Conservation of genetic resources in tropical forest management Principles and concepts Based on the work of R.H. Kemp with scientific review by G. Namkoong and F.H. Wadsworth The designations employed and the presentation of material in this pubUcation do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of Its frontiers or boundaries. M-30 ISBN 92-5-1 03309-9 AN rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechani- cat, photocopying or otherwise, without the prior permission of the copyright owner. Applications for such permission, with a statement of the purpose and extent of the reproduction, should be addressed to the Director, Publications Division, Food and Agriculture Organization of the United Nations. Viale delle Terme di Caracalla. 00100 Rome, Italy. FAO 1993 iii FOREWORD Growing populations and pressure for social and economic development are leading to increasing rates of destruction and degradation of natural habitats, including forests and woodlands. The loss of natural resources and the degradation of land are already affecting the economies and well-being of the people in many countries, especially in the tropics. The loss of habitats are leading to accelerated rates of loss of genetic resources, which are fundamentally important in the adaptation and improvement of plant species presently under cultivation and those whose value is yet to be ascertained. It is today widely recognized that the existing Protected Area system on its own is not sufficient to provide the necessary geographic and biological coverage to conserve either the exceptional diversity of tropical forests or the genetic resources of their main component species. It is also recognized that, to succeed, conservation must be seen not as a constraint but as an integral part of development. Although forest management interventions will cause more rapid changes in the composition of ecosystems than natural forces and, at times, will also accelerate or alter successional changes, such interventions can be rendered compatible with the conservation of the genetic resources of the species under use. The sustained utilization of forests to meet present-day needs coupled with the maintenance of a network of areas dedicated to the protection of ecosystems and their functions, provides the only solution for lasting, genetic conservation. In accordance with its mandate, FAO has for the past 40 years published a number of manuals on the sustainable management of tropical forests, complemented over the past 20 years by guides on the conservation of forest genetic resources. The technical feasibility of both tropical forest management and conservation have been stressed repeatedly in these documents. The present book constitutes a first step towards a more systematic approach to the provision of guidelines for harmonizing sustainable utilization and conservation of genetic resources of tropical forest trees. Many presently prescribed forest management interventions could with minor adjustment be made less harmful to conservation concerns. Conversely, some compromises could be made in existing methodologies for the conservation of forest genetic resources which could help achieve the main aims of conservation while at the same time meeting pressing, present-day needs for the goods and environmental services provided by the forest. This document outlines present forest management practices, illustrated by case studies from three tropical countries. It briefly reviews available strategies and methodologies fof the conservation of forest genetic resources in the light of their compatibility with sustainable use of the resources targeted for conservation. It is planned to publish, in the near future, a companion volume to the present book, in which more specific guidance is given on the management of specific forest types and tree species, in situations in which varying degrees of priority are given to production and conservation respectively, while at the same time meeting at least some of the needs of both of these two complementary aspects. f.P. Lanly Director Forest Resources Division IV Acknowledgements The present study is based on the work of Mr. R.H. Kemp of the United Kingdom and was carried out under the technical guidance of staff of the Forest Resources Division of FAO. Professor Gene Namkoong (U.S.A.) and Dr. F.H. Wadsworth (U.S.A.) provided scientific review of the document, which also benefited from the comments of Dr. J. Wyatt-Smith (U.K.) and colleagues in IUCN, Unesco, the Royal Botanic Gardens, Kew (U.K.) and the Oxford Forestry Institute (U.K.). Thanks are due to Dr. D. Boshier for providing information related to Cordia alliodora (see Appendix 1 ) , and to the Overseas Development Administration (ODA, U.K.) for providing information and material related to the Case Studies on Ghana and India. Valuable inputs were further made by J.R. Palmer, M.E.D. Poore and T.J. Synnott, among many others. Greatest credit for achievements in the field on which information in this document is based, however, is due to the dedicated efforts of national forestry and other staff in the countries concerned. The deep interest of professional foresters in the nature, complexity and functioning of the natural forests and their openness and willingness to discuss these issues and to share their in-depth knowledge for common benefit, is gratefully acknowledged. TABLE OF CONTENTS Page Executive Sumary vii Glossary xi Key to Abbreviations xiv PART I 1 . Introduction 1 2. The Nature of Forest Genetic Resources 5 2.1 Levels and structure of genetic diversity 5 2.2 Ecosystem conservation 6 2.3 Conservation of target species 7 2.4 Conservation of provenances 7 2.5 Values of genetic diversity 8 2.6 Use values and option values 9 2.7 Precautionary values 9 2.8 Existence value 10 2.9 Location of conservation areas 10 2.10 The link to production forests 11 2.11 Size of conservation areas 11 2.12 Dynamic conservation 12 2.13 Disturbance and succession 12 2.14 Logging and genetic diversity 13 3. Impacts of Management in Production Forests 15 3.1 Continuity and control 15 3.2 Economic and market influences 16 3.3 Forest inventory 19 3.4 Forest dynamics 21 3 . 5 Regeneration 23 3.6 Silviculture 25 3.7 Harvesting 28 3.8 Non-timber forest products 30 3.9 Involvement of local people 32 4. The Future of Tropical Forests 35 4.1 Population and land use 35 4.2 Timber demand and international trade 36 4.3 Tropical forests and environmental concerns 39 4.4 Protected Area systems 41 4.5 Buffer Zone forest 43 5. Strategies for in situ conservation in procuction forests 45 5.1 National policies 45 5.2 Management information 48 5.3 Management systems 49 5.4 Management plans 52 PART II CASE STUDIES 59 6. Ghana 61 6.1 The economy 61 6.2 The environment 61 6.3 Diversity 62 6.4 Management for timber production 62 6.5 Policy: linking production and conservation 64 6.6 Non-timber forest products 64 6.7 Forest revenue systems 65 6.8 Forest inventory 67 6.9 Setting priorities 68 6.10 Management and harvesting 70 6.11 Regeneration and silviculture 71 6.12 Reproductive biology 72 6.13 Integration and security 72 7. Brazil: the Amazon forests 73 7.1 Legal framework 73 7.2 Setting priorities 75 7.3 Management options 76 7.4 Secondary forest and non-timber forest products 77 7.5 Information, research and coordination 78 8. India: the Western Ghat forests, Karnataka 79 8.1 National policy 79 8.2 Western Ghat forests 80 8.3 Strategy for integrated development and conservation 81 Appendix 1. Methodology of a study of the reproductive biology and genetics of Cordia alliodora 85 Box 1. The role of logged forests in the conservation of species richness and genetic diversity 14 Box 1. Management of diversity through diversity of management 56 References 91 Cover photo; Recently logged forest, Sarawak (Malaysia). R.H. Kemp vli EXECUTIVE SUMMARY General Considerations 1 . There is today a growing realization at national and international levels of the value of forests as a renewable resource and of their role in the production of a range of goods and environmental services. The latter include the role of the tropical forests as a source of genetic materials for the adaptation and improvement of plant species presently under cultivation and use, and those whose value is yet to be ascertained. 2. Increased populations and pressure for social and economic development in most tropical countries contribute to a continuing trend of diminishing areas under forest cover despite concern for the tropical forest. They also severely limit the possibilities for extending existing systems of fully protected areas. 3. While the exceptional biological diversity of tropical forests constitutes a unique national and global asset, the extent and integrity of this diversity are therefore rapidly diminishing. 4. The continuing availability of diversity and genetic resources is fundamental for the sustainable development of nations. Conservation of genetic resources of forest trees and other woody species is closely related to all other forms of diversity, and is essential for sustaining the productive and protective values of the forest. 5. In spite of their importance at national as well as local levels, forests containing socio-economically valuable tree species have rarely been targeted when planning and establishing Protected Areas. Therefore, managed production forests play a key role in programmes aimed at the conservation of genetic resources of such tree species and are a necessary complement to conservation efforts undertaken through Protected Area management. 6. Management interventions in the forest can be aimed mainly at the production of timber, wood and other products, the protection of soil and water, or the conservation of biological diversity and genetic resources. Productive and protective purposes can be rendered compatible with conservation concerns through sound planning and inter-sectoral coordination of activities at the national level. 7. International concern for the conservation of biological diversity is likely to lead to increased support to tropical countries through both aid and trade channels. Since it is now widely recognized that the Protected Area system on its own is insufficient to provide the necessary geographical and biological coverage, such assistance may in the future be directed increasingly to support forest management carried out with due concern for genetic conservation. Conservation of Forest Genetic Resources 8. Genetic resources are associated with the different levels of diversity that exist in nature, from ecosystems to species, populations, individuals and genes. These levels interact closely and all must be considered when conservation objectives are defined and when corresponding action is specified. V111 9. Conservation of genetic resources in situ is dependent on maintaining the essential functional components of the ecosystem. It implies the planned and systematic management of identified target species in a network of conservation areas which will include Strictly Protected Areas as well as managed forests and multiple-use reserves. 10. Tropical forests are dynamic and subject to change through natural disturbance and succession; the aim of conservation is not to freeze a given state but to contain a dynamically evolving system. 11. Lack of information on population biology, reproductive systems, variation and genetics of most tropical tree species limits the possibilities for the deliberate management of their genetic resources. Maintaining a broad genetic base through the conservation of a range of provenances of target species is likely to be the safest available option until more adequate data are available. 12. Conservation efforts must be planned at the national level, but close linkages to regional and global efforts are necessary to ensure success. Conservation of Genetic Resources and Forest Management 13. The sustained utilization of forests to meet present-day needs coupled with the maintenance of a network of areas dedicated to the protection of ecosystems and their functions, provides the only solution for lasting, genetic conservation. Harmonizing conservation and management for the production of goods and services is especially important in relation to tropical tree species which are not included in plantation and domestication programmes. 14. The domination of short-term economic and market forces over ecological and technical considerations have frequently been the cause of past failures to attain sustainability in natural forest management and conservation of the species being utilized in the tropics. 15. Information on forest composition and growth is critically important for both sustainable production and genetic conservation; broadly-based inventories, including botanical surveys, regeneration sampling and information on non-timber forest products, can be combined to serve both objectives. 16. Some silvicultural operations, including canopy manipulations to favour certain species and individuals, can lead to a reduction in the overall diversity of tree species in a stand. However, these practices might also be skilfully used to maintain or restore diversity, in selected areas. 17. Logging is at present commonly the only large-scale management intervention in the tropical forest. It may either reduce or enhance species and intra-specific diversity. It may furthermore contribute to the depletion or conservation of the genetic resources of the principal species being utilized, depending on the timing, intensity, frequency and discrimination employed, and on the effectiveness of protection and management of subsequent regeneration. ix 18. Large scale intensive logging, on short felling cycles, with poor harvesting control, may alter the species composition and may damage both forest structure and site quality. Logging damage is indiscriminate in its impact on genetic resources but heavy logging tends to enhance the opportunity for pioneer species. Should this be the case, special attention may have to be paid to the conservation of rare populations and species of later successional stages. 19. While harvesting of mature trees of good quality is among the stated objectives of forest management aimed at the production of timber, pressing market demands coupled with inadequate forest management practices may lead to highly selective harvesting having negative (dysgenic) effects on the future development of the stand. Silviculture rightly calls for harvesting of "the best", but this must not be done without due consideration to regeneration potential and the quality of the next generation crop. 20. The harvesting of non-timber forest products will widen the range of management objectives and will diversify management interventions. Hence it may increase the possibilities for sustainable management and conservation of genetic resources both by broadening the range of species contributing to production, and strengthening support of local communities for forest conservation through the provision of direct benefits to people living in or close to the forest. Action Required 21 . Continuity of management based on adequate information on forest composition and dynamics, is essential to meet the production and conservation objectives of areas presently under forest. Forest management should be carried out within the framework of established plans acceptable to all land users and executed by well-trained, informed and motivated staff. 22. The potential contribution of each area of production forest to the conservation of diversity and genetic resources of its component species, and the setting of relative priorities between areas in terms of production and conservation objectives, must be determined as part of an integrated strategy, based on national policies, with appropriate regional and international linkages. 23. Integrated action extends beyond the management of production forests and Protected Areas, to aspects of forest industry, marketing and trade, and to other sectors in the context of national development policies. 24. Involvement of local people living in or benefiting directly from the forest is fundamental to success of any conservation and management efforts, in the short as well as in the long term. 25. The formulation of a National Strategy for the Conservation of Forest Genetic Resources is the most appropriate means to secure the necessary integration of action at the national level, and to define appropriate institutional mechanisms for its implementation. 26. The formulation of such a National Strategy for Conservation is an essential nucleus for broader action in the conservation of terrestrial biological diversity. It will also provide a solid and credible basis for the substantial international support which will be required to meet national and global conservation objectives. 27. The scale and complexity of the data needed to meet national conservation, protection and production objectives will require coordination through some form of a national conservation data centre, which should help facilitate and stimulate the collection of related data, make such data easily accessible for field implementation and use, and provide international links. 28. Specific forest management systems must be decided for each forest or unit of management in accordance with their specific roles under the National Strategy for Conservation. This will help achieve an appropriate balance within the national forest estate as a whole, with due weight to ecological as well as socio-economic objectives. 29. Special attention and priority may have to be given to the conservation of forest genetic resources in certain areas of the productive forest estate, for example at the edges of a species 1 geographical or ecological range, where the populations are likely to be genetically distinct, and where they may be particularly vulnerable to disruption. 30. Special attention must further be paid to the generally species-rich secondary forests in various stages of recovery, and to the conservation of genetic resources of species characteristic of the mature-phase or climax forest. This may at times involve the adoption of some constraints on management for production, in specific areas of the natural range of selected species or communities, at any given time. 31 . Many practical aspects of concession allocation, duration and operation, of critical importance to sustainable forest management for timber and other products, also decisively influence possibilities for the conservation of genetic resources. Special ecological, taxonomic or other advice maybe needed in relation to conservation concerns, for example in the location and distribution pattern of seed trees or patches of forest to be left temporarily unlogged, to ensure regeneration and maintenance of desirable levels of inter and intra-specific diversity. 32. In all aspects of forest management and conservation the failure to comply with prevailing management prescriptions has been to date a common cause of unnecessary damage to site, growing stock and regeneration. Strict monitoring and control, based on criteria specified in corresponding forest management plans and the National Strategy for Conservation drawn up with due concern to environmental, technical, social and economic concerns, should be carried out on a continuing basis in each management unit and at national level, 33. Intensified research into many aspects of taxonomy, forest dynamics and the functioning of ecosystems and individual tree species (including reproductive biology), are needed to meet both production and conservation objectives. Prioritization and coordination of research under the National Strategy for Conservation are essential to optimize use of the limited scientific resources available. xi Allele Apomixis Autecology Biological diversity Biome Climax Chromosome Chorology Community Conservation of genetic resources Disclimax Ecosystem Endemism Ex situ conservation Frugivorous Gene GLOSSARY Any of the different forms of a gene which may occupy the same position (locus) on a chromosome (see also gene). Reproduction without fertilization, so that the progeny are genetically identical to the parent. The ecology of an individual species as opposed to communities. The variety of life forms, the ecological roles they perform and the genetic diversity they contain. Major regional ecological community. Terminal stage of ecological succession. Threadlike body found within the nucleus consisting primarily of DNA (desoxyribonucleic acid) and a protein sheath, and containing the genes responsible for most hereditary traits. Description and delimitation of the distribution of a species or other taxon. A group of different organisms in close interdependency and inhabiting a common environment. The management of human use of genetic resourcs so that they may yield the greatest sustainable benefit to the present generation, while maintaining their potential to meet the needs and aspirations of future generations see also in situ conservation; ex situ conservation). A successional stage maintained short of the true climax by e.g. fire. A community with its physical environment, interacting as a functional system. Natural occurrence confined to a particular locality. Any conservation method that entails removal of individual plants or propagating material (seed, pollen, tissue) from its site of natural occurrence, i.e. conservation "off- site* 1 in gene banks as seed, tissue or pollen; in plantations; or in other live collections, such as ex situ conservation stands. Fruit-eating. Basic unit of inheritance; the physical entity being transmitted during the reproductive process, and influencing hereditary traits among the offspring. Genes can exist in different forms or states, called alleles. Xll Gene flow Gene pool Genetic diversity Genetic drift Genetic resources Genotype Guild Inbreeding IS situ conservation Isozyme Keystone species Mutation Origin Exchange of genes between populations owing to the dispersal of garnets or zygotes. The total sum of genetic material of an interbreeding population. The heritable component of variation. Genetic diversity occurs at the gene level, the individual level, the population level, the species level and the ecosystem level. Genetic diversity is one component of biological diversity (see above) Random changes occurring by chance in the genetic make- up of small, isolated populations. The economic, scientific or social value of the heritable materials contained within and between species. Genetic constitution of an individual (particular set of alleles). A group of species having similar ecological requirements and roles in the community e.g. pioneer species. Mating of closely related individuals. Conservation of genetic resources of target species "on- site 11 , within the natural or original ecosystem in which they occur, or on the site formerly occupied by that ecosystem. Although most frequently applied to populations regenerated naturally, in situ conservation may include artificial regeneration whenever planting or sowing is done without conscious selection and in the same area where the seed or other reproductive materials were collected. Variants of a given enzyme performing the same catalytic function but exhibiting the presence of different genes (syn. isocnzyme) . Organisms which are critical in the functioning of an ecosystem (syn. keystone mutualists, pivotal species; see also Chapter I, page 6). Sudden heritable change in the gene or chromosome constitution causing changes in number, structure, size or sequence. For an indigenous stand of trees the origin is the place in which the trees are growing; for a non-indigenous stand the origin is the place from which the repro- ductive materials were originally introduced (see also provenance) . Outbreeding Mating of non-related (or distantly related) individuals. Kill Phenotype Population Provenance Secondary forest Species Stand Stochastic Sustainable forest management Tropical Hoist Forest The observable character of an individual resulting from interaction of the genotype with the environment. A group of interbreeding individuals occupying a particular area and usually separated to some degree from other similar groups. Population of a species referred to by its locality of occurrence; the place in which any stand of (indigenous or non-indigenous) trees is growing (see also origin). Forests which have suffered various degrees of disturbance as a result of shifting cultivation, or various intensities of logging, as opposed to supposedly virgin forest, or mature-phase forest which has achieved more or less the climax stage of development. One or more populations, the individuals of which can interbreed, but which cannot exchange genes with members of other species. A community of trees possessing sufficient uniformity of composition, constitution, age, spatial arrangement or condition, to be distinguishable from adjacent communities, so forming a silvicultural or management entity. Founded on the properties of probability and chance or random variation. The administrative, economic, social, legal, technical and scientific aspects of the conservation and use of forests within the framework of a technically sound and politically accepted overall land use plan. It implies various degrees of human intervention, ranging from action aimed at safeguarding and maintaining the forest ecosystem and its functions, to favouring given socially or economically valuable species or groups of species for the improved production of goods and services. The term can only be more precisely defined in terms of the management objectives of a particular forest, however, it should always incorporate the principles of sustainable development, to "meet the needs of the present without compromising the ability of future generations to meet their own needs". It therefore has essential objectives in common with the conservation of genetic resources. Both concepts can be most effectively and efficiently applied in the context of the national forest estate as a whole. The closed high forests in the tropics, where the dry season is short (4 months) or non-existent. It includes both rain forests and monsoon (seasonal) forests. XIV KEY TO ABBREVIATIONS ECG Ecosystem Conservation Group (presently includes representatives of FAO, Unesco, UNEP, UNDP, IUCN, WWF) FAO Food and Agriculture Organization of the United Nations IUCN The World Conservation Union (earlier known as the International Union for Conservation of Nature and Natural Resources) UNDP United Nations Development Programme UNEP United Nations Environment Programme Unesco United Nations Educational, Scientific and Cultural Organization WWF World Wide Fund for Nature (earlier known as World Wildlife Fund International) WRI World Resources Institute CHAPTER I INTRODUCTION International concern for tropical forests has risen sharply during the past decade. This period has seen the start of the Tropical Forestry Action Programme (TFAP) as a framework for action in tropical forestry, and the creation of the International Tropical Timber Organisation (ITTO) as a forum of consumers and producers of tropical timber moving in international trade, and discussion of the problems of deforestation at the highest international levels, in the World Commission on Environment and Development, in the UN General Assembly, in the Intergovernmental Panel on Climate Change, at summit meetings of the group of seven leading industrialised countries, and last but not least the United Nations Conference on Environment and Development in Rio de Janeiro, Brazil in June 1992. At the same time the FAO Forest Resources Assessment 1990 Project has revealed that the rate and scale of tropical deforestation in the past decade have been even greater than was previously predicted (FAO 1992c). Central to any solution to this problem must be the reconciliation of the need for development and increased living standards in the tropical countries, based on the sustainable utilization of each nation's resources, including the natural forests, with the conservation of their capability for renewal and adaptation to changing conditions and needs: their genetic resources. The problem is complex, being bound up with broader aspects of land use and population increase, as well as the impacts of international debt, trade and aid relationships. Part of the difficulty in the search for quick and effective solutions is the very complexity of the tropical forests themselves. In terms not only of the many varieties of dry forests, moist forests, hill and coastal formations, but also the different social and economic conditions that surround them, there is not one but many thousand forest problems, each in need of separate analysis and management. One aspect of this complexity of the tropical forests that has now attracted international interest is their biological diversity. This brings a new appreciation of this unique value of the natural forest ecosystems, in addition to their productive and environmental values, which have often proved insufficient to ensure their protection. Diversity considerations are generally poorly, or not at all, represented in formulae used to determine the economic value of an area, and have been discounted in competition with the immediate needs for land, food production and revenue from intensive timber production, for example through fast-growing plantations, in place of natural forest. As human populations continue to increase the opportunity cost of setting aside large areas purely to conserve biological diversity for ethical reasons and often un-specified "future needs", becomes increasingly difficult to defend. At the same time pressures to maximise short-term timber volumes and revenue from production forests may increasingly threaten the continued existence of natural forests under selective harvesting systems, in which funds and revenue are re- invested in care and management of the areas harvested. However to the extent that the conservation of genetic diversity can be seen as one of the services of natural forest under multiple-use management, including appropriate levels of sustainable timber production, the reconciliation of developmental and conservation objectives may be easier to achieve. The exceptional wealth of the tropical forests in biological diversity makes this a strong possibility, given adequate provision for effective management. In the wake of preferences for other land use options, the progressive reduction in size and increasing isolation of the remaining areas of natural forest in most countries imposes the need for increased efficiency in their management. This is partly to ensure the maximum levels of production, to defend the forest against increasing demands of increasing populations, but also to the increased stresses on the ecosystem resulting from the loss of the "buffering" effects of large areas of surrounding forest. This changed situation may have direct impacts on the environmental conditions around the forest, and also indirect influence on the breeding and seed dispersal systems of tree species dependent on animal vectors. The component of the genetic diversity in the forest which is of actual or potential use, either for production or to maintain the forest as a functioning system, constitutes its genetic resources. The conservation of genetic resources is absolutely fundamental to the long-term success of all other forms of diversity conservation (Riggs 1990) and is also essential to the sustainable and productive management of the forest ecosystem in which they occur. To this extent therefore in situ conservation of forest genetic resources should reinforce the conservation and management of production forests, and vice versa. At the same time the systems of National Parks and other fully Protected Areas also have an essential role, particularly in the conservation of biological diversity of very uncertain use value or existence value, as well as the genetic resources of species of economic importance. However the coverage of the tropical forest biome within such systems is limited and they will never satisfy the full needs of genetic resource conservation. New approaches are therefore needed to integrate national conservation activities for maximum effect in both production forests and fully Protected Area systems. FAO has the leading international role in linking forest management to the conservation of forest genetic resources, particularly through the work of the FAO Panel of Experts on Forest Gene Resources, established in 1968. Folliwing publication in 1975 of the book, " Methodology of Conservation of Forest Genetic Resources" (FAO 1975), "Guide to in situ Conservation of Genetic Resources of Tropical Woody Species" was published by FAO in 1984 (Roche and Dourojeanni 1984). These documents reviewed the issues and activities involved in genetic resource conservation in forestry. Three case studies, from Cameroon, Malaysia and Peru, were published in the later document, which further reviewed the role of Protected Area systems, including those within production forests, such as the Virgin Jungle Reserves in Malaysia, which were discussed in detail. The principles and procedures set out in both documents still provide valuable guidance to the subject. Since then the compilation and publication of reviews of management practices in tropical moist forests in Africa (FAO 1989b), Asia (FAO 1989c) and Latin America (FAO 1992a) together with other broader reviews of the subject from both the managerial and the ecological viewpoints (e.g. Mergen and Vincent 1987; Wyatt-Smith 1987a; Poore 1989; Whitmore 1990; Gomez-Pompa et al 1991), have provided new opportunities to consider the possible role of production forests in relation to the conservation in situ of forest genetic resources. The FAO Panel, at its 7th Session in December 1989, endorsed proposals to update and expand the guidance given in the 1984 publication on in situ conservation of tropical woody species. In doing so it was decided to concentrate on the role of tropical forests under management for the production of timber or other products and services in the conservation of the genetic resources of the woody species. These, and particularly the largest and longest-lived trees, are dominant in determining the unique genetic potential of the system. The diversity of tropical forests is a product not only of the large number of species present on a given area but of successional change over time, from the colonisation of gaps or cleared areas by pioneer species, through complex successional stages, to the mature "climax 11 forest. Economic forces and market demands have produced management systems aimed to simplify and to truncate the natural complexity and successional stages in the forest, to concentrate the growth potential into relatively few species, and short cutting cycles, with very limited reinvestment in management activities. This situation, in which short-term economic considerations have frequently overridden ecological concerns, has been governed by the apparently low value of the production per unit area of forest, as expressed in fees and log values in the forest. This has been generally incompatible with the level of investment in management needed to conserve the genetic resources even of the most important economic species, except perhaps and incidentally, those of the pioneer and fast-growing gap-opportunist guilds. Even on the basis of the very limited knowledge of the species composition and intraspecific variation in the forests it is clear that large parts of the useful or potentially useful genetic resources are in danger of being lost. Resolution of this problem is dependent on a wide range of integrated action, within which scientific research and professional forest management are only part of the solution. The rapid pace of change in market demands and opportunities has often rendered invalid both the initial objectives of management and the programmes of data collection or silvicultural action undertaken to achieve them. The danger of adopting too narrow and short-term objectives is even greater in respect of genetic resource conservation, which has to consider possible changes in needs and opportunities over much longer periods than one or two rotations. These are likely to include rapid advances in the means to handle, manipulate and recombine genetic material. At the same time advances in information technology are already transforming the possibilities to handle and interpret large and complex arrays of data for the better understanding and use of functional relationships in the management of the forest. These developments greatly improve the opportunities for multiple-use management, including the conservation of genetic resources. A very important aspect of such multiple-use approaches, both for the long-term security of the forest ecosystem and for the conservation of genetic diversity, is the incorporation of non-timber forest products and the interests of local communities within the management system. Whereas in the past, with more limited resources for managing complexity in the system, and narrower management objectives, the response to the forest's diversity was to simplify the system, by concentrating action on few species, often with unpredictable and unexpected results, much of the current and future value of the natural forest is seen to lie in its genetic diversity. A more appropriate future response to the management of diversity will be greater diversity of management. This may be achieved in various ways and at various levels, from multiple-use of the same area of forest, either simultaneously or in successional stages, through separate management by compartments or working circles, to the integrated and diversified management of the entire national forest estate, embracing production forests, protection forests, genetic conservation areas, Protected Area networks, and areas combining two or more of these functions. Even for the management of genetic resources of species whose natural range transcends national boundaries the national borders define the ultimate unit of effective management, whose freedom and effectiveness to act may be further enhanced through international cooperation. It is ultimately the national government which holds the power to formulate the necessary policies in land and resource use which govern the possibilities both for sustainable forest management and for the conservation of the nation's biological civersity and genetic resources. Each country, and to a large extent each forest area, is unique in terms of its genetic resources and of the appropriate strategies at both national and local level to manage the forests, both for production and for conservation objectives. Attempts at management solutions at the level of the individual forest, alone and in isolation from its setting in national development policies, not only for forest and land use but embracing forest industry, trade and the linkages with other sectors, can only have very limited chance of success. For this reason a detailed case study of the problems facing the conservation of forest genetic resources in one country (Ghana) and of approaches to their solution, with shorter contrasting examples from two other countries (India, Brazil), form a large part of this overall study of policies and activities needed to link conservation with forest production through appropriate management. These case studies are presented in Part II of the present document. In Part I, Capter II brief ly summarises the essential elements of forest genetic resources. Chapter III considers the nature and extent of the impacts of management for timber production on the genetic resources of the forest, and some possibilities for multiple-use management. Chapter IV attempts to foresee the future for the natural forests, in terms of both international and domestic demand for timber and other products, and the probable influence of other factors such as population increase, and environmental concerns, including international interest in the conservation of biological diversity. Finally in Chapter V suggestions are made for the formulation of National Strategies for the Conservation of Forest Genetic Resources. Appendix 1 provides an example of the careful research needed to provide information on the reproductive biology and genetics of individual target species within a national strategy. It is hoped to complement the present study in the future with more detailed guidelines aimed at harmonising sustainable forest management for productive purposes with the conservation in situ of the species, provenances and populations under use, for the benefit of these two complementary objectives. CHAPTER II THE NATURE OF FOREST GENETIC RESOURCES In the same way that the term "forest resources" refers to the usefulness of the forests for the production of timber or other products for human benefit, the term "genetic resources" implies that elements of the genetic variability of the trees and other plants and animals will be used to meet human needs and objectives. However as compared with the harvest of timber or the collection of other products from the forest, which can be used immediately, the benefits from the genetic resources can be used not only in current programmes but are essential for future development in the next and subsequent generations; and for the continued adaptation of the resources to changing environmental conditions and human needs. The other important aspect of the genetic resources of natural forests, especially the tropical forests, is their great diversity, and this range of variation provides the basis for selection and improvement of the products and other benefits to meet future needs, so far as they can be foreseen. Therefore the greater the uncertainty over future demands, the greater the potential value of conserving diversity. Apart from the relatively small number of tree species of current economic importance, or domestic importance to local communities, which must be the main focus of conservation programmes, there are likely to be several hundred or even thousands of other species present with lesser or unknown values. Some of these might form an important part of future harvests of wood, timber and other forest products, in response to changing environmental conditions or market demands. 