species of economic interest;
• improving management practices;
• planting species of economic interest in agroforestry systems;
• enriching the forest in valuable species through management.