2.1 Levels and structure of genetic diversity Genetic diversity occurs at various levels of organisation from the ecosystem, through its component species, their sub-specific populations (provenances), family groups and individual genotypes to the molecular level, of the gene. While ecosystem conservation may also achieve the conservation of some included species and genotypes others might be lost unless, for example in the case of important timber trees, the species and its genetically distinct component populations, are also targeted for specific conservation measures (FAO 1989a). Within such populations there is likely to be substantial variation between individual trees (genotypes). Depending on the pattern of distribution of such variation through the stand, as a result of the nature of the species' breeding and dispersal systems, highly valuable genetic resources at this level may in turn be lost, even if the population as a whole survives. In this respect, special care is needed to ensure that progeny of the best individuals of desirable species, when logged at maturity, is adequately represented in the existing regeneration, to avoid dysgenic effects and loss of diversity. At the level of the gene allelic differences could be the basis of valuable traits, for example in resistance to insect pests or to severe environmental stress, of great potential value for adaptation to changing environmental conditions and for future use. The rapid advances in genetic engineering may permit new combinations of genetic characteristics which could transform the possibilities for highly productive or otherwise desirable plantations, making best use of deforested and degraded land. The chances of recognising such genetic variation at that level in the natural forest are clearly remote. However the strong probability of such potentially valuable genetic resources occurring in highly diverse populations is a further reason to guard against inadvertent loss of diversity. It is therefore essential that all levels of genetic diversity be considered, and to the extent appropriate and practicable, included in the objectives and the activities of a conservation programme (Namkoong 1990). Moreover it is the organisation and structure of genetic diversity at the various levels that underlie both the functioning of the ecosystem and the approaches to the conservation of the genetic resources of individual species (Riggs 1990). The genetic structure of a species is defined by the way in which genetic variation is distributed between and within populations. This structure is the result of mutation, migration, selection and gene flow between separate populations and is strongly influenced by the genetic system, embracing the mating system and dispersal systems for pollen and seed. Information on the diversity and distribution of genes within species and their local populations is essential to the management and conservation of their genetic resources, but such information is very limited or non-existent for most tropical tree species. 2.2 Ecosystem Conservation As stressed above, the conservation of the diversity of the natural forests is dependent on maintaining all essential functional components of the ecosystems in situ. These are likely to involve a range of ecological interactions, particularly symbiotic relationships and interdependent connections, for example between tree species and their animal pollinators, seed dispersers and so forth. Although the objective may in many cases be the conservation of particular target species and populations, in practice this is likely to involve conserving whole communities, at least until we have a more complete understanding of ecosystem dynamics. Such research as has been done commonly indicates hidden complexities and interactions, for example, among so-called "plant-web"- or "food-web" systems (Gilbert 1980; Terborgh 1986; Whitmore 1990). A wide variety of animals, including many groups of invertebrates, birds, bats and other mammals, are involved in pollen and seed dispersal of trees of known economic importance and may in turn be dependent for their survival, in the areas of forest where they are needed, on sources of food or breeding niches provided by quite different and apparently insignificant trees or other plant species. The disappearance of such "keystone" or "pivotal" species might then lead to the loss of plant species, including timber trees, dependent on the animals for pollination or seed dispersal (Howe 1990). Although the great majority of species in mixed tropical forests may play no essential part in ecosystem function, and may therefore be in that sense redundant, the present level of knowledge and understanding is inadequate to determine with certainty all the key components of the ecosystem. The precautionary principle therefore requires that management and harvesting practices should conserve as wide a spectrum of species of as yet uncertain linkages and values as is practicable. Such an approach is consistent with the growing international concern for the conservation of biological diversity in general. Tropical forests and woodlands, in addition to the genetic resources of woody species they contain, are the habitat for a wealth of other plant and animal species, all of which contribute to the total sum of genetic diversity and resources on earth. We may therefore distinguish four arbitrary categories of forest genetic resources for conservation in situ; (i) the principal economic tree species (ii) other trees, plants and animal species of known but lesser value in the national economy or to local communities (iii) key functional species for sustainability of the ecosystem (iv) other elements in the total biological diversity, of as yet uncertain value. 2.3 Conservation of target species Just as the overall objectives of forest management for productive purposes will give preference to the desirable trees of the principal economic species these species will also be a main target for in situ conservation of the forest's genetic resources. This will aim to maintain viable breeding populations, and to favour reproduction of the better individual trees, insofar as this may be judged from their phenotypic appearance in the forest. This objective should be a normal component of all sustainable forest management but in practice is likely to increase the cost of harvesting and management operations as they are currently conducted in most countries. To add further to the costs of management and harvesting, through greater care to avoid damage to seed trees and regeneration of additional "non-economic" species, would therefore require the imposition of special conditions, through incentives or constraints. Nevertheless the maintenance of viable breeding populations of "keystone" species and those important for domestic use by local communities may be essential for long-term sustainable management of the forest for all its values, including the production of timber. Some overall loss of biological diversity must be inevitable as a result of harvesting and management practices, at least for a period and in the areas affected by these operations. A single management system, if consistently practised, will affect all stands similarly. The timing, frequency, intensity and extent of logging will be the principal factors which determine the severity of the impact on overall biological diversity. However the use of diverse management systems can increase the variation between management units. The impact on the principal target species will depend also on the level of understanding of the pattern of variation within the species, both within and between different populations in the forests, and of its interactions with other species in the ecosystem, and then the willingness and capability to adjust management and harvesting operations to minimise losses in diversity. 2.4 Conservation of Provenances The practical implications of the need to understand the genetic structure include the likelihood that genetically distinct populations of a species may exist in different forest areas within the species 1 range (i.e. different provenances of the species), as a result of isolation and/or adaptation to different environmental conditions. These distinct populations may differ substantially in theirsocio-economic value and production potential, or offer possibilities for improvement by combination of their qualities through breeding programmes. The design of in situ conservation strategies is therefore particularly dependent on estimating the likelihood and location of such intra-specific variation. Another important aspect of the genetic system of individual species in determining patterns and levels of diversity within provenances and populations is the extent to which pollination takes place between different individual trees (outbreeding) or the frequency of self-pollination, or even the production of seed without fertilisation (apomixis). Although very few species have been adequately studied many scientists who have undertaken such investigations have concluded that there is a strong tendency for cross- pollination and outbreeding among tropical tree species (Bawa 1974; Bullock 1985; Bawa et al 1985; Bawa and Krugman 1990; Janzen and Vasquez-Yanes 1990). Some authorities believe that the high gene flow between individuals resulting from the predominance of cross-pollination contributes to the high levels of genetic diversity characteristic of the tropical forests. Others have attributed high levels of speciation in tropical rainforests to the isolation of the individual trees of a species, with consequent reliance on inbreeding or apomixis (for a discussion of these views see Whitmore 1990 and Bawa et al 1990). The practical implications are that without specific safeguards, based on adequate knowledge of the genetic systems and variation patterns of important tree species, the effects of logging or extensive deforestation on the population, through alterations in the levels and patterns of outcrossing or inbreeding, are likely to increase with increased severity of disruption of the populations. As a result there could be a lack of adequate fertile seed production, or excessive inbreeding, which could endanger the viability of the species or the provenance in the longer term. The effects might not necessarily be negative, since in some circumstances the removal of closely related individuals could encourage wider outcrossing. However the essential aspect is that the effects will be unpredictable in the absence of adequate information on the species' population biology. The absence of such information in respect of almost all tree species in most areas of tropical forest must severely limit the possibilities for effective management of their genetic resources. The safest practicable conservation strategy must be to maintain the broadest possible genetic base in the species as a whole by conserving as wide as possible a range of provenances over the natural geographical and ecological range. In doing so it is also desirable to protect the integrity of each such population against genetic pollution which can be caused by the introduction of outside provenances of the species, such as might occur for example in the course of gap planting or enrichment planting within the forest, or following the establishment of plantations using outside, hybridising species or provenances in an adjacent area within the range of natural pollen distribution. 2.5 Values of Genetic Diversity The relative value of different individual trees or populations of a species is very difficult to judge in natural stands, where uncertainties over age and past history are compounded by accidental differences in site quality and competition with other vegetation. For the few tropical tree species which have been the subject of provenance trials and breeding programmes highly significant increases in productivity and related economic and social benefits have been achieved, based on selection from the very diverse provenances sampled. These gains have been achieved through careful comparative trials and the application of the results in fast-growing plantations. The genetic resources of other tropical tree species to be conserved in situ in the natural forests might provide the basis for similar increased productivity if subjected to trials ex situ to determine their heritable qualities and performance in given environments. 2.6 Use Values and Option Values Economists recognise two main types of value: use values and non-use (existence) values. Use values may be further sub-divided into those available for known and immediate uses and those which might become available in the future (option values). While the direct use value of the genetic diversity in the forests can best be measured in respect of the few most marketable species the option value of species not currently in demand may be quite high. The latter could become important to adapt to possible climatic changes, for example, or to meet changed market demands for timber or non- timber forest products (NTFP) . There have been many examples in recent years of species earlier regarded as unmarketable or as "weed" species, but which are now highly valued. The effect is to increase the options open to the forest manager, as a result of having conserved a wider range of genetic diversity in the natural forests than those species previously known to be of immediate use (see the Ghana case study in Part II). 2.7 Precautionary Values Among the important indirect (non-consumptive) use values of the genetic diversity of tropical forests is its possible contribution to the stability of the ecosystem particularly in the face of climatic changes at global or regional levels. Virtually nothing is known of the functional role of overall diversity in ecosystem stability, or of the acceptable levels of species loss or the thresholds of irreversible change with decreasing levels of genetic diversity in natural forests. Similarly it is impossible to forecast with certainty the response of tree species to the likely climatic changes. These high levels of uncertainty, combined with the dangers of irreversibility in the reduction of genetic diversity place emphasis on following the precautionary principle of avoiding unnecessary loss of such diversity and of genetic resources. The value of conserving the genetic resources of a range of populations of a species of established socio-economic importance may also be greater in the context of expected climatic changes. Particular interest attaches to the edge of a species 1 natural range, where the local populations may be adapted to more extreme environmental stresses, for example in the transition zones between moist and dry forest types, or between dry savanna woodland and more arid thorn scrub or desert formations. Moreover recent developments in molecular genetics and genetic engineering may give additional value to such populations which, although they may exhibit very slow growth and poor stem form or other production limitations, may contain valuable genetic material conferring characteristics such as resistance to drought, or to high salt levels in the soil. Similar considerations apply to populations of species in other transition zones, for example, in the inter-tidal zone of coastal forest formations, where the effects of sea-level rise associated with global warming could be severe. 10 2.8 Existence Value The final category of value - existence value - is likely to be relatively high for the rarer and more precious tropical tree species most vulnerable to genetic impoverishment, or even to extinction, through intensive harvesting without adequate management, and without adequate attention to the conservation of their genetic resources. It is such existence values, coupled with concern for the existence of tropical forest ecosystems as a whole, that are uppermost in the minds of the public and the media in many developed countries, particularly in regard to tropical rain forests. While there may be future benefits from the conservation of a wide spectrum of genetic diversity, including some of possible indirect value in ecosystem stability, and others of uncertain option value, in respect of as yet unidentified market opportunities, there are likely to be immediate financial costs. These may be both direct costs of protection and management, and also indirect costs of opportunities foregone in the immediate term through lower levels of production from the forest as a whole. In the short term the stability of the forest ecosystem is unlikely to be affected by a reduction in species diversity. Although this could lead to the progressive elimination of e.g. some slower-growing cabinet or joinery timber species, especially if accompanied by more intensive harvesting systems that tend to favour fast-growing pioneer species, the overall levels of total wood production may also be largely unaffected by loss of species diversity. 2.9 Location of conservation areas The greatest diversity of tree species is found in the lowland humid tropics, for example at Yanamomo, in the Peruvian Amazon, where 283 species of trees of 0.1 m in diameter or larger were recorded on a one hectare plot (Whitmore 1990). Historical and environmental influences have led to the concentration of species diversity in certain areas, as a result of evolutionary pressures, related for example to long periods of environmental stability or to periodic disturbance, isolation, the removal of barriers between populations, migration and other influences. Although the level of information on species numbers and distribution patterns is still very sparse biogeographic divisions between the world's major terrestrial ecosystems may be used as an initial stratification, and with other available data provide a basis for selecting areas of high levels of diversity, or of exceptional degrees of endemism, or both together, as high priorities for ecosystem conservation. These two criteria, together with the degree of depletion or of threat to the genetic resources of the area, have been used by various authors and organisations to single out countries and locations to be accorded high priority (e.g. Myers 1988; Reid and Miller 1989; McNeely 1990). Such areas are most appropriate for conservation within fully Protected Area systems, such as Nature Reserves or National Parks, rather than areas managed for the production of timber or other products. The criteria suggested above may be the most appropriate to select a single location of high priority for ecosystem conservation but if more than one area can be selected as part of an integrated system of reserves the criteria should be extended to include a diversity of sites to capture also intra-specific variation of target species. Moreover the association of adjacent areas of natural forest under productive management as "buffer zone" to the fully Protected Area may provide a valuable extension to the range and size of populations of many tree species, and to the efficient coverage of their intra-specific variation. 11 2.10 The Link to Production Forests The role of managed production forests in genetic resource conservation is particularly important for the conservation of intra-specif ic variation at the population level, of tree species of known or probable value. Ideally the location of areas serving such conservation objectives should be determined from data on patterns of variation, or of genetic structure and gene flow, since the latter should reveal the geographic scale over which populations may diverge from each other. In the absence of such information, some reasonable assumptions can be made from geographical and ecological data. Differentiation between populations may develop in response to selection pressures resulting from local environmental conditions. There is thus generally a correlation between geographic/ecological factors on the one hand and inherent morphological or physiological properties of the local population of a species on the other (FAO 1989a). While the fully Protected Area system of Nature Reserves, for example, may include some part of the range of a species the effective conservation of the gene pool (i.e. the total sum of genetic materials) of the species as a whole requires the inclusion of a much wider range of populations representative of possible genetic differences which can only be surmised from their geographical or ecological situations (Frankel 1970). This is likely to require a number of conservation areas distributed over the entire natural range of the species and to the extent that such a strategy might be practicable in any given case it is likely that most of these would have to serve multiple objectives, including production of timber as well as other forest products. The desirable number and location of such conservation areas must be determined for each species and in the absence of specific data it may be assumed that for widespread and strongly outcrossing species a few locations in each major ecological/geographical zone might be adequate. For strongly inbreeding species, and those exhibiting scattered and isolated occurrences, more conservation sites might be needed. 2.11 Size of Conservation Areas Much thought has been given, particularly in respect of populations of large animals, to the minimum size of a viable population that would allow scope for continued undirected evolution, and related to that the minimum conservation area needed. Other calculations have been made based on known or probable levels of inbreeding and the size of population needed to minimise consequent loss of genetic variability, over a given number of generations. A commonly quoted figure is a minimum of 50 breeding adult individuals for short-term maintenance of fitness in the population, and 500 to sustain long- term genetic adaptability to change (FAO 1989a). A figure of 1 000 indivi- duals has been suggested as desirable to maintain "evolutionary potential", while at the other extreme it has been estimated that a genetically effective size of less than 50 may be adequate for several generations and that nearly all the genetic variability of a population can be conserved temporarily in only a few breeding individuals (Wilcox 1990). For long-lived species such as trees and those under some degree of management over much of their natural range, namely those of recognised economic or local value, individual population size is less important than the distribution of conservation sites to sample likely patterns of diversity across the species 1 range. In the absence of information on patterns of variation within a species the conservation of populations of a few hundred individuals at the extremes of the geographical and ecological ranges is the 12 most practical option. However for highly diverse rainforest containing many hundreds of tree species each of which may exist normally at very low freguency 9 conservation areas of 5 000 ha have been suggested as a "rule of the thumb' 1 , based on estimates that this would cover 95% of the species (Ashton 1984). 2-12 Dynamic conservation An essential feature of in situ conservation is that it provides for "continuing evolution" (Frankel 1981). In this sense both the genetic resources themselves and the practice of their conservation are essentially dynamic and should not be seen as an attempt to preserve a fixed and finite resource. At the same time conservation implies the avoidance of the rapid erosion of genetic variability, for example through the extinction of species or unique populations, or strong directional change in the genetic composition of a population as a result of severe reduction in numbers, causing increased inbreeding and/or genetic drift, as a consequence of isolation. As the possibilities for setting aside substantial, additional fully Protected Areas for conservation are progressively reduced, the necessity for the more deliberate management of socio-economically important species must continue to increase. 2.13 Disturbance and Succession Changes in species composition by the ecological succession of different plant communities on the same area are a common feature in any type of vegetation, as the colonising pioneer species and communities give way to later stages. Local extinctions of pioneer species are a necessary part of this process but provided that the natural or artificial disturbances in the climax formations leave opportunities for cyclical recolonisation by pioneer species in other areas the overall composition of the forest over large areas is not affected. Such natural processes in forest dynamics are generally accepted to occur in the most biologically diverse formations, notably in the tropical rain forests, where the most species-rich areas are likely to be those which include patches of secondary forest in various stages of recovery following disturbance, as well as patches of mature-phase forest (Whitmore 1990). The goal for genetic resource management is thus to maintain a dynamic system (Namkoong 1986), which may entail the deliberate removal of trees in the later stages of natural succession, as well as the deliberate conservation of some elements of the mature-phase forest. Depending on the exact systems of management, and the degree of understanding of the forest dynamics on which they are based, genetic diversity and specific genetic resources may be enhanced or reduced in specific areas of forest, over specific periods of time. The lack of active management, for example through the complete exclusion of human intervention, may tend to reduce genetic diversity within any given area, although in certain circumstances such (usually temporary) non-intervention may be a conscious management decision aimed at conserving specific genetic resources within the framework of an overall conservation strategy. 13 2.14 Logging and Genetic Diversity Selective logging in mixed tropical forests might, in theory, be managed to maintain an optimal balance between the various stages of ecological succession to allow for maximum genetic diversity and the conservation of the genetic resources of both pioneer and late succession species. This might be achieved either by clear-cutting, at very long intervals, to allow each felled area in turn to revert eventually to the mature condition, or by careful opening of small gaps by the removal of individual trees, or various possible intermediate patterns and levels of harvesting that may favour the advanced regeneration of different species. However this presupposes not only the willingness to subordinate short-term financial gain to long-term ecological objectives but also a degree of understanding of forest composition and dynamics. In addition to the immediate effect of logging on the advanced regeneration and environmental conditions, such as light, temperature and humidity on the forest floor, all of which may affect the regeneration of different species in different ways, there will be further effects on the density and spacing of populations of the species felled. These changes may influence flowering and fruiting patterns and the mating relationships within the population. In addition to such direct effects there may be impacts on the populations of pollen vectors or seed dispersers of tree species through the removal of "keystone" plant species on which these populations depend. One often inadvertent and usually severe effect of human intervention which may follow logging is increased susceptibility to fire. While some forest formations are adapted to survive periodic burning, and may be fire disclimax communities with particular qualities of genetic resources of potential value associated with their capabilities to colonise fire-prone areas, the adverse impact of indiscriminate fire in other more complex forest formations can severely reduce the genetic resources of the more valuable tree species. In extreme cases whole populations may be lost through fire following the felling of all adult trees of a species in the area concerned. Nevertheless with adequate control and based on a sufficient understanding of the ecological processes involved logging and timber extraction can be used to assist the conservation of a wide spectrum of genetic resources of the principal tree species. Both the efficiency with which this can be achieved, and the security against accidental loss of substantial elements of the gene pool, will be dependent on the management of a network of conservation sites in both the production forests and the fully Protected Area systems, extending across the natural range of the principal species. 14 Box1 The role of logged forests In the conservation of species richness and genetic diversity Most remaining unlogged areas of forest outside the limited extent of Protected Area systems will be exploited for timber within the next few decades. The logged forests, except where excessive machine operations, fire or illegal cultivation have caused intensive damage, still retain much of the original plant diversity and are often apt for recolonisation by the major fauna (Johns 1988 and 1992; Whitmore and Sayer 1992). The effect of the initial logging on the forest structure, composition and regeneration have been well documented and the procedures to minimize damage and improve cost-effectiveness fully described (e.g. Nicholson 1979; Dykstra and Heinrich 1992). Ideally harvesting operations in production forests aim to mimic processes of natural gap formation in order to provide a sustainable yield of timber without radically altering the composition and structure of the forest as a whole. Retention of unlogged areas amounting to only 5% of an intensively logged forest may be adequate to conserve populations of vertebrate species regarded as intolerant of logging (Johns 1992), and a single cycle of togging need not reduce species richness among the tree populations (Whitmore and Sayer 1992) provided adequate regrowth is present and not severely damaged during logging, or seed for subsequent regeneration is available in the soil seed bank or adjacent areas. Major impacts on both species richness and the genetic resources of the prime timber species, especially those more shade-tolerant and characteristic of mature-phase forest, are likely to be associated with the second cycle of logging and subsequent harvesting. The length of the cutting cycle will be critically important to the maintenance of satisfactory breeding populations of the principal tree species. Even when small areas, such as Virgin Jungle Reserves may be left unlogged the progressive effects of forest fragmentation, and isolation, on gene flow, inbreeding rates, genetic variation, seed fertility etc. will decisively affect the genetic resources. There may be impacts on key pollinators or seed dispensers, as well as on soil structure and even soil chemistry (House and Morrtz 1991). Coordinated long-term studies are needed to guide the management of logged forest for the conservation of species richness and genetic resources, including where necessary remedial management (Ng 1983). 15 CHAPTER III IMPACTS OF MANAGEMENT IN PRODUCTION FORESTS Although forest management can be defined in various ways (Vanniere 1975; Philip 1986; FAO 1989b; FAO 1992a), there is general agreement that it should consist essentially of taking firm decisions about the future of a particular area of forest, planning and implementing action to achieve those objectives and monitoring the results. It depends ultimately on the national forest policy and its basic components are the definition of objectives, planning, control, protection and records. It is also concerned with the allocation of resources to meet the defined objectives and it is commonly the wide disparity between the objectives and the resources provided to meet them that causes the greatest impacts on the forest. As used here it is taken to include both the controlled harvesting of timber and associated silvicultural operations aimed at the sustainable provision of specified goods and services from the forest. An essential aspect is that the land should remain in forest use after harvesting and that the operations should be conducive to adequate regeneration and to the maintenance of the environmental and social functions of the natural ecosystem. This implies involvement and prior consent by all land users, control and continuity of purpose, which are also essential elements for the conservation of the forest's genetic resources. 3. 1 Continuity and Control Genetic resource management can only be effective if it is an integral part of land use management as a whole (FAO 1989a). The objectives of in situ conservation may involve various approaches to land use, including multiple use production areas and timber production forests as well as fully Protected Area systems, in the attempt to reconcile the dual requirements of present day demands for revenue and basic human needs (food, wood and other forest products), with long-term conservation objectives. Ideally this requires a comprehensive national land-use policy and plan, embracing an appropriate forest policy and based on a national inventory of the forests, including specific attention to plant species of socio-economic importance and conservation concern. Failing that a broad zonation of vegetation, following the principles of the Biosphere Reserve Programme (Unesco 1984), may be used to focus the conservation objectives to the most important areas, which may then be given special reserved status, within fully-protected categories of land. In reality the pressures of increasing human populations and related economic development programmes on the land and natural resources severely restrict the areas likely to be set aside and retained as National Parks or equivalent, fully protected reserves. The potential contribution of production forests to the conservation of genetic resources is therefore important and may be high, given a conservative and sustained approach to forest management, based on natural regeneration systems. The deliberate inclusion of genetic conservation objectives in the management plans of production forests may be essential, to secure the necessary range and diversity of sites for an efficient national conservation network. The initiation of tropical forest management, first in the Indian sub- continent, and subsequently in Africa (FAO 1989b; FAO 1989c) included the establishment of firm administrative control over the forests, for example with the designation and demarcation of a "Permanent Forest Estate 11 whose boundaries and use could be changed only by decision at the highest levels of 16 national authority. Although this authoritarian approach has sometimes provoked local opposition, and there is now general acceptance that the security of the forest cannot be assured without the consent and involvement of local people dependent on the resource, the establishment of the Permanent Forest Estate has been a powerful element in the conservation of genetic resources. Although in many cases the objectives of sustainable timber production have not been sufficient to ensure the retention of all of the reserved forest estate from reallocation to permanent agriculture, or from illegal encroachment, in most tropical countries legal protection for productive forests has had an important and positive impact on the conservation of their genetic resources. However there have also been severe negative impacts from the often excessive exploitation (unaccompanied by adequate control and management to ensure regeneration) resulting from short- term political, financial and economic pressures. 3.2 Economic and market influences Most recent reviews of tropical forest management (Masson 1983; Mergen and Vincent 1987; Schmidt 1987; Wyatt-Smith 1987a; FAO 1989b; FAO 1989c;FAO 1992a; Poore 1989) conclude that management of the tropical forest as a renewable and truly sustainable resource is technically possible and that the past failures or abandonment of attempts at the sustainable production of timber in tropical forests have been due to socio-economic or political pressures. These have frequently imposed severe constraints on the resources available for the various aspects of management referred to above, resulting in part from the failure to recover adequate revenue from the exploitation of the resource, and/or failure to reinvest such revenue in the regeneration and management of the forest (Repetto and Gillis 1988). At the root of the problem is the weakness of the national economy in most tropical countries combined with the scarcity of capital to invest in development. The forest itself has been treated as a source of finance to support development in other sectors of the economy and while in total this source of funding has been sufficiently large to make exploitation worthwhile, the apparent returns per unit area of forest are commonly too low to secure the necessary level of investment for sustainable management, or even to retain the land against pressures for alternative land use. Even when the arguments for the long term economic benefits from reinvestment in forest management may be accepted urgent short-term financial constraints, coupled with short-term interests of concessionaires, logging enterprises and other concerns, may lead to excessive exploitation without regard to forest regeneration. This is encouraged by short-term concession agreements which provide no motive for longer-term planning. In order to recover the investment in large-scale machinery and road construction the logging companies frequently seek maximum returns from harvesting at minimum costs, and with minimum concern for environmental impacts. Where market demand is highly selective the exclusive concentration on the extraction of the best phenotypes of the most valuable species will have negative (dysgenic) effects on subsequent generations, if such interventions are carried out without due consideration to regeneration potential and the quality of the next generation crop. Where a broader range of species, including many lesser-known or lesser- used species, are marketable there may be potential for greater revenue, providing the opportunity for increased reinvestment in forest management. In practice, however, although the more intensive felling has tended to 17 increase profits, reinvestment has remained static or even been reduced; the forest has in these cases been left to recover naturally without regard to species composition. If logging damage to the advance regeneration and to the site conditions for seedling establishment were closely controlled the harvesting of a wide range of timbers might be more compatible with the conservation of a similarly broad range of species, than the very selective logging characteristic of earlier exploitation regimes (see for example the Ghana case study Part II). However without strict controls over road construction, logging plans, timber marking, harvesting and extraction there must be a danger of severe ecological impacts on the site capability, and on the genetic resources especially of the slower-growing species characteristic of the mature-phase "climax" forest. The introduction of heavy mechanical equipment for timber exploitation, together with increased demands for a wider range of tropical timber species, has had an overwhelming impact on the species and genetic diversity of some natural forests. The effect is to shift the composition of the forests in the felled areas towards relatively few predominantly fast-growing pioneer species characteristic of the earlier stages of the ecological succession. Management for the in situ conservation of genetic resources must give priority to the principal economic species, some of which at least will be fast-growing pioneers or gap-opportunist species, which will be favoured by a certain degree of canopy opening. However the effect of large-scale and extreme reduction of the forest canopy which is associated with heavy mechanical logging of a wide range of species is most likely to favour very short-lived pioneers with timber low in density and durability. It is certain to discriminate, at least in the short term, against the multitude of slower- growing shade-tolerant species, including some high quality cabinet woods and veneer timbers. The overall impact of these interventions is therefore to reduce the range of species of known economic value, uniformly in favour of a narrower band of fast-growing, low to medium density species. At the same time there is likely to be a reduction in the biological diversity of the ecosystem as a whole, both flora and fauna, unless certain areas of the forest are excluded from simultaneous, intensive logging regime. Economic pressures to maximise the returns from the investment in logging equipment, roads and related infrastructure have frequently overridden such ecological considerations . Volume yields in naturally regenerated tropical moist forest are likely to be on average only 2 to 3 m 3 per hectare annually, although in some cases silvicultural operations and rational management might increase this by a factor of three or four times, where there is ample regeneration of marketable species and where site conditions are favourable (Wyatt-Smith 1987a). A common feature of management systems aiming to increase the yield of valuable species in this way is the elimination, usually by poisoning, of unwanted and competing trees. This and other "refining" operations, if maintained consistently through successive cutting cycles, will also reduce the diversity of species in the areas treated but need not do so over the forest as a whole, if selected areas were excluded from such treatments or were treated to favour different components. In most existing operations the loss of yield involved in such exclusion would be regarded as unacceptable on purely economic grounds . The impact of the international trade in tropical timber on the genetic resources of particular genera and species has been assessed in a recent study undertaken on behalf of the International Tropical Timber Organisation (ITTO) 18 by the World Conservation Monitoring Centre (WCMC 1991). This study compiled data and expert observations on 1 868 tree species, and assigned them to conservation categories in accordance with the IUCN system. Of these 304 species were classified as threatened at a global level and 190 as threatened in two or more countries. The reliability of the data may be questionable since there is little available information based on detailed forest inventory and the assessments are therefore based largely on the opinions of specialists in the various geographical and botanical fields. Particular attention was given to the family Dipterocarpaceae, for which the study claimed to have achieved a relatively complete assessment of the conservation status. Even allowing for some uncertainty and possible exaggeration of the degree of depletion or threat the study revealed a serious impact on the genetic resources of this important family of high quality timbers. The very intensive logging now being carried out in several of the countries in which species of this genus occur naturally, must give cause for concern unless urgent attention is given to their sustainable management within which harvesting is only one step; and to genetic conservation measures, carried out in parallel with their use. The ITTO study covered countries in Africa and Asia in 1991/92. In Latin America another aspect of the market impact on forest genetic resources has been identified where the demand for mahogany (Swietenia macrophylla) has led to highly selective harvesting of good phenotypes of the species over large areas without provision for subsequent protection of the regeneration and management to ensure its diversity and quality (Monbiot 1991). The danger of severe dysgenic effects and extinction of local populations is clearly a possible consequence of the market forces in this case. In general the effect of economic and market forces has been to impose the reduction of species and genetic diversity in the timber-producing forests. This has been the result of very narrow, selective markets, and failure to re-invest part of the revenue in forest management, thereby enforcing low-cost "blanket" management systems with short-term objectives. The inadequacies of economic theory and analysis relating to natural management systems in tropical mixed forests (Leslie 1987) relate particularly to the need to widen the scope of the analysis beyond the short-term benefits to the agencies and individuals most directly involved, to include the broader interests of the nation and society as a whole, and of future generations. This applies particularly to the values attached to the genetic resources of the forests. To the extent that the timber production areas can also be managed to conserve genetic resources this additional objective should reinforce the economic case for natural forest management to be governed by ecological as well as economic considerations. Even the much narrower range of overall genetic diversity that will result from repeated extensive harvesting and refinements of the forests is likely to be much greater than if the same area were converted into forest plantations, and certainly substantially greater than that which would result from most alternative forms of land-use. However, many species and populations, especially those characteristic of the primary or mature-phase forests, will be lost unless the current severe constraints on the funding of natural forest management are reduced. Increased funding (including the re- channelling of revenue back into the forest) would allow more deliberate and diverse management of the tree populations, possibly at the expense of some loss in harvested volume. This loss might be partly offset by an increase in unit value from the management of selected areas on longer rotations for higher quality timber. 19 Whereas rethinking of the economic and financial constraints on natural forest management seemed unrealistic earlier, and may still be so for most of the production forests, the economic value of genetic diversity is now increasingly recognised as of international concern. That value, and therefore the costs that might be borne for conservation, will be dependent on the location, composition and existing condition of each specific area of forest. 3.3 Forest Inventory The scientific basis for the conservation of species and of their genetic resources depends essentially on the study and interpretation of taxonomic information on genetically determined differences and affinities, their patterns of natural distribution (chorology) and the ecological basis for their occurrence. These three interdependent sets of data which should form the basis for drawing up sound conservation strategies, are inadequate in the tropics and in many cases non-existent. Very often the only data available are the result of forest inventories concerned primarily or exclusively with stocks of harvestable timber. The techniques for efficient surveys of the standing timber resources in the forest have shown considerable progress in the past 30 to 40 years (FAO 1989b) and there has been increasing emphasis on the need for National Forest Inventories and for detailed and more comprehensive inventories of selected areas. However, all too often such inventories have provided little of the information needed to plan the long term, sustainable management of the forest, but have been confined to the determination of the marketable volume of a limited number of so-called economic or potentially economic species (Masson 1983). Little regard has been paid up to now to the determination of the actual composition of the forest or its condition after logging. Inadequate information on the species composition of the forest is one of the main problems confronting natural forest management, both in terms of the forest's full economic value and its regeneration potential (Wyatt-Smith 1987b). Some of the important questions to be answered in this regard are the adequacy of existing levels of seedlings, saplings and advanced growth of the marketable species as the basis for the future crop, both before and more importantly after logging. Assessment of regeneration was an integral component of the earliest attempts at natural forest management, for example in Malaysia and Nigeria, but latterly the tendency has been to reduce rather than to increase the time and manpower resources devoted to such surveys, and thereby to limit the level of detailed information available even in respect of the preferred economic species. However in recent years, with the development of conservation biology as an applied science, it has become increasingly clear that more detailed studies and surveys must be an inseparable component of forest management systems to produce an adequate scientific basis for conservation action. Diagnostic sampling to determine the stocking and silvicultural conditions of young stems of "desirable" species below the exploitable girth limit is a common practice in the more deliberate attempts at natural forest management (FAO 1989b) and there is a tendency for inventory and resource assessment in advance of logging to be taken further, to extend beyond commercial timber resources to non-timber products, including those of interest to local communities (FAO 1989c). Such action is essential to obtain a more complete valuation of the forest in support of the conservation of a wider spectrum of valuable genetic resources (see also the Ghana case study) . 20 The central and traditional component of forest inventory, namely the accurate assessment of the allowable cut and sustainable yield of timber, is vital to conservation of the resource, to ensure that the rate of harvest should not exceed the regenerative and growth capacity of the forest. This requires an accurate knowledge of the growing stock, its distribution by species, size classes and location, and an understanding of how these change with logging and silvicultural treatments. Availability of accurate forest inventory and stock maps in reducing logging damage and protecting the regeneration is also essential in the conservation of genetic resources. However such inventory data are insufficient as a basis for more positive action to identify and conserve the most valuable components of genetic diversity. Given the high proportion of the cost of inventory operations that relates to the actual access to and work in the forest the additional cost of collecting a wider range of data and observations in the course of the timber survey can be relatively low. Moreover the widespread availability of powerful electronic computing facilities for the handling of large and complex sets of data has largely removed that constraint on the collection of an increased range and volume of field data, in the course of forest inventory. In general 10% of the tree species in a tropical forest comprise at least 50% of the stand (Ashton 1988). Rapid extensive surveys may give guidance to conservation priorities not only in terms of species richness but also in respect of the genetic resources of a forest. Recent studies by Hawthorn in Ghana indicate the potential value of broad botanical surveys in association with forest inventory in developing strategies for the conservation of forest genetic resources (see also the Ghana case study) . For the forest inventory to serve its full purpose in assisting the conservation of forest genetic resources it must attempt to assess the relative genetic value of a given area of production forest, for example in regard to the range of distribution of selected species or forest types, in relation to other managed or reserved areas, including the fully Protected Area system of National Parks etc. This information could help determine the most efficient combination of the minimum number of sites needed to cover species, populations and communities at the minimum level essential for conservation of the desirable range of diversity. Given the very large numbers even of tree species in the tropical forests, and the need to include surveys of some other socio-economically important plant species, or of species essential to the functioning of the ecosystem, and therefore to the overall management of the resource, it is often necessary to adopt a compromise between detailed biological surveys and more general assessments, based on variations in landscape or other environmental features. It may be assumed, for example, that much intra-specific (provenance) variation, of potential economic significance, follows patterns of variation in the environment and in the plant community as a whole. A key to efficient and cost-effective inventory is at the planning stage, to ensure inclusion of an appropriate range of botanical, ecological and sociological expertise, for example, both in the design and in the execution of the survey. Given the range of expertise and strength of interest in many universities and research institutes in both tropical and industrialised (often aid donor) countries in such scientific investigations in tropical forests, the additional human resources needed may often be available at comparatively little extra cost, compared with the expenditure on the basic inventory, and in relation to the value of the additional information to be derived from such expert involvement. 21 A particular aspect of this planning phase which may require expert involvement is the planning of data capture, handling and analysis. The widespread availability of small but powerful computing capabilities for data analysis has also transformed the possibilities for obtaining an understanding of forest composition and genetic diversity from limited data. Approaches to the development of inventory procedures and growth models for the management of tropical forests are advancing rapidly (e.g. Vanclay 1989; Alder 1990) and will progressively gain the capacity to incorporate broader information relating to the management of the genetic resources of the forest. Simulation of the variability and complexities of population distributions is already possible using some stochastic models such as those designed to measure the average number of occurrences in a given area, and to reveal patterns of their variability (Jeffers 1982). This information can be an important aid to selecting locations for in situ conservation. Sustainability as an objective of forest management demands that forest inventories be planned to look far beyond the assessment of marketable volume to the establishment of baseline data for the continuous monitoring of forest condition, and the implications for the conservation of genetic resources. Among these are the retention of sufficient individuals of "keystone" species to maintain their own breeding populations and therefore the long-term contribution of that species to the functioning of the natural forest ecosystem. This requires the identification and recording of such species in pre-logging inventories, so that sufficient numbers, with an appropriate distribution in the forest, are marked for retention, based on a sufficient understanding of the forest dynamics. 3.4 Forest Dynamics i The aspect of forest dynamics which has attracted most attention in relation to timber exploitation is the growth and yield of the principal marketable species. The lack of reliable annual growth rings in the timber of most tropical tree species, and therefore of convenient "short cuts" to the determination of increment, together with the complexity of the growing conditions and species composition in the forests, has made predictions of yield difficult (e.g. Kemp and Lowe 1970). Despite the progressive development of systems of Permanent Sample Plots (continuous forest inventory plots) in representative samples of forest in a number of countries the basis for calculations of growth and yield is still generally weak. Probably the best level of prediction has been achieved in Queensland, Australia (Vanclay 1989). However even in those intensively studied forests there is very little if any reliable information on the relative importance of genetic and environmental factors in determining growth rates of individual trees in the natural stands. The influence of competition from other vegetation on the relatively small proportion of "desirable" stems has been the basis of management interventions to effect improvement thinnings to favour the growth of the potential final crop trees (Hutchinson 1987; Maitre 1991; FAO 1989c). The first step in this treatment, as applied in Sarawak (Hutchinson 1987) is to group the species into "wood quality" categories, to arrive at lists of desirable species, and then to assess and attempt to predict their response to different levels of release from overhead shade and competition resulting from commercial logging or subsequent thinning of the stand. For this purpose the species are allocated to ecological groups according to their degree of shade tolerance and/or capability to increase growth rapidly in response to release. 22 Classical forest management started from attempts to understand and utilise such ecological processes and interactions f and particularly the natural cyclical changes in the progression from gap colonisation by pioneer species to the mature or late successional forest condition. Arbitrary divisions into gap-phase, building-phase and mature-phase forest development are now commonly recognised (Whitmore 1990). The concept of "forest gap-phase dynamics" appears to apply as a general model, possibly with local variants, in a wide range of forests in all tropical regions and although it is an oversimplification to recognise only "pioneer 11 or "climax" species the definition of two broad groups or "guilds" based on their behaviour in response to gap creation has proved useful in approaches to natural forest management (Whitmore 1991). It is also important in approaches to in situ conservation of forest genetic resources, particularly those of the principal economic tree species. In all tropical rainforest floras there are fewer pioneer than climax species (Whitmore 1990) and the pioneer genera are less likely to include narrowly endemic member species of very restricted geographic range. Their efficient systems of seed dispersal are probably responsible for their generally wide range of distribution and reduce the possibilities for the development of localised genetically distinct populations. The complex interrelationships in pollination, seed dispersal, food web and plant web systems described earlier are important both for the attempted regeneration and management of timber species, particularly those in the "climax" guilds, and for in situ conservation of forest genetic resources. Without adequate understanding of the ecological processes and relationships the attempts at natural forest management must rely on "blanket" application of silvicultural treatments with unpredictable and possibly perverse results. However it has latterly proved impossible to justify the necessary expenditure on detailed ecological studies in terms of the increased yield of timber alone over the period of one or two cutting cycles or short rotations. Nevertheless recent reviews of management methods have emphasised the inadequacy of "blanket" canopy treatments and the necessity for improved knowledge of the autecology of individual species, and not only those with marketable timbers (FAO 1989b). The field of reproductive biology, embracing studies of pollination, seed dispersal and predation, and the dynamics of regeneration banks of seed, seedlings and saplings, which are clearly important in genetic resource conservation, have also been identified as necessary to the forest manager (Palmer 1989). The case study of Cordia alliodora in Appendix 1 shows the level and intensity of research needed to provide reliable information for a single species. The number of species requiring such study even in a single forest, is very large, and the resources of skilled manpower are very limited. Seed dispersal biology is directly relevant to natural forest management, particularly in respect of the mature phase species, many of which have high value timbers, and are characterised by the production of large seeds. There is a growing body of evidence that the frequency and density of such species in the natural forests is limited by seed predation and that chances of survival are improved by dispersal away from the vicinity of the parent tree (Terborgh 1990). A good example is the dependence of Virola surinamensis on distribution by toucans and other large frugivorous birds, without which it would be unable to recruit seedlings, and the species would face local extinction (Howe 1990). An understanding of the identity and 23 behaviour of animal dispersers, and their possible dependence on other tree species for food or nest sites, is an area of common concern to both forest managers and forest geneticists concerned with conservation. 3.5 Regeneration The regeneration potential of the desirable species is of key importance in the management of the productive forest estate and clearly fundamental also to the conservation of their genetic resources. For in situ conservation natural regeneration is the clearly preferable strategy, although some aspects of artificial regeneration, such as enrichment planting using seed or seedlings randomly obtained from the same natural stand, may sometimes be acceptable and desirable. The encouragement of natural regeneration is likely to be the cheapest option for timber production objectives also, provided that it can be easily and confidently obtained. In practice, however, it has proved to be one of the most difficult and uncertain aspects of management in tropical forests, despite having been the subject of much field study and experimentation for over a century. The problem is not peculiar to the tropical forests, but is evident from the large areas of previously forested and now barren land in the Mediterranean and some temperate regions. Grazing and browsing by domestic livestock have had a major influence in such areas, and in the more densely populated tropical forest regions, for example in India, even in tropical forests of high production potential, such as the Sal (Shorea robusta) and evergreen forests (FAO 1989c) despite the long history of research and study by trained staff. Similar failures have been common elsewhere in Asia, except in some of the richer dipterocarp forests, where as a result of the range of desirable species whose silvicultural characteristics are favourable to promotion by controlled canopy opening, satisfactory, although by no means predictable, natural regeneration has been more often achieved. Failure to ensure adequate regeneration of the selected economic timber species proved a fundamental problem in many African forests (Nwoboshi 1987; Kio and Ekwebelan 1987; FAO 1989b) and was a principal reason for the development of forest plantations in African countries, as an alternative to natural forest management . Among the most important aspects of regeneration behaviour are the frequency of seed years, the length of seed viability, seedling survival, the pattern of distribution and abundance of seedling regeneration on the forest floor, the tolerance of shade and response to light and the competitive potential of the preferred species under different degrees of canopy opening. The complexity of the interactions in the natural forest is such that the provision of a set of environmental conditions guaranteed to favour the regeneration of a few selected species is virtually impossible, without detailed study of the actual state of each limited area of more or less uniform conditions within the diverse mosaic of the forest, and of the autecology of all the principal species. In practice diagnostic sampling often by means of systematic parallel transects through the forest, is used to determine the overall adequacy of stocking of young stems of desirable species, which is found to survive undamaged after logging is complete, as the basis for the next crop. A diversity of sites and silvicultural systems will be likely to create increased diversity, thus being an advantage from the genetic resources point of view. 24 A shift in approach to the management of the regeneration stage, from earlier attempts to control the composition and development of the seedling population to the acceptance of the largely accidental composition actually obtained, is reflected in the procedures for regeneration sampling. For example in the earlier development of the Malayan Uniform System (MUS) , which was among the first and most successful attempts at management in the tropical moist forests, a three-stage linear sampling regime was used to determine different stages of regeneration development before logging was undertaken. This procedure was later modified by merging the three stages into a single linear regeneration sampling operation after logging was completed (Wyatt- Smith 1987b). The change represented the acceptance that economic imperatives rather than silvicultural considerations, are decisive in determining management practices. The MUS as originally designed made provision for situations where the level of seedling regeneration was found to be inadequate, by proposing deferment of logging until adequate regeneration was obtained. However in practice the inadequacy of regeneration stocking before felling was not allowed to hinder the progress of exploitation (Ismail 1966). In the Malaysian dipterocarp forests the effective fruiting years are the massive long-interval ones (Appanah and Salleh 1991), and it is preferable to log following a mass fruiting year. Silvicultural operations to kill unwanted "weed" species and to liberate juveniles of desirable species can be increased following abundant fruiting, for maximum benefit in the improvement of future timber production. The abundant regeneration which may be achieved by deferment of logging until after a heavy fruiting year offers maximum possibilities for effective silvicultural operations to improve the composition of the final crop; it will also favour conservation of the genetic resources of the species favoured for commercial utilization. It seems probable that the small amount of seed of given species produced between mass seed years may be provided by a small and possibly distinct segment of the population; as the regeneration resulting from such seed will not be representative of the population, its value for conservation will be limited, unless complemented by seedlots/regeneration over a range of years.. The importance of leaving se6d trees of good phenotypic quality at the time of logging, particularly if regeneration sampling has revealed low levels of established seedlings and advance growth of the desirable species, is a further instance of close coincidence of interest between the objectives of production and those of genetic resource conservation. In practice, however, this vital aspect has been commonly overlooked or overridden by the pressures for maximum harvest yield and profit. The retention of a number of large seed bearers after the main logging operation does present some disadvantages in subsequent management of the stand if they are so numerous as to cause depressive shade or competition (Catinot 1986), or if they are subsequently harvested, with related extraction damage to the regenerating forest. However the effective loss in production is slight in comparison with the dangers of progressive deterioration in the genetic quality of the population through effective reliance for regeneration on residual, probably less vigorous and less desirable phenotypes to complement soil seedbanks and existing seedling regeneration, if these latter ones are inadequate. Despite the practical difficulties and dangers of attempting to select genetically "superior" individual trees in natural tropical forests the real dangers of dysgenic effects from the selective, systematic removal from the breeding population of the best and most vigorous phenotypes should not be ignored. More information based on solid research on this aspect is urgently needed. 25 Reliance on natural regeneration systems in production forests undoubtedly offers important opportunities for in situ conservation. The regrowth that follows heavy logging will favour pioneer (gap-phase) species and large-scale clear cutting typically leads to the complete dominance of soft-wooded, low density timber trees (Jordan 1986). As primary forests are progressively cleared or exploited the populations of many prime hardwood species characteristic of climax, mature-phase forests must decline unless specific measures are taken to maintain them. The role of the seedling populations in the natural regeneration of such species is particularly important. For example the Asian dipterocarps generally require at least a small gap in the canopy to develop to adult size but the chance of seeds naturally being placed in such a gap is low due to poor seed dispersal and infrequent seed years (Ashton 1982). This, combined with the lack of seed dormancy (Ng 1980) gives exceptional significance to the ability of dipterocarp seedlings to survive under the low light intensity on the forest floor until an adequate gap in the canopy occurs. The loss of the seedling population before such an opportunity occurs can have profound influence on the future species composition of the forest. The same is true for other species groups of economic importance in some African and neotropical forests (Whitmore 1991). An understanding of seed dispersal biology, and seed and seedling physiology, at least in so far as they are expressed in the incidence and behaviour of seedlings in the forest in relation to light conditions, must be the necessary basis for the conscious manipulation of species composition for both production and genetic resource conservation objectives. Pioneer species typically have seeds that can withstand desiccation and may become dormant for long periods in the soil. Wherever a soil seed bank has been looked for in lowland tropical rain forest one has been found (Whitmore 1990). Some pioneers have small, copious and easily dispersed seed which allows for frequent replenishment of the seed bank, or for the rapid colonisation of canopy gaps soon after they occur. Generally, therefore, the need for management interventions to favour the regeneration and conservation of the genetic resources of pioneer species is adequately met by normal logging and silvicultural treatments. 3-6 Silviculture Success in silviculture aimed at the production of timber, has been defined as the achievement of "a stand of fine-timber trees brought to maturity and producing natural regeneration on a site where it has matured before, and where the soil shows no sign of deterioration 11 (Dawkins 1988). On this definition silviculture in the originally closed tropical forest has been more successful over the past century than has been commonly believed, not only in Myanmar (Burma), India and Malaysia, but to a limited extent in some African and neotropical forests. However in terms of impacts on the genetic resources, and especially on the total range of biological diversity of plants and animals in the forests, silvicultural operations, particularly the use of arbor icides, can have far more persistent and discriminating influence than crown manipulation or logging. Insofar as the treatments achieve predictable increases in the regeneration, growth and representation in the final crop of the principal economic species, they are likely also to favour the conservation of their genetic resources. However the treatments have often proved to be unpredictable even in respect of the species they were intended to favour, and in many cases involved the attempted elimination of species which have subsequently proved valuable both for their role in the functioning of the forest and for their acceptability in the international timber markets. Conversely the success of the silvicultural treatments has 26 sometimes been due to their accidental encouragement of the regeneration of species which were not at the time considered desirable, but which are now in substantial international demand. Silvicultural systems in natural tropical forests may be broadly divided into two main groups. Monocyclic systems, also known as shelterwood systems, aim at a single comprehensive harvest of all marketable timber at the end of the rotation, with reliance on seedling regeneration to form the next crop. Probably the best developed and best known example is the Malayan Uniform System (MUS) under which, as originally designed, the initial logging was followed by poison-girdling of virtually all the remaining trees, down to a specified minimum girth at breast height, by which means the canopy became progressively more open, and conducive to satisfactory growth of the generally abundant regeneration of desirable species, principally dipterocarps. Various attempts at shelterwood systems have been made in all three tropical regions (FAO 1989b; FAO 1989c; FAO 1992a; Schmidt 1991) but problems were commonly encountered with severe climber infestations and failure to induce adequate regeneration of the principal economic species. The increased demand for a wider range of marketable species has made the failure to induce regeneration of the few chosen species less critical and the use of heavy mechanical harvesting and extraction equipment, coupled with the increased market demand, has increased the degree of canopy opening by logging alone, making the need for poisoning of residual trees less necessary. The overall ecological effect of such monocyclic systems is to favour "desirable" pioneer or near-pioneer species, including some with light, pale, general purpose, marketable timber. There are likely to be adverse impacts on the breeding populations and genetic resources of the slower-growing heavier hardwood species characteristic of climax forest not favoured by this silvicultural system and the shorter the rotation (or the more restrictive the selection of "desirables"), the more severe the impact will be over time. Polycyclic systems involve the selective removal of a limited number of stems on two or more occasions over the full rotation cycle, thereby maintaining a less uniform stand of mixed ages, with reliance on advanced regeneration for the next harvest. Such systems are theoretically capable of incorporating mature-phase climax species at the expense of accepting rather lower volume growth rates, but possibly higher value increments and therefore of conserving a broader spectrum of genetic resources in terms of species and timber qualities. However there may be some danger of dysgenic effects within populations of individual species if the selective felling removes the fastest-growing and most desirable individuals, leaving less vigorous and possibly defective stems to regenerate, in the absence of adequate, existing seedling regeneration and/or soil seed banks. Moreover if the desirable species are a small minority of the larger trees in the forest it may be necessary to undertake (expensive) operations to favour the immature trees of the valuable species, to avoid progressive impoverishment of the stand. However, potentially deleterious influences of selection on genetic quality of the species or the stand in this silvicultural system can be avoided by responsible management and harvesting practices, as were practised for example in Queensland, Australia. Under the Queensland Selection System the deliberate selection of trees to be retained in the forest, and the enforcement of strict logging controls to avoid damage to these selected trees, was designed to guard against possible dysgenic effects of selective logging. Effective operation of selection systems requires skilled and frequent tending of the desirable components of the forests and, especially, skilled and responsible logging practices, which help preserve existing advanced regeneration from accidental damage. 27 It has been suggested that for the first 30 to 40 years after the initiation of proposed systems of management the distinction between mono eye lie and polycyclic systems lies more in the future intentions and expectations than in real and irrevocable differences in practices and in the structure of the forest (FAQ 1989b). The essential question in regard to the conservation of genetic resources is the extent to which the harvesting practices and silvicultural systems allow for the retention of a wide spectrum of potentially valuable genetic diversity. This is most likely to be achieved if different forests, and different sections within the same production forest, are subjected to different systems, based on ecological principles to favour the regeneration and bringing to maturity of different elements of the main "guilds", including the climax species. Such management systems would increase the complexity and cost of harvesting and marketing, and might therefore be judged as uneconomic on recent thinking, which has tended to favour very low cost and minimum intervention approaches, since the apparently low level of financial returns per unit area of natural forest has been considered unable to bear the cost of intensive skilled management (see Mergen and Vincent 1987). This implies a management strategy restricted by the forced acceptance of the fortuitous composition of the residual growing stock after logging, with any subsequent tending limited to "blanket" operations applied without discrimination across large areas of initially diverse forest. If accompanied by the search for additional uses and market outlets for the range of timbers actually produced, the resulting balance between revenue and expenditure may appear favourable. However in terms of the impact on the genetic resources such "blanket" operations are likely to cause a progressive loss of over-all diversity, particularly through the impacts on the breeding populations of the slower-growing species characteristic of mature-phase forest. However, genetic losses are not inevitable since the composition of seedling regeneration and advance growth left after a single logging operation, provided that it does not remove too high a proportion of the growing stock (e.g. 20 to 30 m 3 per hectare on average), is likely to contain representatives of all species and guilds. These will probably be adequate to permit the restoration of genetic diversity and the deliberate encouragement of selected elements through subsequent tending, especially if different sets of elements are favoured among the management units. The practice of "liberation thinning" (Hutchinson 1987; FAO 1989c; Maitre 1991) is an example of the deliberate selection and promotion of individual stems to form the final crop, based on the residual advance growth after logging. It allows for the encouragement of a range of desirable species, presently according to lists based on timber qualities, and in accordance with assessment of the ecological requirements of selected species in terms of their likely response to treatments aimed to manipulate the overhead canopy and competition. Once the release of the selected "leading desirables" to form the final crop has been assured this system can permit the retention of a wide range of other species and therefore, if applied over large areas of diverse forest, can be consistent with the conservation of a broad range of species and genetic resources (Hutchinson 1991). Depending on the criteria used to select the "leading desirables" the system could be used to achieve ecological and conservation objectives, at the expense of some reduction in yield of the faster-growing pioneer component of the crop in some areas of the forest. Ng (1983) has drawn attention to the increasing need for "remedial management" in Malaysia in order to attempt to restore and to maintain the structure and composition of areas of mature forest. While current trends to 28 maximise production from the timber-producing areas are not conducive to the adoption of less intensive harvesting, the growing strength of both the national economy and the conservation consciousness in Malaysia may permit such ecological considerations to be applied in selected forest areas within a decade or so. The possibilities for such "remedial management tf will be dependent on research in progress or to be initiated now. The case studies in both Ghana and India illustrate other approaches to "remedial management" of production forests, to restore and maintain a sustainable and productive system. While the forest areas successfully managed in this way will certainly be different from the mature natural forest, their contribution to the conservation of genetic resources, within the context of a National Strategy for Conservation (see Section 5.1) which includes a range of forest conditions and management systems, must be very high in comparison with any alternative form of land use that might realistically be considered. Provided that the forest after logging is allowed to regenerate naturally, and is not subject to conversion to other land use, the options for conservation in situ will remain open. However they will be most heavily influenced by the care exercised in the logging, its intensity, and the interval allowed before the next harvesting operation. 3.7 Harvesting Timber exploitation in tropical forests was initially highly selective and based on a combination of animal-powered extraction and transport by river. Where such systems have persisted, as is the case e.g. in many parts of Myanmar, they will have been largely compatible with the conservation of ecological values and genetic resources. However the increasing, global use of heavy mechanical equipment, more demanding in the intensity and width of road construction, and more severely damaging to regeneration and the soil, has had very severe impacts on the sustainability and functioning of the forest ecosystem. Although the effects of intensive logging, and the associated environmental damage, may be less discriminating in their de facto impact on genetic resources than highly selective harvesting and silvicultural "refinement" operations, they tend to revert the forest to a less species-rich phase corresponding to the earlier stages of ecological succession. This effect is made worse by the severe compaction of the soil resulting from the careless use of heavy equipment, which may leave substantial areas of bare hardened and eroded surface hostile to seedling development. Coupled with the harvesting of a larger number of species, as for example in the dipterocarp forests in Malaysia, Indonesia and the Philippines, where the ease of grouping of species by timber quality has increased market opportunities, the logging damage has sometimes been so severe that both advance growth and seedling regeneration of desirable species were virtually eliminated (Masson 1983). Since most dipterocarp seedlings do not develop readily on bare exposed soil the effect of intensive logging, where up to 40 per cent of the area might be bared by careless operation of heavy equipment, was to lose nearly half the potential regeneration. The intensity of timber harvesting also determines the degree of canopy opening, which has strong influence on the successful development and composition of the regeneration. Large gaps in the canopy favour pioneer species while low intensity selective logging more closely mimics the natural processes of forest dynamics and scarcely alters the species composition (Whitmore 1990). However too frequent repetition of harvesting even on a 29 light selective system may adversely affect the breeding populations of the slower-growing species if the number of mature reproductive individuals which are present before the subsequent felling cycle is severely reduced. Repeated, intensive logging at short intervals may eliminate species characteristic of late-stage mature-phase and climax forest and is liable to produce a combination of fast-growing pioneer species of low timber value with masses of climber tangles and areas of bare soil. Such a mosaic of cleared areas and low perennial weed growth may be invaded by fire, particularly in semi-deciduous and monsoon forests, with catastrophic effects on the regeneration of most timber species and their genetic variation. The constraints on investment in forest management referred to earlier have led to increasing reliance on the logging operation as the principal means of influencing forest composition, structure and development, rather than the (expensive) thinning and refinement operations of the more complex silvicultural systems. Even when silvicultural operations are carried out after logging their effectiveness is determined largely by the state of the canopy, soil and regeneration left by the exploitation. Skilful and responsible harvesting, undertaken with understanding of ecological principles in forest dynamics, can itself serve silvicultural and conservation objectives. However all too often such aspects are entirely disregarded by logging personnel paid on a task or output basis and concerned to maximise rates of extraction regardless of the effect on the forest and the site. Further damage may be done by repeated harvesting without allowing sufficient time for recovery and regeneration. This has happened where market demand has arisen for species considered uneconomic at the time of the initial selective logging, and concessionaires have been allowed to re-enter the forest, regardless of the impact on the regrowth. Failure to adjust the harvesting operation to meet long-term management and silvicultural objectives is the greatest and most dangerous weakness in existing attempts at tropical forest management (FAO 1989c). Conversely the development of a productive partnership between the timber operators and the forest managers is the most essential component of a management and conservation strategy. Without this the possibilities for the conservation of genetic resources in production forests must be severely limited and restricted to accidental residual populations. Recent studies (Kerruish 1983; Jonkers 1987; FAO 1989b; FAO 1989c; FAO 1992a: Jonsson and Lindgren 1990) show that much of the damage caused during exploitation could be readily avoided at little if any additional cost. Although the choice of equipment is important, as powerful tractors and cable extraction systems can both cause severe damage, it is above all the lack of planning, training, supervision and appropriate incentives to the proper use of equipment which are responsible for most damage. Some studies in Sarawak have shown that the introduction of orderly logging reduced the area of forest severely damaged by 44% and at the same time effected a saving of 20% in the logging costs. Similarly in Surinam damage from skidding operations was reduced by 40% in properly controlled logging, while overall productivity was improved by 20% (Jonsson and Lindgren 1990). The amount of damage is broadly related to the number of trees felled rather than the total volume of timber extracted (Whitmore 1990). Insofar as the effect on the advance regeneration is accidental it tends to be distributed over all tree species in an essentially random way, so that in terms of impact on the genetic resources it is unselective (Johns 1988). However the impact on species already rare and subject to selective logging 30 will be potentially severe, if the future breeding populations are thereby further reduced. The seedlings of shade tolerant species, which depend for successful regeneration on survival for long periods under the forest canopy rather than rapid colonisation of gaps or sprouting from seed banks dormant in the soil, are particularly vulnerable to damage by heavy logging equipment. This compounds the adverse impacts on such species from extensive and sudden canopy opening. Since it is the heavy-seeded climax species which are most often dependent on animal seed dispersal the extent to which logging disrupts animal populations may also further affect these timber species. Stock mapping, timber marking, regeneration surveys and operator training, allied to careful planning of road making and logging operations, could be designed to conserve selected species and populations. Some studies have also indicated that quite small areas of logged forest within or adjoining logging concessions may be critically important to the survival within the area of keystone animal species (Johns 1989). Although the determination of the influence of different logging practices and intensities on the species composition and genetic resources in the forest requires ecological and autecological research there is no doubt that the degree of care exercised in the harvesting operation has the most profound influence on the future options open to management and conservation actions. The nature of the changes needed to control and provide incentives for responsible logging is clear in terms of the length and nature of timber concession agreements, levels of stumpage fees and so forth. Equally important is the interest and involvement of local communities in and around the forest, whose activities in the wake of logging operations providing increased access to the forest can strongly affect subsequent regeneration. 3.8 Non-timber Forest Products (NTFP) The importance of the many non-timber forest products extracted from natural tropical forests is now widely recognised. The term now current for such products, NTFP, generally embraces all materials of a biological origin excepting timber which is being extracted on an industrial scale. The range of products includes foods, spices, medicines, fodder, essential oils, resins, gums, latexes, tannins, dyes, rattan, bamboo, fibres, a great variety of animal products and ornamental plants. Food and fodder sources in the natural forest are particularly important as dietary supplements, to reinforce seasonally dependent agricultural systems, and in times of drought or other emergency conditions (FAO 1989d). They often represent the highest evidence of value of the forest as forest in the eyes of the local people, and are therefore an important factor in the conservation of the total resources of the forest, including its genetic diversity. The NTFP may also represent a major source of economic benefit in the national economy. Because those products which are used locally, very often the major component of NTFP, do not enter the market in which traded values are recorded, it is difficult to quantify properly their actual or potential value and therefore this is certainly greatly underrated. The potential value of future products which it may be assumed remain to be identified in tropical rain forests, for example possible pharmaceutical or cosmetic products, is often invoked in the context of the option values of biological diversity. However even without allowing for such possible future additional benefits the actual quantified value of NTFP is very significant. Southeast Asian sources probably account for most of the several billion dollars in annual world trade in NTFP (de Beer and McDermott 1989). Available estimates of export values indicate that the total figure for Indonesia alone in 1987 was at least 31 US$238 million. Probably US$100 million of this figure relates to rattan, of which Indonesia supplies 90% of the world demand. However deforestation and forest exploitation are eroding the resource base and it has been estimated that about one third of the rattan species in Malaysia and Indonesia are under threat of extinction (Dransfield 1987). Characteristic attributes of NTFP include their great variety and relatively high value per unit weight or volume, as compared with most tropical timber. Their harvesting is more labour-intensive and requires relatively little capital investment. Although the yield per unit area of forest is usually low it can be, in the case of some such products, harvested annually on a sustainable basis, with little or no disturbance to the soil or ecological functioning of the forest. Often the maintenance of a forest canopy is a necessary condition for the production of NTFP and therefore the possibilities exist for the simultaneous development of both timber and non- timber resources, with the latter providing an earlier economic return and a continuous source of income for local populations while the timber crop is coming to maturity. Selective logging can have a positive effect on some NTFP, such as rattan and edible fungi. Rattan grows best in gaps in the canopy, which can result from selective timber extraction. Insofar as the conservation of biological diversity and genetic resources of the tropical forests is dependent on systems of management that mimic as closely as possible the natural ecological conditions and processes, rather than drastic alteration of the forest condition through intensive timber exploitation or clear-felling, the simultaneous harvesting of NTFP may therefore be important in bearing the costs of conservation within production reserves. However it will be essential to undertake adequate inventories of the non-timber resources, in association with normal forest inventories, and to specify precise objectives in the management of each area of forest. The degree of precision needed in assessing the NTFP resources may be determined by the level and methods of harvesting proposed. If the products are to be gathered by local people freely and informally the level of information needed will relate principally to the regeneration and sustainability of the resource and qualitative assessments may be sufficient (see also Ghana case study). The interest in NTFP in the context of in situ conservation of forest genetic resources is therefore twofold - for their contribution to the feasibility of the conservation and management of forest resources; and for their genetic resources and their intrinsic value as components of the genetic diversity of the ecosystem. This implies the need to develop systems of multiple-use management of the forest (FAO 1984; FAO 1985a). Forest trees producing edible fruits or other products are frequently distributed widely but at low density per hectare and special attention may be needed to maintain viable breeding populations. Moreover they are likely to be involved in food- web systems, whereby seed dispersers or pollinators of other tree species may be dependent on food supplies from the fruit-bearing species. Conversely important fruit trees, such as the Brazil nut (Bertholettia excelsa) have known dependencies for their satisfactory pollination on certain large nectar- gathering bees which in turn are dependent for successful mating behaviour on wild orchid species. The loss or scarcity of orchids thus threatens the fruit production of Brazil nut trees (Prance 1985). Management for timber production as the primary objective should be planned to be compatible to the greatest extent possible with the production of NTFP (FAO 1989b). This implies not only a wider information base from broadly based forest inventories but also much more complete understanding of 32 the forest dynamics. High hopes have been raised for the future of so-called Extractive Management Reserves in the Brazilian Amazon, based on a range of NTFP resources, of which the best established are rubber and Brazil nuts. The range of products already identified is certainly large and some studies have indicated a high commercial potential (Peters et al 1989). However there are considerable uncertainties over the replicability and sustainability of extractive management systems and some recent studies indicate that careful selective extraction of timber is likely to be a necessary component of overall management to secure sufficient levels of income from the forest. While the development of NTFP resources may have positive impacts on the conservation of genetic resources in a wider sense, the production of NTFP and of timber is not necessarily always fully compatible. Favouring NTFP can, at times, lead to a marked decrease in levels of timber harvesting, at least in the short term, since many timber trees are also sources of fruits or other extractives; the cultivation of some high value products such as cardamom (Elettaria cardamom) in the forest may hinder the establishment of regeneration of timber species in these areas (FAO 1984); etc. Nevertheless the possibilities for different systems of multiple use of natural forests are of undoubtedly high potential in terms of their contribution to conservation of ecosystems and the in situ conservation of genetic resources of a variety of species. While it may be difficult and, in some cases at least, impossible to combine the different management objectives on the same limited area of forest, with the same intensity of timber harvest, canopy opening and population refinement, different working circles within the same forest may be used to maintain a mosaic of different ecological stages or conditions. In some cases working circles may overlap and in others they must be kept geographically separate. Such zonation is also compatible with the development of "buffer zones" around the forest, and "core zones" devoted to strict protection. This is dependent on high levels and intensity of management (see also the India case study in Part II). 3-9 Involvement of local people Very little tropical forest is truly virgin forest in the sense of never having been inhabited by man (Webb 1982). Most areas have been subject to forest clearance and cultivation and are now composed of a mosaic of patches in varying stages of development towards the "climax" condition. A Nigerian forest described by Jones (1955-56) had apparently not reached a steady state after about 250 years (Whitmore 1991). In some areas, particularly where clearance was very extensive and prolonged by repeated fires, the species composition of the vegetation has suffered a major change, for example to a derived savanna woodland, or even grassland, possibly maintained by grazing and browsing as well as occasional burning. However in the absence of such repeated destruction of the regeneration, tropical forests show considerable resilience in recovery after clearance. On the other hand, unless given areas are allowed to remain as, and return to, the mature-phase (climax) forest, the genetic resources of species characteristic of this phase may become threatened due to the long timescale needed to reach it. The establishment of forest reserves to be managed for timber production, to the exclusion of traditional shifting cultivation, was initially a safeguard against such a danger. This has been compromised by the increased intensity of timber exploitation and the widespread failure to apply adequate harvesting controls and lack of subsequent management interventions to ensure effective regeneration in harvested forests. At the same time increasing encroachment or illegal logging which have reduced at least parts of the forest in many (most) tropical countries to a very degraded condition. As the increasing 33 human populations put more pressure on the scarce resources of fertile land the threat to the remaining forests has continued to increase (see also India case study). Recent reviews of forest management recognise that activities must take full account of the needs of the rural communities and that no attempted management system can be sustainable without the broad approval of local people in both planning and implementation (FAO 1989b; FAO 1989c; FAO 1992a). Such approval is unlikely to be achieved without the provision of some tangible benefits in the short as well as the longer term. As long as the perceived benefits of logging are that it allows access to the forest for illegal farming and theft of timber, particularly the undersized stems and advance growth of valuable species left as a result of selection felling to girth or diameter limits, the combined effects of the legal exploitation and subsequent depredations must be increasingly damaging to the genetic resources of valuable species. However the incorporation of extractive management of NTFP, together with the development of small scale rural enterprises based on the selective extraction of timber, could provide local employment and income. Increased participation of local people, combined with less intensive logging regimes, could then help to conserve a wider spectrum of genetic diversity iri situ. However the fundamental requirement is the full endorsement by the local population that the land should remain permanently under forest. Such theoretical models of the positive involvement of local communities in fully participatory management of production forests on a sustainable basis are still at only a planning or pilot stage in a few countries. The OEPF (Organisation de Ejidos Productores Forestales) project in Quint ana Roo, on the Yucatan peninsular in Mexico, has been cited as an example of the involvement of local communities in the management of timber production forests, formerly under concessions to logging companies. Here the communities (ejidos) are directly involved in all aspects of forest management and share in the profits generated through the sale of forest products. A high level of local support for the forest management is reported, related to the provision of employment and income (WWF 1991). However technical problems related to the regeneration and other silvicultural aspects of the project are also reported to require substantial external assistance (WRI 1991). Following earlier reviews of possible approaches to multiple-use forest management in India and in Ghana (FAO 1985a) new initiatives are now under way in both countries, including elements of both the involvement of local people and the conservation of biological diversity and genetic resources. In the Western Ghat forests in southern India draft proposals for multiple-use management are based on the zonation of the forest and surrounding land into five management zones, of which the central zone (Zone I) is to be dedicated principally to the conservation of biological diversity and genetic resources. This approach reflects principles recognised in the "buffer zone" concept (Sayer 1991) as well as the participatory approach of social and community forestry schemes. Another example of ways in which local people, particularly those long settled or indigenous to the area, can have positive impacts in the conservation of the forest and its genetic resources is through the use of local knowledge in taxonomic, ecological and phenological studies. Such studies are essential to the adequate understanding of forest dynamics for conservation objectives. With appropriate orientation to the lines of scientific research needed and some training in the categories of information required, the intimate knowledge of the forest and of many of the species 34 which is often held by local people provides a valuable basis for taxonomic and ecological studies. Many tropical botanists can testify to the skill and value of local "tree finders" in taxonomic and ecological studies in all tropical regions. The scale of the data collection needed, even for a small selection of the many thousand species present in most tropical forests, is out of all proportion to the scientific manpower and financial resources available within each country and internationally. An interesting example is the use of so- called "para-taxonomists" in the current programme for a national inventory of biological diversity in Costa Rica. This is under the overall direction of the Costa Rican National Biodiversity Institute (INBIO) which initiated the programme in 1989 with the joint objectives of conservation and the exploration of potentially valuable organisms. Data from new collections is brought together with earlier information in a computerised database which can provide information on important aspects of a species 1 biology and autecology linked to taxonomic identifications. The mounting of intensive short courses in M para-taxonomy l! has transformed the rate of collection and subsequent specialist identification of arthropod species. In addition to the value of the information and material collected this approach can provide an important link between the local communities and the forest managers (Tangley 1990). In the programme in Costa Rica, in contrast to the situation in many African and Asian countries, the local population is not indigenous to the region and lacks the long history of forest use still found in most other areas in Central and South America. The involvement of the para-taxonomists provides a link between the communities and the better understanding of the natural resources of the country. Since the most damaging impacts on the natural forests came from immigrant populations and technologies this improved understanding through involvement in gathering data on the biological diversity has both direct and indirect benefits. 35 CHAPTER IT THE FUTURE OF TROPICAL FORESTS When considering possible action and investment in the conservation of ecosystems, species and genetic resources some assessment of the future demands on the forest, and the land it presently occupies, is essential. The existing growing stock and particularly the advance growth at the time of harvesting will determine the species composition and possible selection of the seed bearers at the end of the cutting cycle, perhaps 15 to 30 years, or up to 45 years or more ahead in some cases. However the management and conservation of the genetic resources of the timber crop are concerned with the influence on the next reproductive phase and on subsequent generations, taking into account such factors as the possible effects of inbreeding, genetic pollution of locally adapted genepools by outside (introduced) pollen sources, or genetic drift on the evolution of the populations. This implies consideration of the management and production objectives that may apply perhaps 150 or more years in the future. Any attempt to predict the socio-economic and environmental situation of the production forests so far ahead must be so uncertain as to be of very questionable value. Even within the past half century the changed nature of the demands and opportunities surrounding the forests has radically altered approaches to management. Moreover the rapidity and scale of the changes, particularly in aspects of technology relating to the use of timber and wood and to the possibilities for manipulating the crop, including aspects of genetic engineering, are still apparently increasing. In addition the probability of significant change in global and regional climates adds further uncertainty, since the exact nature of such changes and therefore the nature and extent of their impacts on the forests, and on the environments for future production forests, are largely unpredictable. It is thus not to the prevailing socio-economic, market and environmental conditions, nor those of the past decades, that the genetic resources of the forest must be matched, but to those well into the next century. Failure to keep this fact in mind could lead to a dangerous limitation of the conservation objectives and of the necessary genetic base to adapt to unforeseen demands. 4. 1 Population and Land Use Despite concern for great uncertainty related to future environmental conditions and market demands described above, the inevitability of the increase in human populations and in the severity of the related impacts on natural resources, including natural forests, must be the central concern in forward planning. Unless some significant additional improvements beyond current expectations can be made in agricultural productivity on the existing crop lands, very large areas of additional arable land will be needed each year to feed the increasing population. On past and present experience this land will be taken largely from existing forests. Economic analysis of the productivity and contribution to national and local needs from the natural forests, in the face of the increasing demands on land, is likely to conclude that the natural forest is unacceptably expensive in its use of land. At the same time shortage of capital in most tropical countries, and the high discount rates applied to project appraisal in the use of external funds, are a strong deterrent to investment in the rehabilitation of the extensive areas of already deforested and degraded lands. The effect on the forest "land 36 bank" is therefore likely to be as devastating in the near future as in the recent past, and as the forests are further reduced the impact on their genetic resources is made more severe. Again on past experience this is likely to be the situation even if, as is increasingly the case, agricultural production on the deforested land becomes unsustainable due to the inherent intractability and low fertility of most areas still under forest. The eventual costs of attempting to restore productivity to the deforested lands are likely to be very high, and the success may well depend on the availability of the genetic resources of some of the original perennial vegetation, shrubs and trees. On the assessment of some agricultural crop geneticists we have already come very near to the limits of lands utilizable for agriculture (Hawkes 1990) and must look to the more effective use of marginal lands through adaptation based on selection and breeding, for the needed increase in food production. This underlines the importance of the genetic resources of woody multi-purpose species in dry and semi-arid lands, which becomes even more significant in the context of global warming and with the likelihood of more extreme climatic events and stresses. In this respect the conservation of critically important and endangered fragments of dry forest, and of their genetic resources such as i..a. those in the Pacific lowlands of Central America, is dependent on the recognition of their importance for long-term sustainability of land use. Similar considerations apply to many remaining areas of tropical forests and rainforests on sites of low inherent fertility, such as the deep sands underlying some Amazonian rain forests and the podzols of heath forest formations. 4.2 Timber Demand and International trade While trends in population and land use are clear, predictions of future levels and patterns of timber and wood consumption, even a matter of a few decades ahead, are more problematical. The long-term predictions required in the context of genetic resource conservation must be extremely uncertain. Most forecasts of timber trends have a shorter horizon but a recent study by Arnold (1991) undertaken for the U.K. Forestry Commission, provided some valuable indications of longer term trends. In global terms, and also for each of the major industrialised regions, the rate of growth in industrial wood consumption has been slowing down - worldwide from 3.5% annually during 1950/60 to 2.2% in 1960/70 and 1.1% in 1970/80 (Sedjo and Lyon 1990 in Arnold 1991). This apparently reflects a number of long term trends which seem likely to continue, namely that the uses and markets for some applications, such as housing in much of Europe, North America and Japan, seem to be approaching maturity, while improvements extending the life of wood products, reductions in wasteful uses of wood, and increases in the role of substitutes for timber, are having an increasing influence on levels of demand. The results of the analysis reported by Arnold (1991) show that consumption is rising faster in the developing countries than in the industrialised regions and that substantial increases in the demand for industrial wood are expected in Africa, Asia and Latin America in the remainder of this century. The forest-based sector in developing countries experienced rapid industrialisation between 1950 and 1980 but increasing domestic demand, linked to the expanding populations, absorbed the major part of the increase in output (ECE/FAO 1986). Since the domestic markets in tropical timber- 37 producing countries are less demanding in terms of quality than their export markets they are able to utilise a wider range of species and, as the total wood resources from dwindling areas of natural forest diminish, they may accept low-grade material of mixed origin, rather than the high quality and consistent supplies required by the international markets. At the same time the pressures on land have led several tropical countries, especially those with high population pressures, to propose the effective replacement of natural forest by industrial plantations to meet timber production needs (Nwoboshi 1987; Kio and Ekwebelan 1987). The implications of these trends for the management of natural tropical forests are for the acceptance by the domestic market of mixed, general- purpose timber, to be provided at minimum cost with little regard to the selection of species, and perhaps accompanied by the progressive reliance on artificial regeneration of fast -grow ing species, including exotics, rather than natural regeneration of the native forest. Unless active moves are made to counteract it, the management of natural forest is likely to tend towards the "log and leave 11 approach (Poore 1989) with limited reinvestment in control either of logging or of subsequent protection of the regenerating forest. This pattern of operation is already followed in many tropical forests, reflecting the low-input and low-output levels of funding and revenue. Investment may be more readily attracted to industrial plantations and so far as these also will be intended primarily to meet domestic needs for wood and timber, the species selected will be fast-growing general-purpose ones, capable of satisfactory growth on the degraded land not required for food production. Pioneer species, such as the tropical pines, are the natural choice for such plantations, with the advantage of having already been the subject of substantial research into their genetic variation and adaptation to a range of sites, through provenance trials (Barnes and Gibson 1984; Gibson et al 1989). If this trend were followed to its' logical conclusion the conservation of the genetic resources of tropical timber species carried out in production forests, would be limited to the pioneer and fast-growing gap- phase species capable of maintaining viable populations through repeated cycles of largely indiscriminate fellings, at intervals of probably 15 to 20 years on many sites. The international trade in tropical timber has been an important source of revenue for many tropical countries, based in the past largely on the first cut in natural forest, including a high proportion of high-quality hardwoods with exceptional or even unique qualities. A majority of the highest value timbers are among the slower-growing late-phase ecological groups, characteristic of the mature-phase and climax forest. Some important commercial hardwoods, particularly among the Meliaceae (e.g. Cedrela odorata, Entandrophragma spp, Swietenia spp, Terminalia ivorensis, T. superba, Milicla excelsa Lsyn. Chlorophora excelsaj etc) are natural gap-phase species but are unlikely to form a large proportion of the crop under short-cycle "log and leave" systems of management. Without availability of continuing supplies of substantial volumes of such medium-value hardwood species in future timber harvests the prospects for profitable export-oriented management of tropical forests seem poor. The bulk of the world's current supply of industrial wood comes from the forests of the north temperate zone where in total, in contrast to the tropical forests, the net increment exceeds the removals and the volume of the growing stock is increasing. Moreover the developing trend in this zone to remove productive land from agricultural production in response to economic 38 and market related pressures, is expanding the potential for further increases in both area and productivity of the forests. While there may be some constraints on industrial wood production from these temperate forests if decisions are made to restrict timber harvesting in favour of amenity and recreational objectives, and due to hazards as a result of possible pests, diseases or air pollution, the likely effects of global climate change seem favourable for increased growth rates in forests in the north temperate zone. Moreover the strong concentration of scientific expertise, and the high level of forest management generally in this zone, offer the best basis to anticipate the effects of climate change, to mitigate the adverse impacts and to take advantage of the new opportunities for faster growth. It seems likely, therefore, that the future trade in tropical timber will be increasingly restricted to the export of higher quality woods to the industrialised world, with some continuing and developing trade in general purpose timber between developing countries (Arnold 1991). Leslie (1987) has drawn attention to the special economic importance of the high value timbers which are unique to the natural tropical forests, and for which, therefore, there are no adequate substitutes. He points out that the market outlook for these timbers is the direct opposite of the weak market prospects for most of the products of the alternative agricultural or plantation management systems which compete for land with the natural forests. In these circumstances the relative economic prospects for natural forest management, if based on such high-value timber species, can only improve. Some recent studies of the possible incentives to sustainable tropical forest management highlight the potential importance of securing or transferring a greater proportion of the ultimate value of the products, as realised in the market outlets in the industrialised countries, for reinvestment in the forests (e.g. OFI 1991). An important aspect of this is the development of efficient secondary and further processing of forest products in the countries of origin. Exports of processed products have been expanding, despite some evident constraints on market opportunities in the main user regions. There are dangers in the expansion of industrial capacity beyond the properly sustainable levels of the allowable cut in the forests. Nevertheless with adequate quality control and marketing arrangements high- value secondary and further processing seems the best option to maintain a unique market niche for selected tropical timbers, thus providing an incentive for sustainable management of natural tropical forests. Reports of species such as ebony, teak and rosewood being exceptionally traded at prices between US$ 5 000 and $ 7 000 per m 3 (ITTO 1991) indicate the existence of a high value market niche which is likely to remain firm as the availability and supply of such high quality timbers from the natural forests continues to decline. Concern among the public and the media in the industrialised countries over deforestation and degradation of the tropical forests, and particularly the impact of logging on tropical rainforests, is likely to have an increasing impact on international trade. On current trends it seems likely to reduce substantially the traditionally strong demand for tropical timber in Germany, the Netherlands, the U.K. and the U.S.A., and even in Japan. The International Tropical Timber Organisation (ITTO), at its 8th Council Session in 1990, adopted the target date of the year 2000 by which to ensure that all tropical timber in international trade should come from sustainably managed forests. At the same session it approved a set of international guidelines for the sustainable management of natural tropical forests (ITTO 1990) elaborated by an Expert Group in which individual experts, international 39 organizations (FAO) and an NGO (WWF) were represented. At its 10th Session in 1991 the ITTO Council initiated action to develop Guidelines for the Conservation of Biological Diversity in Production Forests, to complement the already existing, general guidelines on sustainable management in natural and plantation forests. To the extent that the timber producing countries were able to achieve the ITTO target of ensuring sustainable management of the production forests by the year 2000, and installing internationally accepted certification for timber to that effect, the related consumer resistance to tropical timber imports might be removed. The criteria for sustainable management seem certain to include aspects of ecological and environmental concern, including the conservation of biological diversity and of genetic resources of the tree species being marketed. Depending on the criteria adopted for judging sustainable management, production forests managed to maintain a broad spectrum of species and their intra-specif ic variation, and including a range of successional stages over the national territory, are more likely to qualify for public and political approval than more intensively logged on a short cutting cycle, without concern for genetic conservation. Countries which are seen to incorporate the conservation of forest genetic resources, and concern for the broader range of biological diversity, within their management systems in production forests will be best placed to secure favourable markets for timber. 4.3 Tropical Forests and Environmental Concerns The link between tropical forests and the stability of local, and possibly regional and global climatic conditions is widely accepted, if still imperfectly understood. Studies in the Amazon and in West Africa have shown the importance of transpiration from tropical forests in influencing local rainfall, and their significance in the hydrological cycle (Salati 1987; Shuttleworth 1988). It is thought that tropical forests may play a key role in the general circulatory systems of the atmosphere, with related influence on precipitation patterns. Their role as a major store of carbon is clear, although the mature-phase forests are presumed to be in approximate equilibrium in their sequestration and discharge of carbon dioxide. In that respect the establishment of fast-growing young forests is clearly a potentially greater influence on possible global warming, to the extent that it may be the result of increasing carbon dioxide concentration in the atmosphere. The species composition of tropical forests, except insofar as it may be essential to the functioning of the ecosystem on a given site, is unlikely to be of critical importance to their role in regional or global climatic stability. There is also little evidence for links between species diversity and ecosystem function (di Castri and Younes 1990). With the exception of certain "keystone" species there is apparently a high level of species redundancy in the functioning of the highly diverse tropical forests, and it seems likely that well-managed secondary forests, composed largely of relatively fast-growing species characteristic of the earlier stages of ecological succession, could adequately fulfil the forest's environmental role. 40 However it is also increasingly accepted that the loss of biological diversity is itself an environmental problem. The danger for such loss is immediate, and therefore as urgently in need of increased international action as is the threat of global climate change. In 1988 UNEP, together with other members of the Ecosystem Conservation Group (then consisting of FAO, Unesco, UNEP, IUCN, and WWF International) initiated action towards the preparation of an International Convention on Biological Diversity. From the outset the special importance of tropical forests in this connection was recognised. The fundamental requirement of the in situ conservation of genetic resources and the conservation of ecosystems and natural habitats was taken as a central principle, linked to general obligations on all parties to a possible Convention to conserve natural habitats, species, viable populations and genetic resources in situ. There was also widespread recognition of the need to integrate the conservation of biological diversity with development, and the possible role of forests managed for the production of timber and other products in this connection. The various drafts of a convention prepared by the Ecosystem Conservation Group, were subsequently discussed by national Governments in the context of preparations for the United Nations conference on Environment and Development (UNCED) 1 . WRI, IUCN and UNEP, in consultation with FAO and Unesco, also prepared a "Biodiversity Strategy and Action Plan 11 , calling for a decade of action, and for the necessary financial resources to be made available internationally (WRI 1992). Discussions of a possible international "umbrella" convention, charter, protocol or other agreement aimed at the conservation of forests, as suggested by the group of seven leading industrialised countries at their meeting in Houston in July 1990 were actively pursued in the FAO Committee on Forestry in September 1990, and at the FAO Governing Council meeting in November 1990. Subsequently it was agreed to pursue these discussions in the forum of the preparatory committees for the UNCED meeting, and this led to the formulation of a draft statement of principles for the management, conservation and sustainable development of all types of forests 2 . Although not legally binding the "Forest Principles" may be used as a basis for a more formal international agreement in future. Any such international instrument should pay particular attention to the importance of tropical forests in the conservation of biological diversity and genetic resources, and to the implications for the provision of both financial and technical assistance to the tropical countries in that connection. While genetic diversity of the tropical forests may not be significant in the forest *s climatic function, it is certainly vulnerable to the effects of climate change. The effects may be particularly severe in the major ecotones, where adjacent biomes meet, for example in the transition zone 1 The resulting International Framework Convention on Biodiversity was signed at the UNCED Conference in Rio de Janeiro (June 1992) by 154 countries; it will come into force after national, governmental ratification by 30 signatory countries (see e.g. FAO 1992b). 2 "A Non-Legal ly Binding Authoritative Statement of Principles for a Global Consensus on the Management, Conservation and Sustainable Development of all Types of Forests", was adopted at UNCED, Rio de Janeiro (June 1992). 41 between closed tropical forest and savanna woodland (Holdgate et al 1989). In such situations, where it has been suggested that a change of 3C in average temperature would lead to a shift in habitat type of roughly 250 km in latitude (MacArthur 1972, in McNeely 1990), each species will respond within its own capacity. This will be strongly influenced by the levels of genetic diversity between populations and between individual trees within each species, and places greater significance on the conservation of the genetic resources of the woody species in such transition zones. In the tropical forests pioneer species with light, wind-distributed seed, or those with highly effective dispersal through birds, bats or other animals known to disperse seeds over wide areas, are likely to be able to adapt more easily to climatic change than species with large, heavy fruits that fall intact, and whose seedlings are adapted to survive under the forest canopy. This suggests that the trees characteristic of mature-phase climax forest are likely to be more disadvantaged by climate change than the more widespread pioneer species, especially if the latter also exhibit a high degree of diversity and strongly outbreeding systems. 4.4 Protected Area Systems Increased recognition of the importance of biological diversity, and the exceptional wealth of tropical forests in this respect, may improve the opportunities for systematic inclusion of aspects of in situ conservation of genetic resources in fully Protected Areas, such as National Parks or Native Reserves. The principles were already established in the World Conservation Strategy (IUCN 1980) which has been widely accepted. Central to the Strategy is the recognition of the interdependence of conservation and development. This theme is further developed in the 1990 adaptation of the Strategy (IUCN 1991 a) which calls for a comprehensive system of protected natural forests and for expansion of efforts to conserve forest genetic resources. The use of the term "Protected Area" covers a variety of approaches to the protection and management of natural and semi-natural areas, classified into eight major categories. Several of these categories can permit sustainable harvesting of forest products within the management objectives and practices, both to conserve biological diversity and to provide sustainable benefits to local people and to national economies. This applies for example, to Category IV (Nature Conservation, Managed Nature Reserves or Wildlife Sanctuaries), Category VI (Resource Reserves), Category VII (Natural Biotic Areas or Anthropological Reserves, for Traditional Local Harvesting of NTFP) and Category VIII (Multiple-Use Management Areas or Managed Resource Areas) (IUCN 1990). However the fundamental objectives of the Protected Areas are to maintain the ecological processes inherent in the natural systems and to conserve their genetic diversity and resources for sustainable use. Two other categories of conservation sites are recognised internationally which may overlay the IUCN categories, namely the Biosphere Reserves, under the Unesco Man and the Biosphere (MAB) Programme; and the World Heritage Sites, nominated by countries party to Unesco 1 s World Heritage Convention. The Unesco MAB Programme works closely with FAO, UNEP and IUCN and with the International Council of Scientific Unions (ICSU). All Biosphere Reserves are intended to have a scientific as well as a developmental objective, and may permit harvesting, including logging, where considered appropriate to improve understanding of the scientific basis for sustainable management. Areas accepted as World Heritage Sites may qualify for some financial assistance through the World Heritage Trust Fund, which might entail 42 restrictions on use of the resources. Roche and Dourojeanni (1984) have evaluated the various categories of Protected Areas in terms of their contribution to forest genetic resource conservation. Although Protected Area systems form a central "core" in national and international action to conserve biological diversity, the developments in conservation biology (e.g. Harris 1984; Sou 14 1986; Wilcox 1990) have revealed the limitations of such systems in the conservation of ecosystems, species and genetic resources. Most conservation biologists now recognise that the Protected Area networks, even at the more optimistic assessments of the areas likely to be secured, will not be able to conserve all, or even most, of the species and genetic resources desirable (FAO 1989a; FAO 1992c; McNeely et al 1990). The total area and pattern of distribution needed would far exceed the practical possibilities and the willingness of local communities and governments to set aside such areas from productive use, or to decisively limit such use. The only solution, therefore, is a planned mosaic of Protected Areas, integrated with production forests managed in such a way as to contribute in a complementary capacity to the overall conservation of biological diversity in general, including genetic resources of important component species. The location of Protected Areas is normally based on considerations of natural landscape value, species richness, endemism, degree of threat from destruction of the habitat, theories of Pleistocene refugia and attention to so called "hot spots" in deciding priorities (Wilson 1988; Myers 1988; Reid and Miller 1989; McNeely et al 1990; Wilcox 1990). Their representativeness in respect of forest genetic resources is severely limited both by the lack of information on patterns of distribution, especially at the intra-specif ic level of variation, and by the pressures from other land-use systems. Most remaining areas of primary or climax forest are on sites judged unsuitable for intensive agriculture or very remote from centres of population, or both. In the fertile tropical lowlands, forests have either been heavily modified or eliminated. Where the opportunity to select and set aside Protected Areas still exists the attempt to conserve potentially valuable populations of important species must normally be based on patterns of environmental variation or of variation in plant communities, rather than on detailed knowledge of the genetics and actual variation patterns of their component species. The size and shape of individual Protected Areas also are strongly influenced, and often determined, by outside pressures rather than theoretical considerations of minimum population size. However even small areas can make important contributions to conservation (Simberloff 1982, 1983) and reserves of even less than 10 ha can be effective in conserving viable populations of many plant species (McNeely et al 1990). One advantage of small areas is that they are more manageable, both for the study of their species composition and autecology and for their effective protection, and they offer greater possibilities for diversification over the whole of the natural territory. A thorough inventory of the tree flora, and desirably other flora and fauna (considering occurrence, density and distribution), is essential for efficient conservation, and may enable such small areas to be effectively used in conjunction with other in situ conservation sites, and with ex situ action (see also the Ghana case study). 43 The value of Protected Areas for the conservation of genetic resources is usually weakened by lack of resources and capability for their control and management. Roche and Dourojeanni (1984) have drawn attention to the necessity for effective management of Protected Areas, and the key role of the Managed Resource Areas (IUCN Category VIII) in this respect, since they permit manipulation of populations with a view to the utilization and possibly incremental development of their genetic resources. They have also suggested a review of the legislation governing other categories of Protected Areas, such as National Parks, in order to permit activities such as harvesting of seed and other reproductive materials. They point out that the totally protected area ("core") within many National Parks is often sufficiently large to allow for zonation into different sections, under possibly different management objectives and practices. This principle, of the association of different systems of management within contiguous forest blocks, is very important to the effective integration of conservation and development objectives. It is well illustrated by the Virgin Jungle Reserves within production forests in Malaysia, and by the concept of "buffer zones" around protected forests. 4.5 Buffer Zone Forest Protected Areas, particularly those of less than ideal size and location, can only meet their conservation objectives if the land surrounding them is under appropriate management compatible with the objectives within the Protected Area itself. This may require the establishment of a "buffer zone", which can both meet those objectives, and also provide appropriate benefits to the local people. Experience has shown that legal protection alone of conservation areas is insufficient to prevent encroachment or damaging incursions into the forest, particularly where human settlements are located close to the boundary (Sayer 1991). Moreover in relatively small reserves the tree species which occur naturally at very low densities will be at risk from possible inbreeding or inadequate levels of regeneration, which may be relieved if their populations are extended into surrounding areas of managed forest. Such extension of protection may help conserve a wider range of intra-specific variation. Some attempts at buffer zone development have proved disappointing (Wells et al 1990) but others have been more successful. The benefits, in addition to the extension of the effective population size of some species and the separation of the "core" protected area from human settlements and intensive agriculture, include the development of local support for the conservation objectives as a result of beneficial involvement in use of the buffer zone, including e.g. hunting areas. The definition of the buffer zone concept by MacKinnon (1981) was "areas peripheral to National Parks or reserves which have restrictions placed on their use to give an added layer of protection to the native reserve itself and to compensate villagers for the loss of access to strict reserve areas". However Sayer (1991) interprets the concept more widely to incorporate a range of possible developmental activities which can deliver benefits to local people as well as extending the effect of the Protected Area, at least for some animal and plant species. These activities might in some cases include agroforestry, tree plantations or a variety of other land uses but in terms of the in situ conservation of the forest's genetic resources the most appropriate form of management is clearly modified utilization of the natural forest. The Corbett National Park in northern India is a long-established 44 example of the way in which the maintenance of semi-natural forests of Sal (Shorea robusta) have buffered the Protected Area while providing a harvest of quality timber to the forest authorities and non-wood forest products to the local people. The best realistic prospects of maintaining a buffer zone of natural forest with relatively little change in its species composition, and therefore maximum contribution to in situ conservation, are in the few remaining major forest blocks, such as Amazonia, the Zaire-Congo Basin and parts of Indonesia. Some possibilities still exist elsewhere in Africa, where population pressures around the forest are still not severe, to develop a buffer zone of secondary forest, with a higher concentration of pioneer species. However light, sustained-yield selective logging or harvesting of NTFP, possibly in combination, should be better able to maintain a wide range of total diversity. The important thing is for the management of both the Protected Area and the adjacent land to be planned and managed as an integrated unit. By this means the genetic resources of species of both the late succession forest and those of the earlier stages of ecological succession can be suitably catered for in separate management zones. Clear objectives and priorities must be set for each zone, from the central "core 11 , where conservation of over-all biological diversity may be paramount, through selectively logged forest, with due attention to the conservation of genetic resources of important timber trees, to the outer zones, where production of the wood and non-timber forest products may be the highest priority, with close involvement of local communities (see India case study). A further important role for areas of managed forest outside the main Protected Areas is to serve as "corridors 11 for the movement of animal populations, particularly in the context of possible climate change. If sufficiently broad and secure they might also allow for migration of some plant and species, including tree species. However given the probable speed of change and the extensive fragmentation of tropical forests the movement of genetic resources of important tree species will be more dependent on ex situ conservation strategies. 45 CHAPTER V STRATEGIES FOR IN SITU CONSERVATION IN PRODUCTION FORESTS 5.1 National Policies Genetic resources are a national asset of the country and therefore dependent on clear national policies for their use and conservation. The potential contribution of each production forest, or unit of management within a forest, to the national objectives for conservation and/or sustainable productive use will vary according to location, species composition, size, shape, environmental features and many other factors. Its actual contribution should be determined by the objectives and quality of management, related to its potential conservation value. It is neither possible nor necessary to prescribe equal priority and intensity of in situ conservation in all production forests. At a minimum level sound sustainable management should include provisions for the protection of site conditions, seed trees, seedling regeneration and advance growth of desirable species in appropriate combinations according to management and silvicultural prescriptions. To the forest geneticist a range of possible production management systems may be acceptable in terms of their impact on the genetic resources, depending on the target populations. Extreme refinement of the composition of the forest to favour one or a few species in an originally poly-specific forests, if done wi,th full understanding of the dynamics of the ecosystem and of the effect on its long-term functioning, may be an acceptable means to conserve the genetic resources of the principal species, albeit at the expense of the broader (species) diversity of the forest, which should receive due attention in other areas of the Forest Estate. Even clearfelling and replacement of the mixed forest by planting a single indigenous species, if based on broadly representative seed collections from the same site, might qualify as in situ conservation. However such approaches, would be exceptional to the general rule which would seek conservation through natural regeneration of the component species in the ecosystem, targeted for conservation. In many cases the need to maximise production of wood and other products from limited areas of land requires the concentration of the site capability on the growth of one or a few fast-growing species or species groups, most likely to be from pioneer or gap-opportunist guilds, and possibly exotics. The forest manager might argue that this action supports the allocation of other forest areas to in situ conservation, as part of an overall strategy. However past experience with the putative development of "compensatory 11 plantations to safeguard natural forest has indicated the need for a stronger and more coherent overall plan to ensure that the conservation objectives are also met. As stated earlier the role of individual areas of production forest ^ or in situ conservation must be determined in relation to patterns of distribution and variation of target species and populations, and to the contribution of the network of fully Protected Areas, National Parks etc, within an overall, national conservation strategy. In addition to using production forests to complement the Protected Area system, by filling key "gaps" in species ranges, or forest types, there can be additional value in the conservation role of production forests sited adjacent to a protected area, or forming a "corridor" to other managed or protected areas. 46 The wider regional and international interests in the conservation of biological diversity and genetic resources related to patterns of distribution extending through several countries, and to global recognition of the importance of the tropical forests in this regard, presents an opportunity to secure additional financial and technical support for conservation actions. While these international concerns, particularly when focused on existence values of biological diversity, are likely to extend beyond the more immediate use values of the forest genetic resources of direct interest to the host country, they may be used to reinforce national objectives in the conservation of selected forest areas and populations. However to ensure that full advantage can be taken of these expanding opportunities, in ways that serve the national interest, as well as broader global objectives, a clear and coherent national policy on the conservation of its genetic resources is essential. As seen in the case studies on Ghana and India, for example, the actions needed to secure effective conservation must carefully consider and include the interests of local populations in and around the forest, and will also extend increasingly outside the forests into aspects of industry and trade; and outside the forestry sector as a whole, to areas of responsibility in other sectors of government. They may impinge on central policies related to the allocation of revenue and expenditure, and to both internal and external commercial and trading practices. In these cases the necessary changes to management practices in individual forest areas are dependent on central policy decisions at a high level, involving various sectors. Review and revision of existing laws and regulations may be needed. Although many of the actions, such as the review of the level of forest fees, the adoption and effective imposition of differential stumpage rates, the systems of allocation, duration, size and operation of timber concessions etc, are needed as much for the broad objectives of sustainable forest management, as for conservation objectives, the latter roust also be kept clearly in view in determining overall policy. Similar considerations apply to the changes needed in forest management practices, and perhaps in laws and regulations, to incorporate NTFP and the involvement of local people in such management systems. The potential for this is seen in all three case studies presented in Part II of this document. The provision of a coherent system of incentives to sustainable management of the forests, at all levels from the local populations to other national groups in both the private and the public sector, and to international market conditions and investment policies, relating for example to trading practices, support for local processing, marketing assistance and maximising value of products to the country of origin, will be needed to secure greater attention to conservation objectives in the production forests. These incentives must be the subject of national formulation and may require international assistance (see the Ghana case study) . An urgent task is the determination of priority species, populations, areas and actions for the conservation of forest genetic resources in each country. This must take account of possible ex situ as well as in situ conservation, within a coherent national programme, in accordance with national policies. In a broader context, embracing all natural resource sectors, the formulation of National Conservation Strategies has been used as an aid to reviewing policies and redefining priorities (Poore and Sayer 1987). The breadth of subjects and issues involved in the whole field of biological diversity conservation has proved difficult to contain within a single comprehensive assessment, and the need for more specific definition of 47 policies and practices linking development and conservation has been suggested (FAO 1985b; Prescott-Allen 1986). In the narrower context of the conservation of forest genetic resources, the multiple-use management of production and protection forests for the production of timber, NTFP for local use, environmental protection and the conservation of genetic resources provides a contribution to exemplify and help define the broader Conservation Strategy. For all of the above reasons the formulation of a National Strategy for the Conservation of Forest Genetic Resources is essential, to make the best use of the land and other resources devoted to production, protection and conservation, and to the opportunities for regional and international cooperation. An important task under such a National Strategy for Conservation will be the prioritization of action and coordination of activities, including research, to make the best use of financial and manpower resources available both nationally and internationally. Without inclusion of research within the over-all strategy, there is a danger that the limited scientific expertise will be wasted or inefficiently used, through the choice of irrelevant subjects for research, or failure to integrate the work of different specialists needed to resolve a particular problem. This has been observed particularly in the fields of reproductive biology and genetics of tropical trees in relation to their conservation and management (see e.g. Bawa and Krugman 1991; Wadsworth 1975). Studies of the effects of disturbance on the forest, and the response of the principal tree species of socio-economic value to such disturbance at various stages of their life cycle, are important for both management and conservation objectives. There is a need to coordinate action within a wide range of differing biological fields, using varying degrees of expertise, experience and technological sophistication. However too often ecological and autecological studies have lacked direct relevance to forest management activities, e.g. caused by failure to examine the comparative situations in both logged and unlogged forest. A National Strategy for Conservation could assist in channelling research activities to the areas of highest national priority. The same main elements and lines of research needed are likely to be common to many countries and at least at a regional (or eco-regional) level there may be considerable coincidence of interest in the principal species and genetic resources requiring study. More effective coordination of action, including research at the regional level has frequently been suggested but only infrequently achieved. The preparation of national research priorities in the context of the National Strategy for Conservation could assist more productive regional cooperation. A further task in formulating the National Strategy for the Conservation of Forest Genetic Resources will be the definition of appropriate institutional structures to guide and coordinate subsequent action. The exact nature of the institutional arrangements will need to be decided in accordance with existing national structures and mechanisms. Action to formulate the strategy may best be taken by the national Forestry Department but the need to secure strong inter-sectoral cooperation and high level policy review indicates close involvement with, if not actually within, central government structures. An example is the adoption by the National Planning Commission in Nepal of the National Conservation Strategy Implementation Programme in that country. This is most likely to ensure the necessary inter-sectoral 48 coordination and policy integration at a high level that is essential. However as an initial step the establishment of a National Genetic Resource Unit, as envisaged under the Tropical Forestry Action Programme, TFAP (FAO 1985b) may be sufficient, if an appropriate equivalent body does not already exist. The TFAP is the most appropriate existing international mechanism to assist the national sector review in all aspects, including the more appropriate appreciation of the capital value of the natural forest and its genetic resources. It is the failure to understand or to accept the use values and option values of genetic diversity in the national economy which has prevented the necessary investment in genetic conservation. 5.2 Management Information In addition to the undervaluation of the resource, and the associated lack of investment in its conservation and sustainable management, the principal deficiency in the approaches to management is the lack of information on the forest structure and dynamics. For effective sustainable management data are needed on the composition of each of the main forest types, the silvicultural characteristics of the principal species and of the others which may compete with them at various stages of their development, their regeneration behaviour, growth rates, and response to canopy opening, logging or silvicultural operations. In most cases even the basic knowledge of growth and yield of the principal economic species is lacing. Management f r in situ conservation of genetic resources requires much of the same basic data on ecology and autecology that underlies attempts at silviculture in the natural forest, but with greater emphasis on breeding biology and genetic structure. Nevertheless a broadly-based forest inventory, as exemplified in the Ghana case study, accompanied by surveys of NTFP and of patterns of variation in the floristic composition of the forests, is of fundamental importance for both production genetic and conservation objectives. The inventory can also provide information on species distribution patterns, which are a first step in exploring intra-specif ic variation, to be included in in situ conservation strategies. The network of Permanent Sample Plots or continuous inventory plots required for production management can also be used for e.g. phenological studies and more fundamental research, possibly involving scientists from universities, both national and international see also Section 3.3). There is frequently a - wealth of unpublished data available from a variety of sources, from herbarium sheets to expedition records and academic theses, which could contribute to the information needed on forest composition, species distribution, phenology of flowering and fruiting and so forth. Modern computer-based approaches to information management can be used to store and assist the interpretation of such data, and to guide the efficient collection of additional information to fill the most important gaps (Jenkins 1988; Davis et al 1990). The use of Geographic Information Systems (CIS) can be a powerful aid in the definition and interpretation of species distribution patterns in relation to environmental variables and vegetation types. Similarly computer-based systems for handling taxonomic data have improved the accessibility and usefulness of information on genetic diversity in respect of both taxonomic groups (e.g. ILDIS: International Legume Database and Information Service) and geographical areas. The ILDIS system incorporates information on economic importance and conservation status, as well as botanical and vernacular nomenclature, vegetation types, references to maps etc (Jury 1991). The Botanical Research and Herbarium Management System (BRAHMS) developed at the Oxford Forestry Institute, U.K. has been 49 designed specifically to handle data related to a forest genetics research programme (Filer 1991). Many organisations in tropical countries use CIS and other database management systems on microcomputer. In view of the importance of the efficient handling of data needed for in situ conservation a National Data Centre for this purpose should be a component in the national strategy. This can not only assist but also actively encourage the efficient collection of data from many sources and potential collaborators. Such a Data Centre would assist coordination of action within the country and provide a link to the regional and international data bases such as that at the World Conservation Monitoring Centre (WCMC) Cambridge (U.K.), supported by IUCN, WWF and UNEP, and to the Forestry Department of FAQ, which provides the Secretariat to the FAO Panel of Experts on Forest Gene Resources. An important activity of this latter Panel is the preparation and revision at its regular meetings, presently held approximately every 3 years, of global lists of species, by regions and operational priorities (exploration, collection, evaluation, conservation, breeding and use), most in need of action. At its Seventh Session, in December 1989, the Panel noted the need for more frequent and vigorous review of the priority lists. Close liaison and cooperation between the National Data Centres and the FAO Secretariat to the Panel would be an important element in this process. National Data Centres should also have a key role in regional cooperation, for example in the project proposed for the Conservation and Rational Use of Central Africa 1 s Forest Ecosystems, linking activities in seven countries involved. Similar cooperative action would be valuable in West Africa (Tufuor 1990b), Central America and other regions and sub-regions. 5 3 Management Systems The fundamental principle in the selection of the management system for a particular area of forest, to take account of the needs for genetic resource conservation, is that it should be based on adequate understanding of the forest ecology, and the autecology of the principal component tree species of socio-economic value. A central problem which has plagued attempts to achieve predictable results from management interventions in tropical forests has been the failure to match the operations to the actual nature and state of the forest and the site. This has been the main reason why successful silvicultural experiments, carefully executed, have often failed to translate successfully into large- scale practice over the diversity of conditions in the forest as a whole. The attempt to apply one given management system, either monocyclic or polycyclic, throughout the forests, and often to apply the same silvicultural treatments as a "blanket" prescription, without adequate attention to variation in forest types, associations and stages of development, can never be more than locally, partially and to a considerable degree accidentally successful. Except for areas where intensive "refinement" of the species composition of the forest may be consistently maintained through several felling cycles, the impact on the genetic resources of the forest is also likely to be accidental and at times irreversible. On the other hand, regeneration studies under a variety of systems in different countries have shown a general resilience of the forest composition, provided that logging is not succeeded by further catastrophic impacts on regeneration, such as fire or illegal cultivation of non-forest crops. 50 Available information on the genetics and variation of individual species, including the principal economic tree species in the tropics and sub- tropics, is generally inadequate to predict the effects of logging and silvicultural operations on their genetic resources, except in very general terms* However, as observed in previous sections, short cycles of logging and the indiscriminate use of heavy machinery, are likely to be more damaging to both species composition and reproductive systems than longer cycles (70 to 100 years)* Similarly light selective logging, leaving a well-dispersed population of seed bearers of the principal economic species, is likely to be preferable from both viewpoints to more severe disruption of the forest* The methods to reduce felling damage (topographic and stock mapping, tree marking, road planning and construction, directional felling, choice of equipment etc) can be cost-effective in economic as well as ecological terms. At the simplest level, therefore, careful planning and control of harvesting even without more elaborate management prescriptions, as practised in e.g. the Unit&s Forestieres d'Amenagement in the Peoples Republic of the Congo (FAO 1989b), is likely to serve both production and conservation objectives. Beyond this level of management the financial and manpower constraints related to the low valuation of the natural forest cause increasing divergence between the degree of information and precision needed for the management of genetic resources, in tailoring operations to the actual forest condition, and the level of investment in management permitted by the economic calculations of costs, yields and discount rates. In addition to securing more appropriate levels of revenue, as in the Ghana case study, and applying them to forest management, additional value should be attached to the conservation of genetic resources in situ, to reduce the pressures for intensification of yield. Such pressures, operating both on the length of cutting cycles and through the concentration of growth into a few currently commercial species, necessarily conflict with wider conservation objectives, concerned with species richness, ecosystem conservation and option values. However the actual level of value to be assigned to the genetic resources must be assessed, and the appropriate compromise must be struck in respect of each forest or management unit, taking account of the National Strategy for Conservation. The need for flexibility in management, according to the nature and actual state of the forest, is recognised at least in respect of timber production, in both the Malayan Uniform System, MUS (though often not implemented) and the Malaysian Selective Management System, SMS, which allows for the adaptation of management and possible silvicultural operations to the state of the populations of young trees eligible to form the next crop, after the effects of logging (FAO 1989c). The SMS prescriptions are also intended to take account of ecological considerations and these should include the conservation of genetic resources, as defined for a particular forest area, within a National Strategy for Conservation. Knowledge of the composition, ecology and silviculture of the forests, particularly in Peninsular Malaysia, is almost certainly better developed than that of any comparable area of tropical forest (e.g. Wyatt-Smith 1963; Burgess 1975; Ng 1978 and 1989; Tang 1987; Whitmore 1990; Appanah and Salleh 1991). The practice of liberation thinning to manipulate the performance of selected potential final crop trees, as practised initially in Sarawak (Hutchinson 1987), has been used with apparent success in several countries, notably in Cote d'lvoire (Maitre 1991) and in Surinam (Graaf 1991) at least on an experimental scale. Such approaches can be used to favour different species or species groups according to the prescriptions made for a particular forest or compartment. This is easier to apply in the Malaysian forests, where there is a wide range of 51 potentially marketable species, and a strong information base both for their management in the forest and their grouping for the market. However the same principle may be applied in forests less rich in economic species, if there are one or two common general purpose timbers with vigorous regeneration. This principle is being explored, for example, in Costa Rica, based on Pentaclethra macroloba to support the conservation of the populations of known, valuable species such as Cedrela odorata (Finegan pers. comm. 1991 ) 1 . The implications of the long-term trends in the international trade in timber, discussed in Chapter IV, are that the most secure market for tropical timber is likely to be for high quality species for high-priced veneers, joinery, furniture, musical instruments or other special applications. Although the bulk of such trade may be small prices are likely to be high (ITTO 1991). Highly selective logging of species used for such purposes, most of which are likely to be slow-growing and characteristic of mature-phase forest, possibly in combination with harvesting of NTFP, could be compatible with the conservation of their genetic resources and other genetic diversity best conserved in mature-phase and climax forests. At the other extreme the increasing acceptance, particularly for the domestic or regional markets, of a wider range of currently lesser-used species, including those with light, non-durable but treatable timbers, offers opportunities for other management models, also compatible with the conservation of genetic resources. These may include uniform systems, as in the strip shelterwood management system, operating on a pilot scale in the Palcazu valley in Peru (Hartshorn 1989), where early results indicate that it may be possible to retain a high proportion of the original tree species within a viable production system, with the support and involvement of local communities. The greater use of lesser-known species as an aid to conservation of the depleted resources of more valuable species, and the range of management actions which may be needed to achieve this, are illustrated in the Ghana case study. The management of production forests in support of genetic resource conservation may be most effective in close conjunction with management of Protected Areas, such as National Parks. These provide a reservoir of seed and a haven for fauna involved in seed and pollen dispersal. Although as explained earlier quite small areas of unlogged forest may serve this purpose, for example in supporting adequate populations of frugivorous birds or small mammals involved in seed dispersal, in certain cases much larger areas are required. This applies e.g. to the role of elephants in natural forest ecosystems in Africa in the ecology and distribution of timber trees such as Tieghemella heckelii (Martin 1991). The association of the Bia National Park, in Ghana, with the adjacent timber production forest is a good example of this. At the same time the production forests can provide an effective extension of the range of species found in the Protected Area. The valuable principle of incorporating protection areas, such as the Virgin Jungle Reserves (VJR) in Malaysia, within the production forests is clearly a most important component of any National Strategy for the Conservation of Forest Genetic Resources. B. Finegan, CATIE, Turrialba (Costa Rica). 52 5.4 Management Plans It is at the level of individual forests, working circles and compartments, and through the preparation of detailed management plans, or working plans, that the principles of forest management and genetic resource conservation must be put into operation. To the extent that working plans for the forest are carefully prepared and scrupulously carried out this is a further instance of potential benefit to the conservation objective through their association with timber production. As stressed earlier many aspects of sustainable forest management, from inventory, growth and regeneration studies to responsible harvesting, should simultaneously assist genetic conservation objectives. This applies particularly to the provision made to secure satisfactory regeneration, for example through the timing of exploitation in relation to mass fruiting years or the retention of adequate numbers and distribution of seed trees in the absence of adequate regrowth of desirable species, as well as the avoidance of damage to soil seed banks. The collection and validation of data on growth and the accurate forecasting of future sustainable timber yields is both essential for forest management and also of critical importance to conservation objectives. There are many examples of severe threat to the genetic resources of major timber species through over-cutting of forests on the basis of over-estimates of future levels of harvest, and through repeated "creaming" of the stand. These have sometimes been used to justify large scale industrial development which then creates an insatiable and destructive demand on the forest resources. This is an area where the interests of long-term sustainable forest management and the conservation of genetic resources closely coincide. There are other aspects of management in which the need for care to conserve genetic resources goes beyond, and may at times even be seen to conflict with production objectives. As stated earlier it is neither possible nor necessary to attempt to impose conservation measures with equal priority and stringency in all production forests; to do so would weaken the support for the general concept of genetic conservation among forest managers, and invite widespread disregard for application of related principles in practice. Therefore realistic assessment of the situation in respect of each area of forest is essential, so that firm objectives, activities and criteria for monitoring their effects can be included in each working plan. Where the practices for sustainable forest management and genetic resource conservation closely coincide the prescriptions should apply in all production forests, and it is only necessary to refer to the additional importance of careful attention, for example to the restrictions on harvesting damage to regeneration or the importance of subsequent protection from fire, in the context of the National Strategy for the Conservation of Forest Genetic Resources. In certain forest areas to be designated under the National Strategy for Conservation, additional measures will be needed. These will relate to the assessed importance of the forest in the conservation of the genetic resources of the principal species of known socio-economic value, lesser-known species, NTFP or particular vegetation or forest types. An important criterion will be the location of the forest in the overall species 1 range, and in relation to environmental conditions. The importance of conserving a wide range of provenances has been established earlier, and this may be achieved to a large extent through the application of sustainable forest management throughout the Forest Estate. However particular care may be needed where the species 53 occurs under more extreme environmental stresses, related to climate, soil, altitude etc, or at the edges of its natural range. Not only are these populations likely to be genetically distinct, through adaptation to the local environments, but they may also be more vulnerable to disruption through disturbance, which could radically reduce the long-term viability of such populations. Special consideration should be given to these marginal populations through limiting the intensity of logging, and through intensification of research into their genetic structure and reproductive biology. The other main criterion in selecting areas for special consideration relates to the production management systems in general use in particular forest formations or vegetation zones. If, as is most likely, these tend to favour pioneer or early gap-phase species, and thus to reduce the opportunities for the maintenance of satisfactory breeding populations of mature-phase species, special attention will be needed to conserve the genetic resources of this latter stage of the forest's development. The extent to which this should be done in production forests will depend on the role and de facto contribution to genetic conservation of existing Protected Area systems but it is unlikely that production forests alone could provide adequate levels of geographical and ecological coverage for species and populations concerned. Genetic resource conservation should be an important factor in the design and operation of working circles within the forest. Hitherto the allocation of areas to a protection (conservation) working circle has been based almost invariably on considerations of accessibility, steepness of slope, adverse geological or drainage condition etc. rather than their value for genetic resource conservation. The protection of special sites such as streams ides is desirable for several reasons, including conservation of biological diversity. However the siting of conservation areas such as Virgin Jungel Reserves (VJR 1 s) should attempt to conserve associations representative of the diversity and genetic resources of the forest as a whole, or of the particular forest type. This is likely to frequently conflict with immediate production objectives but the more intensive the form of management in the remainder of the forest, the more valuable the diversity and genetic resources of the VJR or other Conservation Area may be. From the viewpoint of the conservation of the most desirable economic species the practice of leaving small pockets of unlogged forest, and well dispersed seed trees, throughout the production areas is likely to be needed in addition to the allocation of a larger area, such as an entire compartment, as a VJR. The real problem commonly lies in reconciling these various objectives with the interests of the timber concessionaire. The provisions recommended to secure real interest on the part of the timber harvester in the regeneration and long-term sustainable management of the resource should also be conducive to conservation of the genetic resources of the principal economic species. However to conserve desirable option values of the genetic diversity, which have no evident, present-day market value, some additional incentives will be needed. The most promising concept may be the setting of high concession rents, together with clearly defined conservation objectives, related for example to the omission of selected "pockets" of forest from the harvesting operation. To the extent that the conservation objectives were met there could be a refund of a proportion of the fees, or conversely for gross violation the imposition of penalties under the contract. It has been suggested that ideally concessions should not be granted purely for timber 54 extraction, but as a contract between the government and the private sector to manage the forest for a range of products and benefits, including timber (Johnson et al 1991). In all aspects of forest management and conservation the failure to comply with the prescriptions and conditions set has been a frequent cause of damage to the growing stock, and particularly its capacity for regeneration. Close monitoring of operations, both to judge their conformity with prescriptions and to assess their effect relative to the declared objectives, is in the common interest of both sustainable production and genetic resource conservation. It may also be an important aspect of "forest accounting* 1 designed to monitor the condition of the resource and to evaluate the standard of management. This will include aspects involving scientific criteria, for example in regard to genetic resource conservation, as well as others related to technical and socio-economic objectives. Criteria for monitoring should be clearly set in the management plan, taking account of the objectives set under the National Strategy for the Conservation of Forest Genetic Resources. The genetic resource conservation value of a concession area, as determined under the National Strategy for Conservation, should be used in assessments of the quality of sustainable forest management. Systems and criteria for the assessment of sustainability in forest management will be needed at least in respect of forests managed for the export of timber, in accordance with the ITTO guidelines and target mentioned above (ITTO 1990). There might be positive incentives through the market for conforming to the accepted guidelines, which will include consideration of the impacts on biological diversity in the forest. In any case it is in the national interest to ensure that the objectives set under the National Strategy for Conservation in respect of forest genetic resources are achieved. In both the assessment of the genetic resource value and objectives in a particular forest and the setting of criteria by which to measure the achievement, specialist advice of forestry geneticists may be needed. This is most likely to be necessary if management action is required to conserve the genetic resources of given target species or populations, which may involve considerations of population genetics and viable population size. Other specialist advice in regard to forest ecology, the role of "keystone" species in the functioning of the forest and the genetic systems of economic species may also be required. As indicated earlier, and emphasised in most recent reviews of tropical forest management, the involvement of local communities situated in and around the forest in aspects of forest management is likely to be increasingly essential to long-term conservation objectives. The concept of joint planning and management, operated differentially according to the management objectives of the different zones, as outlined in the India case study, allows for a contractual relationship with local people, without weakening the legal status or government control over the forest land. Such a careful balance between the long-term security of the genetic resources in those areas requiring strict protection, and the sustainable development of a variety of indigenous trees and other plants in appropriate management zones, provides wide possibilities for genetic resource conservation. However to an even greater extent than the management of the natural forest for timber production alone such multiple-use systems are likely to require specific formulation in the particular ecological and socio-economic situation, and incorporation of the full range of concerns in the management plans of the forest concerned* 55 The management of areas of production forest acting also as a "buffer zone 11 to a National Park or other Protected Area was discussed in Chapter IV. The most important aspect is for the management of the two kinds of area to be fully integrated and under single effective control within one management system. This may require special administrative arrangements if primary responsibility for production forest on the one hand and the fully Protected Area on the other, come under separate ministries. However the same principle of joint planning and management, with clear objectives and management plans for each zone, should apply. The efficient management of the genetic resources of key economic species represented in a National Park or other Protected Area, may require specif ic management interventions, for example the collection of seed or other propagating materials for ex situ action, or liberation thinning to favour the development of young trees of the species targeted for conservation in situ. Such activities may (at times properly) be contrary to the general Park regulations. This could be the subject of examination under the National Strategy for Conservation and provision could then be made by amendments to laws and regulations to permit action for the conservation of forest genetic resources, with appropriate safeguards and restrictions, in accordance with the National Strategy for Conservation. International concern for the tropical forests is certain to continue and to deepen, as the forests are further reduced. This concern, and the allocation of resources to the extent needed to conserve the existing natural ecosystems and their component species, will come too late to retain them intact in many areas in many countries. However except where large areas are severely altered, for example by fire or continuous intensive cultivation, valuable genetic resources are likely to survive under sustained forest management systems; such resources are invaluable as the basis for the regeneration and/or restoration of functioning natural forests, in various forms and for a variety of uses. 56 Box 2 Management of diversity through diversity of management The Protected Area network, which must be the core of ecosystem conservation, is likely to prove increasingly inadequate to meet long-term objectives for the conservation of genetic resources. In addition to its limitations in geographical coverage, restricted often arbitrarily by other demands on land, it is likely to be further limited in terms of environmental diversity by the size, shape and management practices in each reserve area. Dynamic conservation, in a world in which future human needs and climatic conditions must be increasingly uncertain through successive generations, requires more than the attempt to prevent genetic erosion and to retain a fixed current state of diversity. It implies conserving the evolutionary process, to advance and develop ecosystem diversity, and targeted genetic resources, in directions which seem likely to prove most useful, and to retain a wide range of options to continue this process into the very distant future. Within this concept management actions, including harvesting and siMcultural activities, can be creative of genetic diversity, if they are freed from past pressures towards "blanket* uniformity, centred on a few species with minimal reinvestment of resources skills and finance. Logging operations alone, if they are neither too intensive nor too frequently repeated, but follow known principles for good practice (Dykstra and Heinrich 1992) are likely, in initially diverse tropical forests, to leave levels of complexity in species composition and plant associations which, although they may not exactly follow previous patterns of distribution, can maintain or even increase overall genetic diversity. Given adequate information on the patterns of such diversity between compartments (or in critical cases sub-compartments), as well as across larger forest management units, and within the National Forest Estate as a whole, and given the continued development of advanced systems for handling and interpreting data, future management systems could be tailored to serve dynamic conservation strategies, at the same time as they meet the needs of industry for predictable flows of raw materials. This, however, will require greater coordination of management planning at the national level, as well as the integration of conservation concerns in the management prescriptions of both production forests and the Protected Area network, under a coordinated, National Strategy for the Conservation of Forest Genetic Resources. con 57 fiox_2 continued At national level the combination of a comprehensive National Forest Inventory with a detailed botanical survey, as exemplified in the Ghana case study (see Part II), provides a good basis for such strategic planning, in terms of geographically-related data on the distribution of many tree species and plant associations. At the level of Forest Management Units diagnostic sampling by compartments, designed to collect information to support genetic conservation strategies, as well as forecasts of harvestable production of timber and other commodities, could provide much information, to adequate levels of precision (often presence or absence rather than volumetric or qualitative data), at acceptable cost (Hutchinson 1991). The Malaysian Selective Management System (FAO 1989c) allows for siMcultural operations to be tailored to the actual condition of the forest after togging, taking account of environmental objectives. It is likely that, given responsible harvesting planning and control, largely accidental influences on species composition in the advanced regeneration, superimposed on site-related differences, will leave considerable diversity within individual forest reserves, and even more over the National Forest Estate as a whole. The coordinated management of this existing diversity might go a long way to achieving the objectives of dynamic conservation, linked to productive use of the forests. However, in some areas, carefully targeted siMcultural operations might be needed to conserve the genetic resources of selected populations of trees of socio-economic importance, or to maintain given species richness or ecosystem diversity. This might involve management decisions to prolong felling cycles or leave selected trees or patches of forest unlogged, as well as some silviculture! interventions, for example to release the advanced regeneration of selected individuals of chosen species in particular areas, as part of a coordinated national strategy. The prescriptions for each area to be treated would take account of its comparative value in terms both of production and genetic conservation. Some areas, particularly in the boundary zones of production forests, might be devoted to intensive management of non-timber forest products for national use or local community benefit. This could extend the overall diversity of management systems, with the introduction of the influence of cultural diversity on forest composition, while maintaining the principle of linking conservation with productive use. The underlying principle, coordinated in application at the national level, must be the management of diversity through planned diversification of management. Introduction 59 PART II CASK STUDIES The ecological principles underlying approaches to the sustainable management of natural forests, briefly reviewed in Part I, have been long understood. The same elements, in particular the manipulation of the forest canopy in relation to regeneration and growth, form components of a wide variety of silvicultural and management systems, from strip-shelterwood clear felling operations to highly selective polycyclic management. The factors determining the degree of success, particularly in the attempt to move from the small scale research or pilot-phase operations to management of entire forests, have been fundamentally economic, social and political forces, deciding the level of capability for management, including the adequacy of its information base, and its freedom of action to interpret and to apply ecological principles. Similarly the implications of genetic systems and genetic structure for the conservation of intraspecif ic variation are sufficiently established even though detailed knowledge of their expression in the individual species is almost entirely lacking. A central activity in any strategy for genetic resource conservation must be the gathering of that information, in respect both of patterns of variation across a species range and the actual mating system, seed dispersal mechanisms etc operating within the populations. The almost incalculable number and variety of investigations needed to cover all species of actual or potential economic importance requires rigorous setting of objectives and priorities for each forest area or management unit, and in accordance with the national Forest Genetic Resource Conservation Strategy. In both respects, of management and conservation, every nation and to a degree every forest, is unique. The choice of strategy will depend as much on socio-economic forces around the forest, and on national policies for land use and for development both within the forestry sector and in other sectors, as on the basic scientific principles for management and conservation. Key factors will include the relative importance of the forests for the export of timber and for the needs of local communities; their existing extent and distribution in relation to population density and demands for land; the importance of their influence on local and regional environments; the national and global value of their genetic resources, in terms both of principal economic species and of biological diversity in general; and the nature and degree of threat to valuable ecosystems, species, genetically distinct populations and genes. The following three case studies have been chosen to illustrate the principles outlined in Part I of this publication, in three contrasting situations, and the ways in which sustainable forest management and, more recently, genetic conservation, are being approached in each country. The starting point in all three is seen to be considerations of national policy and land-use but the nature and balance of the activities in progress, and the implication for future action and research vary greatly between them. The ideal basis for the development of a National Strategy for the Conservation of Forest Genetic Resources is a review of the forestry sector, in its wider context with other sectors, and the preparation of a national forestry action plan. This in effect has been the case in Ghana, which was one of the very first countries to be involved in the development of a National Plan, within the framework of the Tropical Forestry Action Programme (see e.g. FAO 1985b). 60 The Ghana case study illustrates many aspects of the common problem of harmonizing genetic resource conservation with the utilization of existing resources for development. Timber is important to the Ghanaian economy and the forests 1 environmental role is important in sustaining agricultural production. At the same time population pressures are rising and demands from local communities around the forests are increasing, relative to the diminishing resource. The resources of the principal economic timber species have been depleted to a critical level in some cases. Responsive action by the government has provided the information base for revised policies and strategies now being introduced and implemented. Within these the conservation of forest genetic resources is seen to be largely dependent on national action to increase the diversity of management in both the forest and the forest industry sectors, and this action to depend in turn on international trading conditions. In view of its wide applicability this study has been treated at length, to illustrate the range and interdependence of actions. In contrast the Brazil case study focuses on a single large region within the country, within which the forest biome and its genetic resources are still largely intact. As a result of this the possibilities exist for integrated programmes of conservation and development, involving planned coordination of action in both protected areas and production forests, and multiple-use forest management, with involvement of local people. The major problems and challenges relate to the scale of action needed, and the complexities of coordinating programmes involving many administrative divisions and possibly even more technical and scientific bodies, generating large quantities of data, as contributions to what should be a single coherent strategy for the region. The Indian case study focuses down even further to a single unique area of natural forest. In contrast to most other tropical countries the forest has been under planned protection and management for well over 100 years but now as the result of increased population pressures, the survival of the natural ecosystem and of viable breeding populations of key species have been judged to depend on a new and more active approach to the integrated development of the forest resources, with involvement of communities within and around the forest. Both this approach and the scientific surveys and research needed to provide the basis for efficient conservation strategies, are essential elements in all three case studies. All three case studies illustrate the emerging principle that the conservation of valuable genetic diversity depends on a greater diversity of forest management approaches, specifically adapted to the existing diversity of ecosystems, species and demands on the forest. They also demonstrate the universal need for the rapid and efficient collection and interpretation of essential information on the composition and dynamics of the forests, as the basis for more exact systems of sustainable forest management compatible with genetic conservation concerns. To complement the above, the case study of Cordia alliodora (Appendix 1) describes detailed studies undertaken on the biology and "functioning" of an important species of present-day economic importance, to illustrate the information needs in this regard and the utilization of data collected in development or refinement of genetic conservation strategies. 61 GHANA 6-1 The Economy The forestry sector in Ghana is the third highest source of export earnings after cocoa and minerals, and accounts for 5 to 6% of the gross domestic product (GDP), providing employment to about 70 000 people (Asabere 1987). Export earnings in 1987 represented 11.4% of the total, with cocoa contributing by far the greatest proportion (60%). However timber is seen to play a key role in the economy (Frimpong-Mensah 1989). Forests cover about 11.3 million hectares or 48% of the total area of the country (Ghartey 1990), and in addition to meeting all its timber requirements they supply some 75% of energy needs. The population of 14 million people is estimated to be increasing at 3% a year and population densities range from 17 persons per square kilometre to over 150, with an average of 63 (WRI 1990). With a GNP per capita of US$ 390 the country has a low-income economy, heavily dependent on the agricultural sector. The high forest zone, roughly one third of the country, supports two thirds of the population. The forest itself, which covered virtually all the 8.2 million hectares of this zone at the beginning of the century, has been reduced to about 1.7 million ha. The principal cause of deforestation has been demand for agricultural land and to a considerable extent the successful retention of a national estate of reserved forests, in the face of such massive conversion, is the result of sound land-use policies. However there is now relatively little high forest left outside gazetted forest reserves and with the population doubling in 22 years, and projected to reach a total of 37 million by the year 2 025 (WRI 1990), the area of land under agriculture is still increasing. 6.2 The Environment The protective role of the forests was recognised from the outset, and particularly its influence on local climate and hydrology; and the dependence of the major agricultural crops, especially cocoa, on the maintenance of the forest cover (Gent 1929, in FAO 1985a). While the principal concern was for the agricultural activities within the high forest zone there is now increased understanding of the wider influence of the moist forests on the savanna zones to the north, covering 66% of the country. These include four vegetation types, progressively drier, from the Derived Savanna, and Southern Guinea Savanna zones, to the Northern Guinea and Sudan Savanna. The rainfall ranges between 800 and 1,200 mm/ year but has shown a decreasing trend over the past 30 years. Most significant are the length and severity of the dry season and the uncertainty and erratic nature of the rains. Since the main rain-bearing winds come from the south west, to displace the severely dry north eastern winds, the rainfall in the savanna zones is influenced by the state of the vegetation in the south of the country, and particularly the forest cover, which through transpiration effectively recycles the precipitation northwards. Although the West African data are less complete the situation is believed to be similar to that in the Brazilian Amazon in regard to the links between forest and the hydrological cycle (Salati 1987). Increasing deforestation in the northern part of the country, with annual fires, has already caused substantial ecological degradation and increasing population pressure seems certain to aggravate the problems. The possibility of global warming is a further reason to retain as far as possible the stabilising effect of the remaining forest cover. 62 6.3 Diversity The rain forest flora, both herbs and trees, changes clinally from west to east across the country over a distance of 300 km, with increasingly dry climate, and changes in soil type (Whitmore 1990). Ghana's biogeographic affinities are Guinea-Congo lian in the southwest, Sudanian in the north, with the important division of the high forest zone by the "Dahomey Gap" to the east, separating to some extent the forests of the Upper Guinea from those of the Lower Guinea. Although the forests are not rich in biological diversity or end em ism, in comparison for example with those of Cameroon, over 2^100 plant species have been recorded in the forests (Hall and Swaine 1981), including about 680 tree species (Hawthorn 1990a). 6.4 Management for timber production As a result of an initial survey of the sector in 1908, following the enactment of a Timber Protection Ordinance the previous year, the reservation of selected areas of forest was proposed but met with opposition from the local chiefs (FAO 1985a; Asabere 1987). However by 1939 some 1.6 million ha had been reserved in the high forest zone, surveyed, demarcated and constituted under the Forest Ordinance. About 70 per cent is made up of production reserves, which are the source of future supplies of timber, as the unreserved forests still remaining are not expected to survive as timber production areas beyond the end of the century. The remainder of the permanent forest estate is reserved for protective purposes, a proportion of this area being made up of plantations. There are 13 wildlife reserves covering a total of nearly 1.2 million ha, of which four are in the high forest zone and nine in the savanna. Within the high forest zone the areas reserved strictly for protective purposes are those which are most inaccessible and which cannot be easily or safely logged (Tufuor 1990a), and are not therefore necessarily representative of the range of plant communities in the zone. The tropical high forest zone, which is the main source of timber, shows a wide range of variation 'in plant communities, which have been broadly grouped into four ecological types: the Wet Evergreen Forest (Cynometra - Lophira - Tarrietia association), the Moist Evergreen Forest, the Moist Semi- Deciduous Forest and the Dry Semi-Deciduous Forest (Hall and Swaine 1981). Three main approaches to the management of the forests have been made over the past 40 to 50 years, namely the Tropical Shelterwood System (TSS), enrichment planting, and the Modified Selection System (MSS) which was introduced in 1956, and has been the main system employed throughout most of the zone, particularly in the Moist Semi-Deciduous Forests, until 1970. The main features of the MSS were stock-mapping of all "economic" trees over 7 ft (2.1 m) girth and selective felling on a 25-year cycle. The yield was regulated by area, with minimum girth limits set according to the species class. Felling was based on the stock maps, starting with the largest trees and going down until the prescribed yield had been achieved. However trees below the minimum girth limits were never permitted to be felled, even if there proved to be insufficient trees above the limits to make up the prescribed yield. Moreover there was a stipulation that the remaining stock should be left well distributed over the compartment, so that in some cases a tree well above the limit might be left as the only one in that part of the area (Asabere 1987). This relatively simple system of control, although based 63 on incomplete data on forest dynamics, was reasonably solid and ecologically sustainable. In the late 1960's the felling cycle of 25 years was increasingly criticised by timber concessionaires as being too long and resulting in the accumulation of over-mature trees subject to decay. As a result it was decided in 1970 to permit the felling of all economic trees above 3.4m girth or above 2.1 m girth (depending on species class) and to reduce the felling cycle to 15 years. The assumption that all trees over these dimensions were over-mature was not based on systematic studies, and such an investigation later revealed that the trees in the highest classes with girths of 3.4 to 4.0 m were far from being over-mature. Rather than reducing waste of timber it appears that the blanket application of salvage felling throughout the production forests, irrespective of stocking, encouraged more wasteful and less efficient practices, with large numbers of logs being left to rot in the forest (Asabere 1987). The greater intensity and frequency of logging, concentrated on the few preferred species, with the removal of all the largest trees (whereas previously some had been allowed to remain) was bound in time to have an increasingly dysgenic effect on the populations. At the same time the tendency to increase the size of gaps in the canopy, as the result of removal of all large "economic" trees regardless of their proximity or situation, must lead to greater structural changes, and to a stronger shift in species composition towards the pioneer species, and with greater danger of dense climber tangles developing. There would therefore be a predictable adverse impact both on the genetic resources of the species being logged and in terms of overall diversity, as the prescriptions applied throughout the forest. From the mid-1 970 f s to the early 1980's the forestry sector suffered a marked decline both in the production of logs and in the total value of timber exports. This accompanied general economic problems, related to the overvalued currency (cedi) and linked to the poor state of logging, transport and sawmilling equipment, lack of spare parts, inadequate investment in roads and infrastructure generally, and inadequate funding of the Forestry Department. From 1983 a series of macro-economic and sectoral reforms, followed by renewed investment in equipment and infrastructure, assisted by multilateral and bilateral funding, led to a progressive recovery in log production and export revenue, which had been substantially achieved by the end of 1987 (Frimpong-Mensah 1989). With the restoration of export-led logging and milling operations there was a danger that exploitation would override concerns for sustainable management, centred on sustained yield of timber. The initiation of the Tropical Forestry Action Plan (TFAP) in 1985 had promoted the concept of joint sector reviews, with assistance from several donor agencies, and in Hay 1986 a TFAP review was started in Ghana, led by the World Bank, with direct collaboration and assistance from FAO, CIDA (Canada) and ODA (U.K.). Following the studies, project preparation, appraisals and negotiations the Forest Resource Management Project was initiated in March 1989, with funding from the World Bank, Denmark and the U.K. This provided the opportunity for the Government of Ghana to undertake a comprehensive review of the forestry sector, linking conservation to production objectives, and related policy considerations. 64 6.5 Policy; linking production and conservation The 1948 forest policy provided a broad basis for both production and conservation objectives, if interpreted in that spirit. It included specific reference to the maintenance of the permanent forest estate, the protection of environmental values, management for sustained yield, research with emphasis on ecology and silviculture, public education, staff training and optimal land use (Kese 1989). Existing legislation (principally the Forests Ordinance (1927) and associated Forest Reserves Regulations; the Concessions Act (1962); the Timber Operations Decree (1972); the Forest Protection Decree (1974); the Trees and Timber Decree (1974) and the Timber Industry and Ghana Timber Marketing Board Amendment Decree (1977) with some provisions of the investment Policy Decree) provided a suitable basis for policy application. The success of the original forest reservation policy was strikingly revealed by Landsat imagery which showed clearly defined reserve boundaries which had been generally respected, in contrast to the surrounding deforestation (Hawthorn 1989). Revision of the National Forest Policy, completed early in 1992, further strengthened environmental provisions relating to soil and water resources, with specific references to the conservation of flora, fauna and biological diversity, and to sustained yield of non-timber as well as timber resources. This implied due attention to in situ conservation of genetic resources . At the same time as the strengthened concern for the environment and biological diversity there are the increasing demands on the economy and on land resources from the expanding population. Revised and strengthened inter- sectoral policies, with appropriate sectoral strategies, are essential to secure the appropriate reconciliation of conflicting production and conservation demands. Grut (1989) from a study of the forestry sector in Ghana and in Guinea, has shown that on an analysis of timber production alone, the most profitable option would be to take out all merchantable timber (i.e. to well below the normal minimum diameter limits set by management) in one felling. Calculations used to reach this conclusion did not consider the substantial, but usually non-market, values of non-timber forest products, and the severe impacts on forest structure, ecology and species composition that might result. 6.6 Non-Timber Forest Products (NTFP) The role of non-timber forest products (NTFP) is critically important to the long-term conservation of the forest resources and their genetic diversity. A detailed study undertaken in the context of the forest Resource Management Project in Ghana (Falconer 1991) shows that NTFP's form the main link between the reserved forests and the communities living immediately around it. The study revealed the importance of a wide variety of products in human nutrition, medicines, materials for house building, household and agricultural implements, fuelwood, fodder, and trade and processing activities based on NTFP's. For house building materials the most important qualities sought were durability and insect resistance, for the principal building poles in particular. These qualities are commonly associated with denser and harder wood more frequently found in slower-growing species (usually non-pioneer species). It was evident that many chiefs and elders in the communities linked the value of the forest strongly to their traditional collection of materials for house building. 65 Forests were also found to be very highly valued for medicines. All the people interviewed used plant-based medicines and 80% relied on them exclusively. Most of those commonly used were gathered around the village and in fallow areas of secondary forest, rather than in the forest reserve. Nevertheless the forest itself was seen as an important source of medicines by the majority of those interviewed. For many households the gathering, processing and trading of forest products provided an important source of supplemental income, especially in the periods of less agricultural activity or when there is an acute need for cash. For some of them NTFP's provided the main source of income, either directly or as a source of materials used for production equipment and raw materials used in off-farm processing activities. Forests were the main source of traded NTFP's in the region although in some areas and for some products fallow lands were important sources of supply. In many communities people claimed that increased land clearance for agriculture, increased bush fires and degradation of the fallow environments (secondary forests) had depleted the NTFP resources so that collectors and processors had come to rely more heavily on the forest reserves. For many products, including bushmeat, the study revealed the importance of the fallow areas as sources of supply. The degradation of the secondary forest was seen as a great problem by many communities and more and more the forests remaining in reserves are likely to become important sources of NTFP's. Flexible forest management systems which can meet local needs as well as those of the timber industry offer the best prospect for long-term security of a wide range of tree and plant species. There is also scope for the propagation and cultivation of some species in buffer zones around the forest, and provided that the source of seed or selected clonal material is taken from the local populations in the forest this could be an acceptable aspect of in situ conservation for the species concerned. 6.7 Forest Revenue Systems A major problem in conservation and sustainable management in many countries has been the apparently low rate of economic return from the natural tropical forest (Mergen and Vincent 1987) despite the evident real value of the resource (Leslie 1987). This is frequently due to failure to capture appropriate levels of revenue from timber (Repetto and Gillis 1988). Grut (1989) shows this clearly in respect of Ghana: both the levels of royalty set and the effectiveness of collection of the fees due have hitherto been extremely low. Even with the very low level of stumpage demanded the revenue actually collected was only one sixth of the amount due (Grut 1989). The very low level of royalties on timber in Ghana, and the need for substantial increases, particularly for the most valuable and marketable species, had been recognised earlier (FAO 1985a). Substantial increases have now been introduced and others are expected to follow. Assuming that these are also implemented to raise average stumpage fees to about 10% of weighted average FOB log export value by 1994, with increased effectiveness in revenue collection rate to 50% by the same date, revenue would be sufficient to meet the entire costs of the Forestry Department, and not only the costs of management of the production forests (Grut 1989). 66 Both the size and length of tenure of timber concessions are key factors in determining the interest and capability of the concessionaire in sustain** able management. The fragmentation of concessions in Ghana was criticised earlier (Asabere 1987) and suggestions made in the context of the Forest Resource Management Project to set a minimum size of 10 000 ha and duration of 50 years are under consideration by the government authorities. Related to these suggestions are consideration of more strongly market-determined allocation of concessions, to secure the best value possible, and a more significant level of concession rent. The latter could be particularly influential in securing conservation objectives, including in situ conserva- tion of genetic resources. It has been suggested, for example, that substantial concession rents would be set, based on a formula derived from weighted FOB values, less costs of logging etc and an adequate level of profit. The concessionaire could subsequently claim back a proportion of the fees paid, in respect of costs incurred in sustainable management and con- servation activities. The latter would then be seen by the concessionaire as a revenue-earning opportunity (Grut 1990). Such reforms of revenue levels and collection can also potentially improve recovery and reduce waste of timber, both in the forest and in industrial processing (Asabere 1987; Chachu 1989). The Government of Ghana has set up a special Committee to evaluate measures to improve forest management and productivity in the timber industry, and the Committee is expected to review the size and related aspects of timber concessions. Ghana had earlier introduced differential royalty rates for 50 timbers. The new royalty fee structure being set involves considerably higher increases in respect of the species in highest demand, and therefore in danger of depletion, than for those currently neglected. This has important implica- tions for the conservation of the genetic resources of the main economic species. However, such action cannot be effective in isolation, as many lesser-known species have lower natural durability and therefore require special care both to ensure that the logs are not left lying in the forest and for appropriate seasoning and preservation treatments. Others which may have more durable but hard and possibly siliceous timber require special processing equipment. The securing of an adequate market return for such lesser-known species may be assisted by local processing (Asabere 1987; Ofosu-Asiedu and Ampong 1990; Parant 1990) for which investment in equipment, infrastructure and training may be needed. Ghana is promoting further processing of timber, and log exports as a proportion of total export volumes have been falling from 63Z in 1988, to 53% in 1989 and 42% in 1990. However in 1989 less than 6% by value of Ghana's forest products exports were processed beyond air-dried timber (ITTO 1990). There are also important forest management and silvicultural consider- ations, as well as ecological concerns, related to the promotion of harvesting of lesser-known species, and these are reviewed below. However the implica- tions of the results of the forest inventory carried out clearly indicate the need to make maximum sustainable use of the neglected timber species, while reducing pressure on the stocks and the genetic resources of the principal economic species. This is in line with the new recognition that harvesting convenience and the demands of the wood industry alone should not be allowed to determine forest management (Chachu 1989) but that royalty rates, girth limits and other objectives and intensity controls should reflect the ecological differences between species in their regeneration and growth, as well as ultimate timber values (Hawthorn 1990b). There is already abundant evidence of the impermanence of classifications based solely on market acceptability and valuation of timber. 67 6.8 Forest Inventory The first survey of forest resources in Ghana was made in 1908 and the first National Forest Inventory in 1947, covering 1 290 square miles of forest (Logan 1947). The main focus of this and subsequent inventories in the period 1952 to 1973 was the then currently merchantable species, of which only 26 "economic 11 species were listed compared with 334 "secondary timbers" (FAO 1985a). In 1985 a new National Forest Inventory was started, primarily also to provide an estimate of the total commercial log volume, in view of concerns that the demands of the rehabilitated timber industry would exceed the sustainable supply from the forest. However broader aims were included, to supply the wider information needed for sustainable management, and to assess the biological productivity and ecological status of the forest, including the collection of data on non-timber forest products, and on non-woody plants and fauna (Adlard 1990). To meet the primary objective stratified random sampling was done over 546 000 ha in 43 forest reserves. Stratification was based on the ecological survey and classification by Hall and Swaine (1981). In addition to measurements of tree diameters, for volume estimations, assessments were made of form and quality, together with crown classification, observations on logging, burning etc. About 420 tree species were recorded, and grouped into three classes according to their current market acceptability, size and frequency of occur rence:- Class I: Species exported from Ghana at least once during the period 1973- 88, including all the main economic species plus some lesser- known species being actively promoted for export (total 66 species) . Class II: Species attaining 70 cm diameter (marketable size) and present at a frequency of at least 1 tree per km 2 , although not previously exported (total 58 species). Class III: All remaining species, not considered to have potential for timber production. All the data were stored on computer and constitute an important database on species occurrence and distribution (Wong 1989). The gross national standing volume of timber of exploitable size was estimated as 102 million m 3 , concentrated mainly in the Class I species. However the bulk of this volume was found to be made up by the less desirable species, while the volume of the traditional economic species was very limited and heavily dependent on one mainstay, Triplochiton scleroxylon, which although in constant demand is not intrinsically of high value. It appears that current exploitation of most "economic" species is unsustainable and that they have a very limited resource life (i.e. the period during which the existing supply of trees of exploitable size will be exhausted). The valuable Meliaceae (Entandrophragma angolense, E. cylindricum, E. utile, Khaya grandifoliola, K. ivorensis, K. anthotheca), as well as Milicia (syn. Chlorophora) excel sa, are likely to be exhausted within 2 to 3 decades at current levels of felling. Among the most valuable species Pericopsis (syn. Afrormosia) elata is estimated to have a resource life of 3 years or less (Alder 1989). The calculation of the species resource life can only be approximate, based on the division of the existing resource by the rate of 68 extraction, and if with a diminishing resource the rate of extraction were allowed to rise, the resource life would be even less. The implication is clear: the future of a viable timber industry in Ghana is dependent on a substantial and rapid move into the marketing of presently under-utilised species, in Class I. This could be achieved by the promotion of perhaps 10 to 15% of the 50 odd presently under-utilised species listed, with maximum reliance on those well represented in the forest, alongside the sustained logging of Terminalia excelsa. If this were done with effective monitoring and controls to limit the logging of the principal economic species to sustainable levels, to allow their populations to recover between logging cycles, there need be no long-term danger to the genetic resources of these species. However their future existence is clearly dependent on positive action in this regard. Growth data for the estimation of increment rates were based on measurement of about 11 000 trees in 256 permanent sample plots. The development of a forest growth simulation model (GHAFOSIM) for Ghana forests is in progress (Alder 1989; 1990) and will be assisted by the progressive accumulation of data from the greatly expanded number of permanent sample plots, in which all trees of all species over 10 cm diameter (dbh) are measured. This will provide a much more complete understanding of the forest dynamics. It is evident that no single management strategy will serve adequately for all the forests but that for efficient linking of production to conservation the management of individual reserves or concessions will need to be decided separately, within an overall picture of the genetic diversity in the forests as a whole. 6.9 Setting Priorities Conservation of the principal economic tree species is clearly a high priority, and is seen to be dependent on developing the markets for presently lesser-used species. However conservation of intra-specific variation in these species is also of greatest concern, including the maintenance of viable populations of genetically distinct provenances. In the absence of direct evidence of provenance differences from field trials and/or isozyme studies the best guide is likely to be patterns of environmental variation. These may be seen directly from environmental variables or indirectly from patterns of vegetation. The way in which population-level variation in individual species reflects patterns in the community as a whole is apparent in many cases (Okafor 1975; Hall and Swaine 1981). The assumption that this occurs is recognised in the botanical survey being carried out in association with the on-going National Forest Inventory (Hawthorn 1991). The aim of the botanical survey is to develop a computerised database of the distribution of vascular plants in the forests of Ghana, thereby securing clearer definition of the substantial f loristic variation within each of the forest zones recognised by Hall and Swaine (1981). This will reveal patterns of intra-specific variation at local as well as national level, and the relationship between them, and thereby assist in designing strategies for the conservation of economic plant species. The survey gives particular attention to patterns of distribution of rare plants and trees as a direct aid to their conservation. The importance of combining taxonomic, ecological and choro logical information is recognised in the study, in order to set priorities for conservation of selected areas. 69 Preliminary analysis (Hawthorn 1991) has defined three sets of conditions under which rare plants are most likely to be found. One of these is among the dry forest species in or near the coastal or Afram plains. Such West African Dry Forest is now of very limited distribution and some species are unknown elsewhere except for occurrences in East Africa, while others are endemic to Ghana. Special consideration should also be given to fire prone reserves in the Dry Semi-Deciduous zone, where the forest plants are especially at risk. The threat of fire to the genetic resources is particularly acute where the population of mature seed-bearing individuals has been severely reduced, seed years may be infrequent and a fire following seed fall or seedling establishment can be catastrophic. This may be the case with Pericopsis elata, particularly at the fringe of the species 1 range. The botanical survey is also developing an objective rating system for the classification of plant species in terms of priority for conservation. This will be used to assist the selection of areas, for example within production forest reserves, where forest management should take particular account of the importance, and the nature, of conservation objectives. The classification being developed will be based essentially in the information on distribution, derived from the earlier surveys of Hall and Swaine (1981), together with the more general observations in the Flora of West Tropical Africa for the wider picture, the substantial herbarium collections at Legon and Kew (Royal Botanic Gardens) and the new data from the botanical survey. This will provide a basis for highlighting the "genetic hot spots" in Ghana's forests, as concerns ecosystems, species and genetic resources of species targeted for conservation. Most categories in the rating system are concerned with degrees of rarity, both within Ghana and in the wider contexts of Upper Guinea (the region within which the contribution of the Ghanaian flora is likely to be most significant), West Africa and Africa as a whole. However for the conservation in situ of the genetic resources of species of known economic importance a separate category is recognised. This includes species that may be fairly common or widespread but heavily exploited, and requiring individual attention and control, particularly in regard to the conservation of intraspecific variation. This category includes all the heavily utilized timber species, canes (rattans) and other species of known economic value (e.g. Thaumatococcus daniellii, a source of sweetening agents and other potentially commercial organic compounds). Under this classification species are allocated a "Red Star" rating if, as in the case of sought-after timber trees, a significant proportion of the mature individuals are affected by exploitation pressures (e.g. species with a short resource life, as listed above); or if it is advisable to "flag" the occurrences of a species in the forest for conservation of specific provenances. An additional and immediate practical outcome of the botanical survey will be a report on the current status of existing Protected Areas and of the degree of protection afforded within the production reserves, with recommendations regarding forest management in these latter ones, particularly in the more sensitive areas and those of exceptional biological value. This will assist the development of comprehensive approaches to reconcile the production and conservation objectives, within the Forest Management Units (FMU) now being established. 70 6.10 Management and Harvesting The Modified Selection System, described briefly above, had apparently succeeded in maintaining satisfactory levels of timber yield but at the expense of the stocking of the main economic species. Data from the on-going inventory and intensified permanent sample plot network will permit the future development of detailed and flexible management plans separately for each management unit (single reserve or large concession area) using simulation models to be derived from continuation of the work initiated on the Ghana Forest Simulation Model (GHAFOSIM). Such an approach, capable of incorporating information from the botanical survey, and ecological studies on regeneration in relation to logging intensity, pattern and methods, offers the best prospect for combining genetic conservation with production, both at the local and the national level. The possibilities for effective integration of conservation and production objectives in forest management plans have been strengthened by the introduction of Forest Management Units (FMU), which group the existing production reserves into units of about 50 000 hectares. Each FMU will be subject to a management plan, incorporating the recommendations on the conservation of biological diversity and genetic resources, emerging from the botanical survey. As an interim measure a forty year felling cycle has been imposed and stock surveys and stock mapping of all compartments are being undertaken to determine the permissible yield. This is calculated for each species in each compartment by a simple formula which takes into account the number of trees above the minimum felling limit, and those in the diameter class immediately below, which will form the next crop. The formula also allows for the 30% retention of the mature crop and 20% mortality in the residual population intended for the next felling cycle (Ghartey 1990). Provided that logging controls are enforced to prevent unnecessary damage to advance growth and to the conditions for regeneration (see below), and that sufficient seed bearers of the important economic species are left where needed, this interim regime should be sufficiently conservative to prevent further degradation of the genetic resources of the forest. However this is ultimately dependent on action to promote the lessor-used species, in order to maintain acceptable levels of harvest while conserving the remaining genetic resources of the most valuable and most depleted populations. In November 1990 the Ghanaian authorities imposed Forest Improvement Levies on six species of log exports, complementing export bans already in place on 19 other species, including the valuable Meliaceae. The levies are intended to conserve the species concerned and to provide logs for domestic processing (ITTO 1991). The success of these moves and of plans for greater reliance on the lesser-used species will have a major influence on the possibilities for future management. The impact of increased logging on the genetic resources of presently under-utilised species will also need to be considered in respect of each forest reserve and management unit, in the preparation of management plans. However the inventory shows that overall over 40% of the potential timber harvest from species in Class I (i.e. those that have been exported at all) is made up by four species still in abundant supply in the forest (Ghartey 1989). A strategy based on the promotion of these up to the limits of the annual allowable cut, together with lesser contributions from some other presently neglected species, should be consistent with more balanced management of the total genetic resources of the forest, provided that satisfactory conditions for regeneration are maintained. 71 6.11 Regeneration and Silviculture Under the TSS system introduced in 1946 the attempt was made to influence regeneration of the relatively few desirable species by manipulation of the canopy and cleaning operations but it was judged unsuccessful and abandoned in the mid-1960 ! s. Since then, apart from attempts at artificial regeneration through enrichment planting, the attempts to influence the crop through silvicultural operations under the MSS were directed to favouring the young trees in the 1 to 5 ft (0.3 to 1.5 m) girth class by improvement thinnings (Asabere 1987). Nevertheless from the viewpoint of long-term sustainable management, and the conservation of the genetic resources of the principal economic species, an understanding of the effect of logging on the species composition of the regeneration is essential. In association with the National Forest Inventory, regeneration studies were undertaken in logged and adjacent unlogged forest within the Bia South reserve (Hawthorn 1990b), to assess the various effects of logging operations on the pattern of regeneration of the tree species. A secondary objective was to describe the state of the vegetation as a whole following logging, as background information on the biotic environment of the developing trees. Since the forest had not been exploited previously, and the only control exercised had been through the minimum girth limit, with all trees above that limit eligible for removal, the logging intensity in some compartments had been very high. However this effect was mainly confined to the most desirable species such as Khaya ivorensis and Entandrophragma utile. Large trees of less desirable species had been left. Logging had been done in the three years immediately prior to the study. The main subdivisions of the species "guilds 11 recognised were between true pioneer species and under storey pioneers (small trees and shrubs) on the one hand, and between non-pioneer light demanders and shade-bearers on the other, although it was recognised that these divisions may be arbitrary separations in what is a continuum of preferences. The report on the study (Hawthorn 1990b) is detailed and reveals a complex situation. Nevertheless some clear conclusions were drawn regarding the effects on early seedling regeneration e.g:-4 (i) the overall species diversity is not likely to be adversely affected by well-controlled logging, although the balance of species is expected to change unless special steps are taken to prevent this. The trend will be declining levels of regeneration of some valuable (non-pioneer) timbers (e.g. Entandrophragma spp, Khaya spp) and greater relative abundance of others (e.g. Triplochiton scleroxylon and Terminalia spp); (ii) large canopy gaps show very poor regeneration of preferred species and low species diversity; (iii) regeneration of most timber species, particularly non-pioneers, requires the retention of parent trees evenly dispersed throughout the forest, for example in small unlogged pockets within logged compartments; (iv) small gaps and skid trails helps to add to overall species diversity. 72 Clear general guidance can therefore be given regarding strict logging controls, to keep road widths and the area of loading bays to a minimum, to avoid the creation of large canopy gaps and to ensure the retention of seed bearers evenly spread through the forest. To avoid damage to regeneration logging should be restricted to a single short period. With such provisions there seems little danger of any sudden loss of genetic resources of the main economic species, and good possibilities to retain a high level of species diversity at least of a wide range of Class I timbers. The most serious threat to genetic resources is undoubtedly fire, and associated studies of fire damage suggest that logging, especially in the drier forest types, may render the forest more liable to fire (Hawthorn 1989). The recommendations for controlling the intensity of logging are therefore even more important in fire prone areas. Although the results of these studies are both valuable and encouraging more widespread and comprehensive data will need to be gathered from the expanded series of permanent sample plots, which could also be the sites for more detailed studies in population dynamics, pollination and seed dispersal systems and agents, and other ecological processes (Adlard 1990). These could provide a scientific basis for possible future silvicultural action to influence the regeneration and development of selected populations maintaining high levels of genetic diversity. 6.12 Reproductive Biology While much valuable information on genetic systems may be gained from well-documented pheno logical observations, more detailed and systematic studies of the key economic species (e.g. Entandrophragma spp, Khaya spp, Milicia excelsa, Pericopsis elata) which are in urgent need of action to conserve their constituent populations, are strongly advisable. Data on reproductive biology, used in conjunction with information on species distribution and variation patterns in individual species, can assist in determining both the siting of in situ conservation areas and sampling strategies for complementary ex situ conservation. The principles involved are summarised in e.g. Bawa and Krugman (1991). Their application is also illustrated in the case study on the reproductive biology and genetics of Cordia alliodora, in Appendix 1 . 6.13 Integration and Security The database developments associated with the forest inventory and the botanical survey provide a framework for planning the integrated management of genetic resources, both within fully protected areas and in production reserves. They also provide a basis for monitoring future changes in the forest under the influence of prescribed harvesting and management regimes. However there will be a need for high-level decisions on the balance to be struck between timber exploitation and the conservation of biological diversity, to ensure that the necessary "trade off 11 between short-term financial costs of conservation (including revenue and profits which may be foregone) and the longer-term social, economic and environmental benefits are settled in the national interest, and where appropriate with international assistance. This is likely to be needed particularly in enabling the essential changes in the timber trade, to largely replace current dependence on the few principal economic species, with very limited resource life, by the promotion of lesser-used species, with greater reliance on local processing and the export of components and finished products. Action in these areas of 73 international trade, linked to investments in local provision of physical capacity and human resource development, will determine the feasibility of forest management systems linking sustainable production and genetic resource conservation. BRAZIL; THE AMAZON FORESTS The Amazon basin as a whole contains the largest single extent of tropical forest in the world, of which approximately 60% (some 350 million ha) lies within Brazil (Dubois 1991). About 280 million ha are closed forests with the potential to produce industrial timber (FAO 1981). At the same time the forests are almost certainly the richest in the world in terms of total biological diversity, and cover a wide range of forest types. The possible contribution of the potential production forests to the conservation of such diversity and of forest genetic resources is therefore very high. According to Cochrane and Sanchez (1982) some 93% of the soils of the Amazon Basin have major fertility deficiencies and although perhaps half of this area is underlain by soils of good physical properties, and might be capable of sustainable agriculture or pasture use, given substantial inputs of nutrients and under careful management, they are susceptible to rapid degradation after clearance of existing forest. In addition studies have indicated the interdependence of the local hydrological cycle and the forest cover (Salati 1987) and it has been suggested that to safeguard local and regional climate, until a fuller understanding of this relationship is achieved, about 80% of the area should be kept under forest cover (Prance 1990). Despite some recent immigration into the Amazon region from elsewhere in Brazil the human population is still low relative to the total area, with over 40% of the Brazilian population concentrated in some 10% of the national territory in the south eastern region. Nevertheless the pressures on currently forested land in Amazonia will inevitably increase and it is essential to develop sustainable systems of land use which will be compatible with the need to retain a forest cover over substantial areas for its environmental influences. The opportunity still exists to develop patterns of land use which optimise the sustainable management of natural forest, with the simultaneous conservation of the rich biological diversity of the region. This requires the integrated planning and development of the systems of fully protected areas and production forest reserves, with both systems contributing appropriately to both conservation and development objectives. 7-1 Legal framework The Forestry Law (1965) makes provision for the creation of reserves both for protection and for production objectives. Under Article 5 it decrees the creation of:- (a) National, State and Municipal Parks and Biological Reserves, to protect exceptional natural features, flora, fauna and scenic values, and to reconcile this protection with educational, recreational and scientific objectives. (b) National, State and Municipal Forests, for economic, technical or social objectives, including possible afforestation. 74 In specific reference to the Amazonian forests the Forestry Law calls for the utilization of the forests to be placed under technical constraints and management plans. In addition the Fauna Protection Law of 1967 makes provision for National, State and Municipal Biological Reserves, and the Law for a National Environmental Policy of 1981 covers the establishment of Ecological Stations and Environmental Protection Areas. The Federal Constitution of 1988, in reference to the rights of the people to an ecologically stable environment, requires the public administration to ensure proper ecological management of ecosystems and preservation of the diversity and integrity of the genetic heritage of the country. It further declares the Brazilian Amazon forest, the Atlantic forest, and three other regions, as a national estate, placing limitations on their utilization, to guarantee preservation of the environment (Schubart 1990). The principal legal category of reserves which offers the best possibility to combine management for timber production, and other forest products, with the conservation of genetic resources in situ is the National Forest. This corresponds approximately to the IUCN Category VIII (Multiple Use Management Area/Managed Resource Area) and has as its primary objectives the sustainable harvesting and management of the forest, principally for timber and other forest products, together with maintaining the protective functions of the forest over water resources, and the conservation of biological diversity, flora and fauna, insofar as this is compatible with the principal production objectives. Appropriate scientific and technological research, and environmental monitoring, are also envisaged. The basic criteria for the selection of an area as a National Forest are the potential for sustainable production of timber and/or other forest products. If rare, endemic or endangered species are present that might be adversely affected by the production operations it is suggested that another more restrictive category of reservation would be more appropriate (Padua 1989). There has been a marked increase in the number of National Forests in Brazil in the past three years, particularly in the Amazon region, where Schubart (1990) lists 24 such Forests up to the end of 1990, covering a total area of over 12.6 million hectares. Extractive Reserves are the other major category where direct productive uses of the forest may be combined with conservation of genetic resources. According to Padua (1989) these are intended to meet the needs of social groups who are dependent on the gathering of forest products for their survival, and who harvest in a sustainable manner according to traditional practices and in conformity with pre-established management plans. Secondary objectives include the conservation of biological diversity, and possible contributions to scientific and environmental education and monitoring. On this interpretation the harvesting of timber is specifically excluded. The traditional products of extractive forest management, as the term is used in Brazil, have been primarily rubber and Brazil nuts, although a wide variety of other fruits, fibres etc may be included. To provide a legal basis for the concept the former National Institute of Colonization and Agrarian Reform (INCRA) issued a Decree in July 1987 which provided guidelines for the settlement of Extractive Reserves as a mode of agrarian reform in the Amazon region (Allegretti 1990). The Decree used the concept of land use concessions, ceding use of the Reserves from the State to the practitioners during a minimum period of 30 years, with specific 75 regulations on land use practices. It also established a mechanism by which the State could mediate between the inhabitants of the Reserve and outside economic interests. Administration of Extractive Reserves on this model is by a group elected by the local inhabitants, either in the form of a cooperative or an association, thereby avoiding subdivision of the land into separate private holdings. According to Schubart (1990) the legal status of Extractive Reserves has yet to be finally decided but action has been taken under the National Environmental Policy (January 1990) to create Reserves of this nature. Insofar as some Extractive Reserves are being created inside National Forests the legal basis for combining their traditional practices with some selective harvesting of timber may be envisaged. The principal categories of reserve forming the fully protected area system in Brazil include National Parka (IUCN Category II), Ecological Stations, Ecological Reserves, Biological Reserves (all equivalent to IUCN Category I), and Environmental Protection Areas (similar objectives to IUCN Category V but including specific reference to biological diversity). All of these have significance in the conservation of ecosystems, species and genetic resources. In addition Wildlife Refuges and Game Reserves might provide incidental protection to plant genetic resources. 7-2 Setting Priorities The First International Seminar on Tropical Forest Management in Brazil, in 1985, in stressing the "need to maintain and preserve biological diversity in any undertaking in the Amazon", drew particular attention to the importance of a better understanding of forest dynamics and the interactions in processes such as seed dispersal, pollination, regeneration systems etc (Siqueira 1989). It also stressed the importance of typological classification and zoning of forests for production and conservation objectives. The principles involved have been discussed earlier, and in particular the vital necessity for gathering data on variation and variation patterns in species and populations, and on autecology and breeding biology, as a basis for action towards in situ conservation of genetic resources of priority species. Compared to some other tropical regions in the New World the Brazilian Amazon has a long history of botanical surveys and substantial holdings in several herbaria (Daly and Prance 1989). Nevertheless the sheer scale of an inventory needed to establish and interpret patterns of distribution of communities, major tree species and their probable intraspecif ic variation in relation to the existing environmental conditions and past history, is vast. The total area of Amazonian forest within Brazil in all of the reserved categories is probably around 5% of the total forest. While the location, individual size and shape of Protected Areas are more important than overall percentage figures there is clearly scope for the selection of a substantial number of additional reserves, given the limitations of the soils and the acknowledged environmental importance of forest cover. At the same time, for long-term security and in the face of increasing population pressures, the development of multiple-use management areas (predominantly National Forests and Extractive Reserves) that can contribute to the conservation of genetic resources in situ and complement action in fully protected areas, will be necessary. Since most of the Amazonian forest biome is still largely intact the possibility exists to plan the networks of production reserves and fully protected areas in an integrated manner, to maximise the effective use of the land and the resources devoted to management and protection. This is an important aspect of the zonation referred to by Padua (1989). 76 Vegetation mapping in Brazil has a long history but has been complicated by the use of a variety of classification systems. Knowledge of the Amazon was significantly advanced by the use of radar imagery in "Projeto RADAM" in the 1970 v s and although the maps produced were purely physiognomic they were extremely detailed. This information was used for the most recent vegetation map of the Amazon (Prance and Brown 1987) which recognises four main sub-types of the rain forest on "terra firma" (about 53% of the region) as well as several types of seasonal transition forests, savannas and savanna woodlands, forests on white sand soils, several types of inundated forests etc. Prance (1977; 1982) has drawn attention to centres of endemism believed to have been isolated refuges for tropical moist forest flora during the cooler and drier periods coincident with the Pleistocene glaciations. While the theory of refuges as an exploration of existing patterns of diversity and species distribution may be contested there is general agreement on the location of the centres of endemism. An important first step towards the preparation of a plan for the selection of conservation sites throughout the country were taken in 1982, with the publication of the Plan for a System of Conservation Units in Brazil (IBDF 1982). This plan has yet to be implemented as a coherent whole, the information presented in it is still valid and is still useful. In respect of the Amazon region a further major step was taken by the International Symposium on Priority Areas for Conservation in the Amazon Basin - Workshop 90, held in Manaus in January 1990 (Rylands 1990; Prance 1990). This meeting produced the initial draft of a map covering the entire Amazon basin, showing 94 proposed priority areas, in three levels of priority for conservation. The final selection of areas was the product of discussion between some 100 scientists working initially in small specialist groups (plant systematics; plant ecology; mammals; ornithology; herpetology; ichthyology; entomology; geomorphology and climate; and units of conservation). The botanists and zoologists subsequently pooled data from their component specialist groups to consolidate the first proposal for selection of areas while the data from geomorphology and climatology were used to identify fragile soils and ecosystems most in need of protection. Finally a synthesis was made from the botanical and zoological priorities to produce the draft map. Much further work is needed in the field to select actual sites for protected areas or for managed multipurpose production forests, in which conservation concerns are fully reflected. This may be progressively assisted by data from an ongoing study on the Biological Dynamics of Forest Fragments, north of Manaus. Although this initial prioritization is based on limited and unevenly distributed sources of information it has the great advantage of being regional in coverage, and can provide an initial framework for more detailed studies. It also provides some guidance for the allocation of priority to conservation objectives in existing production reserves. 7.3 Management Options The earliest attempts at natural forest management for the sustainable production of timber in the Brazilian Amazon are comparatively recent. Experimental silvicultural work in the tropical moist forest at Curua-Una, near Santar&m, Para, began in 1957. It included some experimental logging and regeneration studies on a small scale and the trials then established are still monitored. In 1972 some management activities started in a second area, designated as the Tapajos National Forest (600 000 ha) in 1974 and as a result 77 of studies and assistance from FAO a full management plan was prepared, envisaging selective harvesting, with a 70 year rotation based on natural regeneration, and a substantial component of line planting, to be harvested after 35 years (UNDP/FAO 1980). The aim was to allow logging of 1000 ha annually but the market conditions, heavily affected by an economic downturn in the early 1980 f s and by the availability of logs virtually free from large- scale forest clearance in other areas, prevented the implementation of the plan. However the forest has been well protected and studies have been undertaken which, together with the intensive inventory and studies undertaken initially, suggest the adoption of a polycyclic harvesting system, with cutting cycles of 30-35 years. Provision was made in the management plan for Biological Reserves and phenological studies, as well as 48 sample plots (Carvalho et al 1984), and the spatial distribution of 11 major species and the regeneration of 106 species were studied. Regeneration in the Tapajos forest is generally encouraging and frequently abundant, especially of valuable and fast-growing gap opportunist species such as Vochysia maxima after some opening of the canopy (Viana 1990). The ITTO, in response to a request from the Brazilian authorities to develop a demonstration model for forest management in the region, is providing assistance for controlled logging in the Tapajos National forest, to allow for pilot studies of management systems and silviculture on an operational scale, in concert with continued research into forest ecology and dynamics. An associated research proposal makes provision for identifying and protecting seed production stands of major species in the natural forest, covering a wide area in Amazonia, in collaboration with various organisations, as sources for future enrichment planting. In parallel with the study of forest management systems linked to genetic conservation objectives in specific areas, research into the genetic diversity and genetic systems of individual species will be needed. This applies for example to Swietenia macrophylla, which is apparently endangered by dysgenic selective logging outside the Reserved Forests. Other species likely to be given high priority for such action include Aniba rosaeodora, which has been subject to intensive exploitation for its essential oil content, and appears to show some chemical differences between provenances. In regard to in situ conservation of forest genetic resources the system of Genetic Reserves established in the Jari forest estate, with studies of phenology, ecology, regeneration and mortality, should provide valuable additional data. 7,4 Secondary Forest and Non-Timber Forest Products Although the scientific approach to natural forest management for timber production is relatively recent and limited in scale other models of traditional systems of management, including utilization and management of timber trees, exist which may be incorporated in a comprehensive system of genetic resource conservation. Secondary forests are now widespread in recently settled areas of Amazonia and although frequently seen as examples of land abandonment following shifting cultivation or pasture degradation they are in fact used and managed by a variety of rural communities (Dubois 1990). Some of these forests include a high proportion of economic species, although selective logging of accessible areas along the rivers and streams has seriously depleted the populations, with possible dysgenic effects. Dubois (1990) proposes systems of enrichment planting, combined with the management of natural regeneration, to intensify productivity of a variety of tree 78 species, in association with other crops, in agrof ores try systems dominated by the perennial vegetation. With appropriate attention to seed sources and management practices such systems could contribute to in situ conservation, as well as to the retention of forest cover and sustainable livelihood for the local people. Undoubtedly the longest history of forest management in Amazonia is that of the indigenous communities, harvesting mainly non-timber forest products (NTFP's). The Amerindian groups and "caboclos" (residents of the floodplains, of mixed descent) have detailed knowledge of the forest resources they depend on for survival (Parker et al 1983; Anderson 1990). The indigenous reserves may contribute to in situ conservation of some timber species but in addition knowledge of local traditional practices in the harvesting of NTFP's might be incorporated in management systems in National Forests and Extractive Reserves, with the possibility of retaining a wider base of species diversity in production systems. 7,5 Information, Research and Coordination Prioritization and coordination of research are essential to focus resources where they are most needed, in terms of the areas to be productively managed, and the centres of diversity and endemism most endangered. For this reason, the association of Biological Reserves and Ecological Stations with the new National Forests, with provisions for research in aspects of ecology, phenology and forest dynamics relevant to silvicultural and conservation objectives, to be conducted in advance of the preparation of management plans and actual exploitation, is most likely to support efforts in in situ conservation. An example of this strategy is seen in the recent siting of a Biological Reserve for scientific research under the Museu Goeldi, Belem, in association with the Caxiuana National Forest. Equally important is the establishment of efficient, computerised data- banks, to facilitate the storage and retrieval of information on species distribution, with related environmental and other data bearing on the nature and future use of the genetic material. A good example within Brazil, although outside the Amazon .region, is the proposed cooperative programme of the Projeto Nordeste for a botanical species inventory, to be carried out in nine states of north eastern Brazil. This is expected to involve collaboration between ten or more Brazilian herbaria, research centres and universities, with the involvement of the Royal Botanic Gardens, Kew, U.K., within which the Biodiversity Database will form a coordinating link. The same principle should enable the progressive development of data needed for planning in situ conservation of forest genetic resources in the Amazon region. The involvement of major Brazilian centres of scientific expertise outside the region, such as the Universities of Sao Paulo and Rio de Janeiro, as has been recommended (Daly and Prance 1989), may also be facilitated by such coordinating mechanisms. Skilled human resources are the most vital requirement to gather and interpret the data needed for efficient and coordinated in situ conservation programmes. Because of the intrinsic scientific interest of the Amazon region it is possible to attract substantial international expertise and funding to support expeditions and inventories in this region, to be carried out in collaboration with Brazilian scientists. For example the research undertaken at the Maraci Ecological Station, Roraima, embracing both rainforest and savanna, at the invitation of the Brazilian Secretariat of the Environment, 79 in 1987-88, mobilized the resources of 148 scientists and 55 technicians and resulted in the collection of data involving many new species, and leading to the compilation of the most comprehensive study to date of any forest area in northern Amazonia (Hemming 1989). Such collaborative schemes also offer opportunities for further training of local scientists, both through direct association with other specialists in the field and through postgraduate study offered within the programmes. The institutional focus to determine the selection of the areas and subjects for collaborative research, and to coordinate subsequent action in accordance with national priorities, is critically important in securing the most efficient use of available resources. In view of the very extensive geographical area covered by the Brazilian Amazon, which includes several states, and the number of research institutions and other organisations which might usefully be involved, such central coordination will be critically important to the success of the overall strategy for in situ conservation of forest genetic resources in the sub- region. INDIA; WESTERN GHAT FORESTS, KARNATAKA 8.1 National Policy Forest reservation and management, involving detailed working plans based on inventory growth and yield data and associated research, has been practised in the Indian subcontinent for over 100 years. The concept of Protected Areas in India can be traced back even further to at least the 4th century BC, when the establishment of Abhayaranyas (forest reserves) was proposed (Singh 1985). However in recent times since the establishment of the existing network of Forest Reserves the human population has more than trebled and the cattle population has grown by a factor of 2.5 (Shyamsunder and Parameswarappa 1987; in IUCN 1991b). The country now contains around 15% of the entire human population, expanding at about 2% annually and expected to reach over 1.4 billion by year 2025 (WRI 1990). Through increasing population pressures and the demands of economic development, particularly in the period 1950 to 1980, India has lost large areas of its tropical forests. In 1952 the revised National Forest Policy called for the retention of one-third of the total land area as forests and this remains an objective in the 1988 Policy revision, which reinforced the increasing attention to the role of trees in land-use systems both within and outside the reserved forests. Whereas earlier policies and programmes had placed emphasis on meeting the raw material requirements of wood-based industries the 1988 policy gave priority to environmental conservation and to the needs of local people. Improved utilization of the large areas of degraded lands resulting from past deforestation and misuse had been made a priority action in both national and state government programmes. In 1985 the creation of the National Wasteland Development Board provided a central coordinating point for the massive reforestation programme (target approximately 9 million hectares during the 7th Five Year Plan) to develop fuelwood and fodder supplies. Through reorganisation of forestry research and training, greater attention was focused on aspects of social and community forestry, and to reforestation, in accordance with the revised Forest Policy. At the same time the rate of deforestation was substantially reduced, and conservation efforts were strengthened. 80 Wildlife protection also has a long tradition in India and by 1988 there were 65 national parks and 407 wildlife sanctuaries covering an area of approximately 131 800 km 2 or 4.4% of the land area (IUCN 1991b), although a significant proportion had not yet achieved full legal status. The network of 16 Tiger Reserves, covering 26 000 km 2 , including large areas of tropical moist forest, has been recognised internationally as a most significant conservation success. Nevertheless major problems remain in securing an adequate national network of Protected Areas for the conservation of biological diversity and genetic resources. To complement such areas, considerable reliance will need to be placed on the protection and management of ecosystems, species and genetic resources outside the Protected Area network, and particularly within Forest Reserves while at the same time taking care to not further deepen the already existing conflict between the needs of the rural populations surrounding Forest Reserves and Protected Areas, for fuelwood, fodder and a range of other forest products. Among the most significant areas for special attention in this respect are the surviving tropical rain forests, confined to the Anadaman and Nicobar Islands, to areas in the north-eastern states, and to the Western Ghat forests in southern India. 8-2 Western Ghat Forests The Western Ghat Forests occupy the range of high hills extending for about 1 600 km from the southern point of India up the western coast, to about 21N. Three main regions within them have been recognised (Pascal 1982) with the largest and best conserved areas of the original forest lying in the central region, within Karnataka State. Here the hills range from 700 to nearly 1 900 m a.s.l. The rainfall varies along three gradients: south- north, east-west (with a marked rain shadow on the eastern side) and altitudinally. Annual totals show great variation, but are concentrated within a few months (e.g. at Agumbe the mean of 7 000 mm/ year is concentrated in 128 days) with a dry season lasting from 3 months (at high altitude in the south) to 8 months, depending on locality. All the major rivers of South India arise in the Western Ghats and their environmental significance was the prime reason for early demarcation, reservation and the protection of significant areas of Reserved Forest in this area over the past 120 years, since the 1870's. Despite the legal status of the Reserved Forests, and the efforts of an efficient State Forest Service, large areas were lost to dams, mining and cultivation, including substantial illegal encroachment. Nearly 60% of the remaining forest is classified as degraded, much of it severely, but there is still a significant "core" area of evergreen and semi-evergreen forest (Sinha 1988), of the greatest ecological importance and arguably of at least equally high priority in global terms as any other tropical moist forest in the world. The Western Ghat Forests lie between the African and the Indo-Malaysian rainforest blocks and are isolated from neighbouring areas of rainforest. While the levels of total biological diversity are lower than those present in most other tropical rainforests in Asia, they are rich in endemic species. Of the 15 000 higher plants recorded in India over 4 000 species are found in the Western Ghats (5% of the land area) and of these 1 800 species are endemic, mainly found in the remaining evergreen forests (IUCN 1991b). These include 130 tree species endemic to the Ghats (Bor 1953) and six endemic mammals, including the lion-tailed macaque (Macaca silenus), while 84 of India's 112 endemic amphibians are found there (Inger and Dutta 1987). In regard to trees and shrubs, both the diversity of species and intraspecific 81 variation, adapted to the local environment, are clearly of great potential importance. In no other part of the world is evergreen rainforest found under such seasonal climates (Rai and Proctor 1986), and the adaptations of plant populations, including many valuable timber trees, to the climatic conditions may be even more valuable in the context of possible climate change. Among the most valuable timbers of international significance are Dalbergia latifolia (Indian rosewood), Calophyllum inophyllum, Pterocarpus marsupium, Tectona grandis, Terminalia crenulata, T.paniculata and other species suitable for furniture, doors, windows, interior construction and other high value applications. Many more are of great local importance for house construction, bridges, carts, agricultural and household implements etc. As a result of over-exploitation and particularly illegal logging, overgrazing, fires, encroachment by cultivators and other pressures from the increasing populations, both the effective area and the regeneration of the forests have been severely reduced. In the surrounding areas there are almost as many cattle as people and it has been estimated that at least 10% are dependent on free grazing in the forests. Whereas this had formerly been regulated within permissible limits such control is no longer possible because of the huge numbers involved (IUCN 1991b). The effect on the genetic resources of the tree populations from the reduction and local elimination of regeneration as a result of these pressures is likely to be increasingly severe. 8.3 The Strategy for Integrated Development and Conservation Despite a clear policy, sound legislation and an efficient Forest Service, which had succeeded in protecting the reserved forests for over 100 years, it became clear that their effective regeneration and therefore long- term survival were increasingly threatened. While the intensified programmes of reforestation were expected to mitigate the pressure, further action was needed to conserve the natural ecosystems and their unique genetic resources. In 1988 the Karnataka State Forest Department drafted a proposal for a project on the Integrated Development of Forests in the Western Ghats to attempt to reconcile satisfactorily the many conflicting demands made on the forests, including the central need to conserve ecosystems and genetic resources. The central element of this strategy was the recognition of three sets of management objectives, namely ecological, economic and social ("local needs 11 ). The conservation of the genetic resources of the tree and shrub species relates in different ways, and to different degrees, to all three of these objectives. However the actual operations needed to achieve conservation will vary from strict protection, involving exclusion of other human activity, to agro-forestry programmes. The concept involves a multidisciplinary approach, and the recognition that local people, who are responsible for much of the biotic pressure on the forest, must play a role in the planning, management and protection of the forest resources. The means to achieve the reconciliation of development and conservation objectives is the Joint Forest Planning and Management (JFPM) system, which derives from practices being developed by the Karnataka Forest Department (KFD). 82 The two essential elements are: (a) Joint Planning - a consultative process by which the KFD and local people jointly discuss the ecological and environmental condition of a specific area of the forest, and the scope for it to meet specific needs. (b) Joint Management - for certain areas, where the KFD and forest users will agree on a division of responsibilities for management and a division of the proceeds. The precise arrangements will vary from one specific area to another and as appropriate the joint management may cover all, or only certain categories of the trees and other plants present. A key element is the division of the forest into four zones, with an additional zone comprising an area immediately outside the Reserved Forest to be jointly managed to meet the needs of the adjacent villages. Within that a boundary zone of forest, also under joint management with local people, will serve similar objectives. The main part of the forest is to be managed for production of timber and other products, divided into two zones, depending on whether or not there are forest-dwelling people. Finally the "core" zone is strictly reserved for conservation objectives. An important aspect of the management of all areas where joint responsibilities operate is the preparation of contractual forest agreements according to specific micro-level plans. The ownership and statutory responsibility for management over the entire forest reserve remains with the KFD, even where management responsibility may be shared by contractual agreement. Future working plans for the management of the forest will be prepared by zones. In the "core" zone the plan will be oriented to research, concerned with autecology and the interrelationships between species, and with studies providing a "control" for comparison with the regeneration, growth and yield studies in the production zones, which will be designed to gather the data needed for management, silvicultural and harvesting activities. The satisfactory harmonization of the different objectives and interests is dependent not only on the process of joint planning and management but also on the incorporation of new information to be gathered in the course of the project into the management process, related to improved understanding of both ecological and socio-economic relationships; and on the realisation and distribution of benefits from improved management of the resource. In addition to the biological research directly related to the conservation and management of the forest's genetic resources, aspects of social and economic research, and the incorporation of staff training and public education into the programme are also needed. The conceptual relationship between meeting the needs of the local people, restoring and maintaining the productivity of the forest resource and the conservation of overall genetic diversity and genetic resources, is illustrated below. 83 FOREST RESTORATION & PROTECTION ECOSYSTEM STABILITY CONSERVATION OF GENETIC RESOURCES SUSTAINED YIELD ASSURED SOUND FUTURE ECONOMIC BENEFITS LOCAL LIVING STANDARDS MAINTAINED AND IMPROVED Efficient data collection, handling and analysis will be an important feature of the strategy, using enhanced GIS capabilities, taxonomic databases, and when appropriate modelling of growth, yield and other features of the production system. Where, because of the severe reduction in the regeneration within the forests, and the need to reestablish viable populations, enrichment planting and gap planting is planned, careful selection and documentation of seed sources will be needed to safeguard the genetic integrity of the local populations. In some circumstances, particularly in the outer zones, the use of exotic species incapable of hybridizing with indigenous populations, may be appropriate. However important opportunities exist to reinforce in situ conservation of the genetic resources of the original forest by artificial regeneration using local seed sources. For some species the strategy may also include aspects of ex situ conservation, possibly in association with the Trivandrum Botanic Garden in the adjacent Kerala State, which has an emerging programme for the conservation of rare and endemic plants of selected centres of endemism within the Western Ghats. 85 Appendix 1, METHODOLOGY OF A STUDY OF THE REPRODUCTIVE BIOLOGY AMD GENETICS OF CORDIA ALLIODORA (R & P) OKEN David Boshier 1 1 . Introduction Sound decisions in genetic conservation and progress in long term breeding require detailed knowledge of the taxonomy, population structure, reproductive biology, mating systems, etc of the species concerned. For many tropical forest tree species, little information is available and managers are faced with problems in developing effective strategies. For example, questions that need to be answered are:- Are neighbouring trees in natural stands highly related and is there inbreeding? What, if any, is the incompatibility mechanism to stop inbreeding? What agents effect pollination and how do they affect gene flow and population size? What degree of inbreeding is acceptable? The case study presented here describes the methodology used in ongoing research carried out for one particular species - Cordia alliodora. It is not intended that the methodology described here be used as a recipe for studying any tropical tree species, but that it may provide ideas on how to approach the problem. Techniques used will always depend on the type of tree, its distribution, size of flowers, pollinators etc. Previous work with C^ alliodora, relevant to the study, is generally of a preliminary nature but provides useful information. C. alliodora is an important neotropical tree, which combines high quality timber and value with fast growth on good quality soils. It is used extensively throughout its natural range and its light crown and self pruning habit make it particularly suitable for use in various agrof ores try systems, providing valuable timber and income to small farmers. The species has a wide distribution occurring from northern Mexico through Central and South America as far south as Bolivia, Paraguay, southern Brazil and northern Argentina. In Paraguay, northern Argentina and parts of southern Brazil the closely related C. trichotoma (also classified as C. alliodora var tomentosa) predominates over C. alliodora (Gibbs and Taroda 1983). C. alliodora is found on most of the Caribbean Islands from Cuba to Trinidad, but is almost certainly not native to Jamaica. Through this geographical range the species occurs under a wide variety of ecological conditions, varying from very wet (as much as 6 000 mm per year) to seasonally dry (as low as 600 mm per year); and from sea level to as high as 1 200 m a.s.l. in Central America and 2 000 m a.s.l. at lower latitudes in Colombia. In lowland humid tropical regions it is generally a tall, thin, lightly crowned tree, reaching heights over 40 m and d.b.h. up to 1 m in mature trees, although in the region of 50 cm d.b.h. is more normal. In seasonally dry areas it is smaller, and more poorly formed rarely reaching more than 20 m in height. The overall objective of the present study was to gain an understanding of the reproductive biology of C. alliodora that will give adequate knowledge to make sound long term decisions for in-situ/ex-situ conservation and breeding of the species. Oxford Forestry Institute, South Parks Road, Oxford, 0X1 3RB, U.K. 86 The specific objectives were to: 1) study in detail the phenology of flowering; 2) determine the type of pollination and possible incompatibility systems; 3) study the mating system, gene flow and neighbourhood area found in natural stands. 2- Flowering and Seed Phenology Flowers are hermaphrodite, unspecialized, about 1 cm in length and occur in large panicles. The size of panicles varies, with as few as 50 flowers to as many as 2 000. The petals are white and persistent turning to brown and acting as a parachute in wind dispersal of the seed. The aim of the first phase of the phenology study was to detect any patterns of flowering that may exist within the tree, branch and inflorescence and provide data on length of the flowering and fruiting periods and the proportions of seed reaching maturation. It is important to know how a tree flowers and what effect that might have on pollinator movements. For example: sequential flower opening from the top to the bottom of the crown may mean that early flowers are pollinated by one group of pollinators and later flowering by a completely different group of pollinators. Conversely, random flowering within the crown could stimulate greater movement of pollinators between panicles and different trees. 2.1 Within inflorescences: individual groups of flowers were identified and the stage of flowering observed every three days from flowering to seed fall. Classification of the stages of flowering was based on Mendoza (1965), distinguishing : 1 - start of formation of flower buds 2 - individual flower buds visible, but not open 3 - individual flower buds full 4 - white petals emerge from flower buds, but not open 5a- petals open, stigma receptive 5b- petals still white, stigma withered 6 - petals turn brown 7 - embryo starts to swell. It was thus possible to observe at which stage of flowering most losses occurred, the length of each stage, any sequential pattern of flowering or losses within a panicle, and the effect certain weather conditions might have on flowering and fruiting. 2.2 Within branches and trees: individual inflorescences/branches within a number of branches/trees were observed periodically from flowering to seed production. The same flowering categories were used as in 2.1 (categories 5a and 5b combined) and each panicle/branch classified on the predominant category of flowers. 2.3 Within populations: flowering phenology within a population of trees of the same species is of fundamental importance to understanding gene flow, genetic structure, seed production and yet few studies exist of individual tropical tree species based on large samples. Similarly few studies have looked at the flowering of the same population for more than one year. It is important to know about periodicity of flowering for the species and how individual species flower with respect to one another from year to year. For example, because of a synchrony of flowering two adjacent trees may be unable to mate, which may be the case in one year but not in the following. Similarly some trees may be heavy flowerers in one year and totally sterile the following year. 87 To look at flowering within natural populations, in a 26 hectare plot all trees (216) of a flowering age/size surrounding three C. alliodora selected trees up to a radius of approximately 500 meters were mapped, and labelled. The stand, which was also to be used for the mating system and gene flow study (see 5), was chosen taking into account various factors: (a) the presence of selected plus trees would provide direct information on the degree of diversity sampled when collecting open-pollinated seed from the plus trees, which are part of a breeding population (Boshier and Mesen, 1989); (b) the possibility of easily defining a population, there being a natural break between the trees under study and nearest trees of the same species; (c) ease of access to allow frequent visits. The crown of each tree was observed every three days to determine the start, peak and end of flowering and the percentage of the total number of flowers open by that date. Subjective assessments of flower (scale of 0-5) and seed production (scale of 0-3) were also made. The same observations were made over three successive flowering seasons (Jan-April) to study year to year variation. An index for synchronicity of flowering was calculated, for the whole stand as well as for the plus trees in relation to the surrounding trees (Augsberger, 1983). To look at the effect population substructure may have on flowering synchrony the individual tree and stand index was re-calculated on the basis of increasing population size from a base tree (a plus tree). Two other measures of population synchrony were made: (1) synchrony of the first day of flowering, calculated as one standard deviation around the mean of the first day of flowering; and (2) synchrony of the median day of flowering, calculated as one standard deviation around the mean of the median flowering day. For one flowering season similar phenological observations were carried out in a stand in the seasonally dry Pacific region to allow comparison of flowering under differing climates. 2.4 Between populations Much information on the general phenology of a species can be gained from the examination of herbarium specimens. Even for species for which full collections are not available, studying existing specimens can reveal much about phenology, distribution and likely regions where the species may occur. Herbarium collections of C. alliodora and C. trichotoma were studied and full details noted form specimen labels as well as the stage of flowering. It was possible to classify well-preserved specimens under one of the flowering categories used in 2.2. Logging of the information onto the OFI herbarium database, BRAHMS (Filer, 1991) allows mapping of the distribution of the species as well as studying gross variation in flowering time over this range. 3. Breeding System Many species of Cordia are heterostylous, indeed in one case failure to recognize this originally led to one species being classified as two (C^ thaisiana and C. apurensis, Agostini, 1983). Gibbs and Taroda (1983) studied the C. alliodora * C. trichotoma complex based on herbarium specimens from South America and distinguished the two species on the basis of the pattern of heterostyly. In C. alliodora heterostyly appears to have broken down although there is variation in style length. Opler et al (1975) studied the pollination biology of a number of species of Cordia and found varying degrees of self- incompatibility in the trees of C. alliodora they sampled. 88 A combination of field and microscope work were used to study stylar patterns, both within and between families and their relationship to the incompatibility mechanism. On freshly cut inflorescences from an open pollinated progeny trial, flowers were emasculated as they opened and before anther dehiscence had occurred. Controlled cross, self and related pollinations between half-sibs were carried out and the flowers fixed at varying times after pollination to allow study of pollen tube development on the stigma and through the style. The same controlled crosses were carried out in the field to allow comparison with the results from the microscope study, checking that crosses deemed compatible produce seed and those deemed incompatible failed to do so. The flowers on any one panicle open during a period of 4-6 days. The numbers of seed produced per cross and empty flowers were counted to compare fertility from related and unrelated crosses. The seed was used to mount a nursery trial to look at the effects of inbreeding on initial growth rate. Controlled pollinations were also carried out in the field to look at length of stigma receptivity, and to exclude any possibility of parthenocarpy. It was noticeable that in the dry zone the stigma was receptive for only one day at the end of which it was withered whereas in the wet zone the stigma could be receptive for as long as three days, emphasising the importance of caution in extrapolating results from one region to a climatically different one. 4 Pollinators Opler et al (1975) looked at potential pollinators and concluded that noctuid and geometrid moths and anthophorid bees were the most likely pollinators. Their work was however restricted to the dry Guanacaste region of Costa Rica and it is likely that in the wet lowland regions other pollinators are of importance. As many insects are seasonal, it is likely that different insects act as pollinators at different times during a prolonged flowering season. Collections were made of insects visiting flowers during the day and night at various times in the flowering season. The insects were mounted, the identified to genus and stored for future use as a reference collection. Observations were made on several occasions of numbers and types of insects visiting flowers, to see which were the more important ones. Other studies such as those on the effectiveness of different pollinators in depositing pollen on the stigma, or marking and recapturing to look at fidelity to food source could also be carried out (Frankie et al 1976). Flowers were collected and fixed at different times of the day following anthesis (dawn, midday, dusk) from a number of trees. Microscopic observation of these flowers for the quantity of pollen on stigma and pollen tube growth can provide supportive evidence on the time of pollination and the relative importance of certain pollinators. Similar collections were made in the Pacific region to compare pollination for the species in differing climates. It is important to study variation of nectar flow with time as this can give clues as to which vectors are active pollination and when. A number of trees should be studied as there can be variation from tree to tree both in time of production, quantity, and quality (sugar concentration). Even within an inflorescence patterns of nectar production may exist which will tend to move pollinators in certain directions. To characterize the availability of rewards to pollinators, variation in nectar production was studied on a number of trees at intervals during a three day period after flower opening. The process was repeated on a number of occasions during the flowering season. 89 5. Mating System, Gene Flow and Neighbourhood Area Isozymes may be used a genetic markers to provide estimates of the mating system, gene flow, neighbourhood area and paternity. The interpretation of such data is made more significant when it can be combined with field observations of the spatial distribution of the trees and flowering phenology. For this reason the same population was chosen for studying mating system as for the within population phenology study. Seed was collected during the 1989 season from the trees surrounding two plus trees; during 1990 and 1991 seed was re- collected from the plus trees to look at year to year variation in outcrossing rates and paternity for these two trees. Although sufficient seed could be collected from one panicle, seed was collected from the whole crown, to ensure that the sample was not biased due to particular pollination events. To look at variation of pollination within a tree, at the time of collection, seed from one plus tree was divided into three lots representing the upper, middle and lower parts of the crown. Seed from three panicles on the same branch on the plus tree was also separated as individual lots. A total of 163 trees occurred within a 250 m radius of the plus trees, a number prohibitively large to allow all to be studied for isozymes. A sub- sample of 52 trees was therefore selected on the basis of distance from the plus tree and phenological data obtained during field work (see 2.3). The following trees were eliminated from the sample: (a) non-flowerers (36 trees); (b) trees that flowered completely before or after the plus tree (5); (c) trees classified as 1 for quantity of flowering (i.e. six or less panicles) (12); (d) trees in which less than six days synchrony in flowering was observed (13); (e) trees with scarce flowering (category 2) (21); (f) trees in category 3 for flowering and greater than 200 m from the plus tree (24). To aid in determining paternity for progeny arrays of specific trees, genotypes were assigned to un- sampled trees by assaying leaf material. The leaf material was collected and immediately frozen in the field in liquid nitrogen. To complement estimates of pollen flow from the isozyme work and allow estimates of neighbourhood size and area, seed dispersal was studied for four trees in the same population used for phenology and isozymes. A UV fluorescent dye was used to spray seed in the canopy and so enable tracing distance of seed dispersal from the trees. Seed was collected in 1 m 2 traps, placed at intervals along transects, laid out in the direction and reverse of the prevailing wind. The dye was still easily visible on the seed after six weeks and probably persists much longer. 6. Variation between Populations There are two complementary approaches to studying genetic variation between populations. The first, provenance testing, is generally used in forestry and studies the performance under uniform environmental conditions in one or more planting sites of trees grown from seed collected from different populations. Traits studied are usually of a continuously variable nature and the observed variation is split into genetic and environmental components. Provenance variation for Cordia alliodora has been studied in an international trial coordinated by the Oxford Forestry Institute based on collections made in the late seventies principally in Central America, with assistance from FAO and ODA (Stead, 1980). Over a wide range of sites, provenances from the humid Atlantic region of Central America and in particular Honduras and Costa Rica showed superior growth and form to Pacific region provenances, even when grown in seasonally dry areas (McCarter 1988). The study thus suggested ecotypic differentiation of the Pacific and Atlantic populations, but also indicated that 90 within these broad bands variation within provenances was greater than between. There appears to be great potential for genetic improvement through individual tree selection, the provenance assessments showing a 200-3002 difference within provenances between the best 10X of trees and the mean. The second approach is to study genetic diversity shown as variation in enzymes (and more recently DNA) for specific gene loci. The enzyme survey carried out in (5) above showed differences in the number of loci staining; and variability within loci, between the population under study and a Pacific zone population previously studied. Further studies of isoenzyme variation in the original OFI provenance collections of C. alliodora, being carried out at University of Massachusetts, Boston (U.S.A.), indicate interesting differences between provenances. 7. Chromosome Number More information on variation within and between closely related species can sometimes be revealed by looking at chromosome variation, both in terms of numbers and overall quantities of DNA. The two existing reports of chromosome number in C. alliodora; n=15 (Bawa 1973) and 2n=72 (Britton 1951), differ greatly. The possibility of gene duplication for a number of enzyme systems in the Atlantic population also suggests that some benefit may be gained from looking at chromosome number to see if intra-specific variation does exist in this trait, and if this bears any relation to the Pacific/Atlantic provenance differences encountered in the international provenance trials. Acknowledgements The above study was financed by the Overseas Development Administration of the United Kingdom (R4484 & R4724). The author is grateful to the following people; Ing. F. Mesen, CATIE; Ing. F. Lega, Scott Paper, for facilitating the field work in Costa Rica; to Dr. K.S. Bawa and Mr. M. Chase, Univ. Mass., Boston for collaboration with the isozyme work and Dr. G. Frankie, Univ. California, Berkeley, for entomological advice. 91 REFERENCES Adlard, P.G. Inventory for management. Report within the Forest (1990) Resources Management Project. OFI, Oxford (U.K.). Agostini, G. and Velasquez, D. Notas sobre Cordia thaisiana Agostini (1983) (Boraginaceae) . Ernstia 16: 14-15. Alder, D. Natural forest increment, growth and yield. In; Forest (1989) Inventory Project, Seminar Proc. (Ed. J.L.G. Wong). Forestry Department, Accra (Ghana) . Alder, D. "GAFOSIM": a projection system for natural forest growth and (1990) yield in Ghana. Consultant's report to the Ghana/UK Forest Resources Management Project. ODA, London (U.K.). Alegretti, M.H. Extractive reserves: an alternative for reconciling (1990) development and environmental conservation in Amazonia. In; Alternatives to Deforestation: Steps Toward the Sustainable Use of the Amazon Rain Forest (Ed. A.B. Anderson). Columbia University Press, New York (U.S.A.). Anderson, A.B. (Ed.) Alternatives to deforestation: steps the sustainable (1990) use of Amazon rain forest. Columbia University Press, New York (U.S.A.). Anderson, M.H. Extraction and forest management by rural inhabitants in (1990) the Amazon estuary. In; Alternatives to Deforestation: Steps Toward the Sustainable Use of Amazon Rain Forest (Ed. A.B. Anderson). Columbia University Press, New York (U.S.A.). Appanah, S. and Salleh, M. N. Natural regeneration and its implications (1990) for forest management in the dipterocarp forests of Peninsular Malaysia. In: Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa, T.C. Whitmore and M. Hadley). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Augsberger, C.K. Phenology, flowering synchrony and fruit set six (1983) neotropical shrubs. Biotropica ^5(4): 257-267. Arnold, J.E.M. The long-term global demand for and supply of wood. In: (1991) Forestry Expansion: a Study of Technical, Economic and Ecological Factors. U.K. Forestry Commission, Edinburgh (U.K.). Asabere, P.K. Attempts at sustained yield management in the tropical high (1987) forests of Ghana. In: Natural Management of Tropical Moist Forests: Silvicultural and Management Prospects of Sustained Utilisation (Eds. F. Mergen and J.R. Vincent). Yale University, New Haven, Connecticut (USA) . Ashton, P.S. Dipterocarpaceae. In: Flora Malesiana 9: 237-552. (1982) 92 Ash ton, P.S. Biosystematics of tropical forest plants: a problem of rare (1984) species. In: Plant Biosystematics (Ed. W.F. Grant). Academic Press, Toronto (Canada). Ash ton, P.S. Conservation of genetic resources. In; The Case for Multiple- (1988) Use Management of Tropical Hardwood Forests. Study produced for the International Tropical Timber Organisation. ITTO, Yokohama (Japan) . Barnes, R.D. and Gibson, G.S. (Eds). Provenance and genetic improvement (1984) strategies in tropical forest trees. Proc. IUFRO Joint Working Parties S2.02.08, S2.03.01, S2.03.13. Mutare (Zimbabwe), April 1984. Oxford Forestry Institute, Oxford (U.K.). Bawa, K.S. Chromosome numbers of tree species of a lowland tropical (1973) community. Journal of the Arnold Arboretum 54(4): 422-434. Bawa, K.S. Breeding systems of tree species of a lowland tropical (1974) community. Evolution 28:85-92. Bawa, K.S., Perry, D.R. and Beach, J.H. Reproductive biology of tropical (1985) lowland rainforest trees (I): Sexual systems and incompati- bility mechanisms. American Journal of Botany 72:331-45. Bawa, K.S., Ashton, P.S and Salleh, M.N. Reproductive ecology of tropical (1990) forest plants: management issues. In; Reproductive ecology of tropical forest plants (Eds. K.S. Bawa and M. Had ley ). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Bawa, K.S. and Krugman, S.L. Reproductive biology and genetics of tropical (1991) trees in relation to conservation and management. In; Rain Forest Regeneration and Management (Eds. Gomez -Pomp a et al). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Bor, N.S. Manual of Indian forest botany. Oxford University Press, (1953) London (U.K.).- Boshier, D.H. and Mesen, J.F. Breeding population of Cordia alliodora in (1989) Costa Rica. In; Breeding Tropical Trees: Population Structure and Genetic Improvement Strategies in Clonal and Seedling Forestry. Proc. IUFRO Working Parties, Pattaya (Thailand), November 1988. (Eds. G.L. Gibson, A.R. Griffin and A.C. Matheson). Oxford Forestry Institute, Oxford (U.K.) and Winrock International, Arlington, V.A. (U.S.A.). Britton, D.M. Cytogenetic studies in the Boraginaceae. Brittonia : (1951) 233-266. Bullock, S.H. Breeding systems in the flora of a tropical deciduous forest (1985) in Mexico. Biotropica 1J7: 287-301. Carvalho, J.O.P. Mane jo de regenerate natural de especies f lores tais. (1984) EMBRAPA-CPATU, Belem (Brazil). Carvalho, J.O.P. et al. Manejo de florestas naturais do tropico umido com (1984) referencia especial a Floresta Nacional do Tapajos no estado do Para. EMBRAPA-CPATU, Belem (Brazil). 93 Carvalho, J.O.P. et al. Pesquisas com vistas ao manejo de matas nativas na (1987) regifio do Rio Jari. EMBRAPA-CPATU, Belem (Brazil). Catinot, R. Etudes sur les sytemes d ' amenagement dans les forets (1986) tropicales mixtes d'Afrique francophone. Document prepared for the Forest Resources Divsion, FAO, Rome (Italy). (Unpublished; synthesised in FAO Forestry Paper NQ 88, see FAO 1989b). Chachu, E.R.O. Allowable cut from the forests. In; Forest Inventory (1989) Project: Seminar Proceedings, March 1989 (Ed. J.L.G. Wong). Forestry Department, Accra (Ghana). Cochrane, T.T. and Sanchez, P. A. Land resources, soils and their (1982) management in the Amazon region: a state of knowledge report. In; Amazonia: Agriculture and Land Use Research (Ed. S. Hecht). CIAT, Cali (Colombia). Daly, D.C. and Prance, G.T. Brazilian Amazon. In; Floristic Inventory (1989) of Tropical Countries (Eds. D.G. Campbell and H.D. Hammond). New York Botanical Garden, New York, N.Y. (U.S.A.). Davis, F.W., Stoms, D.M., Estes, J.E. and Scepan, J. An information (1990) systems approach to the preservation of biological diversity. Int. J. of Geogr.Inf. Systems 4(1): 55-78. Dawkins, H.C. The first century of tropical silviculture - successes (1988) forgotten and failures misunderstood. In: The Future of the Tropical Rain Forest (Ed. M.J. McDermott) . Conference Proceedings. Oxford Forestry Institute, Oxford (U.K.). de Beer, J.H. and McDermott, M.J. The economic value of non-timber forest (1989) products in Southeast Asia. IUCN/WWF, Gland (Switzerland). de Graaf, N.R. Managing natural regeneration for sustained timber (1991) production in Suriname: the CELOS silvicultural and harvesting system. In: Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa et al) . Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). di Castri, F. and Younes, T. (Eds). Ecosystem function of biological (1990) diversity. Biology International NQ 22. Special Issue. International Union of Biological Sciences, Paris (France). Dransfield, J. The conservation status of rattan in 1987: a cause for (1987) great concern. In: Recent Research on Rattans (Eds. A.N. Rao and I. VongkaluangTT Kasetsart University (Thailand) and the International Development and Research Centre, IDRC (Canada). Dubois, J.C.L. Secondary forests as a land-use resource in frontier zones (1990) of Amazonia. In; Alternatives to Deforestation: Steps Toward Sustainable Use of the Amazon Rain Forest (Ed. A.B. Anderson). Columbia University Press, New York (U.S.A.). Dubois, J.C.L. The present status of research into management of the rain (1991) forests of Amazonian Brazil. In: Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa, T.C. Whitmore and M. Hadley), Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). 94 Dykstra, D.P. and Heinrich, R. Sustaining tropical forests through (1992) environmentally sound harvesting practices. Unasylva 43/2, NS 169:9-15. FAO, Rome (Italy). ECE/FAO. European Timber Trends and Prospects to 2000 and beyond. (1986) Econ.Comm.for Europe/FAO Agric.and Timber Division. FAO/UN, Geneva (Switzerland). Falconer, J. The significance of forest resources in rural economies of (1991) southern Ghana: summary of a study on non-timber forest products. Forestry Dept., Accra (Ghana) and ODA, London (U.K.) (Unpublished) . FAO: Food and Agriculture Organization of the United Nations. Methodology (1975) of conservation of forest genetic resources: report on a pilot study. FO:MISC/75/8. FAO, Rome (Italy). FAO. Los recursos fores tales de la America Tropical. FAO/UNEP (1981) Informe Tecnico NQ1 . UN 32/6 1301-78-04. FAO, Rome (Italy). FAO. Intensive multiple-use forest management in Kerala. Forestry (1984) Paper NQ 53. FAO, Rome (Italy). FAO. Intensive multiple-use forest management in the tropics. (1985a) Analysis of case studies from India, Africa, Latin-America and the Caribbean. Forestry Paper NQ 55. FAO, Rome (Italy). FAO. Tropical Forestry Action Plan. FAO, Rome (Italy). (1985b) FAO. Plant Genetic Resources: their conservation in situ for human (1989a) use FAO, Rome (Italy). FAO. Management of tropical moist forests in Africa. Forestry Paper (1988) NO 88. FAO, Rome (Italy). FAO. Review of forest management systems in Tropical Asia. (1989c) Forestry Paper NQ 89. FAO, Rome (Italy). FAO. Household food security and forestry: an analysis of socio- (1989d) economic issues. FAO, Rome (Italy). FAO. Management and conservation of tropical moist forests in Latin (1992a) America. Forestry Paper NQ 101. FAO, Rome (Italy). FAO. The United Nations Conference on Environment and Development (1992b) and the Framework Convention on Biodiversity. Forest Genetic Resource Information NQ 19:2-3. FAO, Rome (Italy). FAO. Information Notes on some issues related to forests. 15, (1992c) 2-page notes prepared for the United Nations Conference on Environment and Development, Rio de Janeiro (Brazil), June 1992. Forestry Department, FAO, Rome (Italy) (see especially following notes: "Sustainable Management of Tropical Fores ts , FAO/FORM Febr.1992; and "Forests as habitats of biological diversity and conservation of forest genetic resources", FAO/FORM Febr.1992). 95 Filer, D.L. "BRAHMS": Botanical Research and Herbarium Managment System. (1991) Internal Information Note. Oxford Forestry Institute, Oxford (U.K.) (Unpublished) . Frankel, O.H. Genetic conservation in perspective. In: Genetic Resources (1970) in Plants - their Exploration and Conservation (Eds. O.H. Frankel and E. Bennett). IBP Handbook NQ 11. Blackwell, Oxford (U.K.). Frankel, O.H. Evolution in jeopardy; the role of nature reserves. In; (1981) Evolution and Speciation (Eds. W.R. Atchley and D. Woodruff). Cambridge University Press, Cambridge (U.K.). Frankie, G.W., Opler, P. A. and Bawa, K.S. Foraging behariour of solitary (1976) bees: implications for outcrossing of a neotropical forest tree species. J. Ecol. 64; 1049-1057. Frimpong-Mensah, K. Requirement of the Timber Industry. Proc. Seminar on (1989) Forestry Inventory Project 29-30 March, 1989. Forestry Department, Accra (Ghana). Gent, J.R.P. Forestry in relation to agriculture in the Gold Coast and (1929) Ashanti. Paper No VIII, Forestry Department, Accra (Ghana). Ghartey, K.K.F. Results of the Inventory. In; Forestry Inventory Project: (1989) Seminar Proceedings (Ed. J.L.G. Wong). Forestry Department, Accra (Ghana). Ghartey, K.K.F. The evolution of forest management in the tropical high (1990) forest of Ghana. Paper presented to the Conference on the Conservation and Rational Utilization of the Closed Forest of Central and West Africa, Abidjan, 5-9 Nov., 1990. lUCN/African Development Bank/World Bank. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland). Gibbs, P.E. and Taroda, N. Heterostyly in the Cordia alliodora - (1983) Cordia trichotoma complex in Brazil. Revista Brasileira de Botanica 6:1-10. Gibson, G.L., Griffin, A.R. and Matheson, A.C. (Eds). Breeding tropical (1989) trees: population structure and genetic improvement strategies in clonal and seedling forestry. Proc. IUFRO Working Parties S2.02.08, S2.02.09. Pattaya, Thailand, November 1988. Oxford Forestry Institute, Oxford (U.K.) and Winrock International, Arlington, V.A. (U.S.A.). Gilbert, L.E. Food web organisation and conservation of neotropical (1980) diversity. In: Conservation Biology (Eds. M.E. SouU and B.A. Wilcox). Sinauer, Sunderland (U.S.A.). Gomez-Pompa, A., Whitmore, T.C.and Hadley, M. (Eds). Rain forest (1991) regeneration and management. Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Grut, M. Economics of managing the African rainforest. Paper presented (1989) at the 13th Commonwealth Forestry Conference, Sept. 1989, Rotorua (New Zealand). (Unpublished). 96 Grut, M. Revenue and concession policy for the timber forests of West (1990) Africa. In: Rep. Seminar on Sustainable Development of Tropical Forests, Bali, May, 1990. PCF(VII)/4. ITTO, Yokohama (Japan). Hall, J.B.and Swaine, M.D. Distribution and Ecology of vascular plants in (1981) a tropical rainforest. W. Junk Publishers, The Hague (Netherlands). Harris, L.D. The fragmented forest: island biogeography theory and the (1984) preservation of biotic diversity. Univ. of Chicago Press, Chicago (U.S.A.). Hartshorn, G.S. Gap-phase dynamics and tropical tree species richness. (1989) In: Tropical Forests: Botanical Dynamics, Speciation and Diversity (Eds. L.B. Holm-Nielsen, I.C. Nielsen and H. Balslev). Academic Press, London (U.K.). Hawthorn, W. The flora and vegetation of Ghana's forests. In; Forest (1989) Inventory Project: Seminar Proceedings, March 1989 (Ed. J.L.G. Wong). Forestry Department, Accra (Ghana). Hawthorn, W. Field guide to the forest trees of Ghana. National Resources (1990a) Institute/ Overseas Development Administration, London (U.K.). Hawthorn, W. Forest regeneration after logging in Bia South GPR (Ghana) . (1990b) Report under the Forest Resources Management Project. Overseas Development Administration, London (U.K.) and Forestry Department, Accra (Ghana). Hawthorn, W. Protected forest areas in Ghana: interim notes. Forest (1991) Resources Management Project. Overseas Development Administration, London (U.K.) and Forestry Department, Accra (Ghana) . Hemming, J. The Society's "Maraca Rainforest Project", Roraima, (1989) Brazil. Geographical Journal 155(1);1-12. Hendrison, J. Damage-controlled logging in managed tropical rain forest in (1985) Suriname. University of Wageningen (Netherlands). HIID: Harvard Institute for International Development. The case for (1988) multiple-use management of tropical hardwood forests. A Study for the International Tropical Timber Organization, prepared by the HIID, Harvard University, Cambridge, Mass. (U.S.A.). Holdgate, M.W., Bruce, J., Camacho, R.F., Desai, N., Mahtab, F.U., (1989) Mascarenhas, 0., Maunder, Q.J., Shihab, H. and Tewungwa, S. Climate change: meeting the challenge. Report by a Commonwealth Group of Experts. Commonwealth Secretariat, London (U.K.). House, S. and Moritz, C. The impact of rainforest fragmentation on flora (1991) and fauna. Proc. Inst. Trop. Rainforest Studies, Townsville (Australia) . 97 Howe, H.F. Seed dispersal by birds and mammals: implications for seedling (1990) demography. In: Reproductive Ecology of Tropical Forest Plants (Eds* K.S. Bawa and M. Hadley). Parthenon Press, Carnforth (U.K.) and UNESCO, Paris (France). Hutchinson, I.D. Improvement thinning in natural tropical forests: aspects (1987) and institutionalisation. In: Natural Management of Tropical Moist Forests (Eds. F. Mergen and J.R. Vincent). Yale University, New Haven, Connecticut (U.S.A.). Hutchinson, I.D. Diagnostic sampling to orient silviculture and management (1991) in natural tropical forest. Comm. For. Rev. 70(3): 113-132. IBDF: Institute Brasileiro de Desarrollo Florestal. Piano do Sistema (1982) de Unidades de Conservacao do Brasil. II Etapa. IBDF/FBCN, Brasilia (Brazil). Ismail bin Haji Ali. A critical review of Malayan silviculture in the (1966) light of changing demand and form of timber utilisation. Malay Forester 29(4) : 228-33. ITTO: International Tropical Timber Organization. Guidelines for the (1990) sustainable management of natural tropical forests. Technical Series NS 5. ITTO, Yokohama (Japan). ITTO. Elements for the 1990 annual review and assessment of the world (1991) tropical timber situation. International Tropical Timber Council (ITTC), Paper (X)/3. ITTO, Yokohama (Japan). IUCN: International Union for Conservation of Nature and Natural Resources. (1980) World Conservation Strategy. Prepared by IUCN, UNEP, WWF, in technical collaboration with FAO and Uncesco. IUCN, Morges (Switzerland) . IUCN: The World Conservation Union. The 1990 United Nations List of (1990) National Parks and Protected Areas. Report prepared by World Conservation Monitoring Centre and the IUCN Commission on National Parks and Protected Areas. IUCN, Gland (Switzerland). IUCN. Caring for the World: a strategy for sustainability. (1991 a) IUCN, UNEP, WWF, in collaboration with FAO and Unesco. IUCN, Gland (Switzerland). IUCN. The Conservation Atlas of tropical forests: Asia and the (1991b) Pacific (Eds. N.M. Collins, J.A. Sayer and T.C. Whitmore). IUCN, Gland (Switzerland). Janzen, D.H. and Vasquez-Yanes, C. Aspects of tropical seed ecology of (1991) relevance to management of tropical forested wildlands. In; Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa, T.C. Whitmore and M. Hadley). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Jeffers, J.N.R. Outline studies in ecology: modelling. (1982) Chapman and Hall, London (U.K.) and New York (U.S.A.). 98 Jenkins, R.E. Information management for the conservation of biodiversity. (1988) In: Biodiversity (Ed. E.O. Wilson). National Academy Press, Washington D.C. (U.S.A.). Johns, A.D. Effects of selective timber extraction on rain forest (1988) structure and composition and some consequences for frugivores and folivores. Biotropica 20(1):31-37. Johns, A.D. Timber, the environment and wildlife in Malaysian rain (1989) forests: final report. Inst. of S.E. Asian Biology, University of Aberdeen, Scotland (U.K.). Johns, A.D. Species conservation in managed tropical forests. In; (1992) Tropical Deforestation and Species Extinction (Eds: T.C. Whitmore and J.A. Sayer). IUCN Forest Conservation Programme. Chapman and Hall, London (U.K.). Johnson, N., Cabarle, B. and Mead, D. Development assistance, natural (1991) forest management, and the future of tropical forests. World Resources Institute, Washington D.C. (U.S.A.). (Unpublished) . Jones, E.W. Ecological studies on the rain forest of southern Nigeria: (1955) the plateau forest of the Okumo Forest Reserve. Journal of Ecology 43:564-94. Jones, E.W. Ecological studies on the rain forest of southern Nigeria: (1956) the plateau forest of the Okumo Forest Reserve. Journal of Ecology 44:83-117. Jonkers, W.B.J. Vegetation structure, logging damage and silviculture (1987) in a tropical rain forest in Suriname. Agricultural University, Wageningen (Netherlands). Jonsson, T. and Lindgren, P. Logging technology for tropical forests: (1990) for or against? The Forest Operations Institute, Kista (Sweden). Jordan, C. Local effects of tropical deforestation. In: Conservation (1986) Biology: The Science of Scarcity and Diversity (Ed. M.E. Soul6). Sinauer Associates, Sunderland, Massachusetts (U.S.A.). Jury, S.L. Some recent computer-based developments in plant taxonomy. (1991) Bot.Journ. of the Linnean Society 106(2) ;121-128. Kageyama, P.Y. Genetic structure of tropical tree species of Brazil. In; (1990) Reproductive Ecology of Tropical Forest Plants (Eds. K.S. Bawa and M.Hadley). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France) . Kemp, R.H. The conservation of genetic resources in managed tropical (1992) forests. Unasylva 43(N9 169/2) :34-40. FAO, Rome (Italy). Kemp, R.H. and Lowe, R.G. The effect of tree size and environmental (1970) variables on increment of Khaya grandifoliola in a forest outlier in Nigeria. In: The Development of Forest Resources in the Economic Advancement of Nigeria (Eds. C.F.A. Onochie and S.K. Adeyoju). University of Ibadan (Nigeria). 99 Kese, K. An overview of Ghana's forest management policies and re- (1989) forestation programme. Address to Int. Hardw. Prod. Assoc.,San Diego (U.S.A.), April 1989. Kio, P.R.O. and Ekwebelan, S.A. Plantations versus natural forests for (1987) meeting Nigeria's wood needs. In: Natural Management of Tropical Moist Forests (Eds. F. Mergen and J.R. Vincent). Yale University, New Haven, Connecticut (USA). Leslie, A.J. A second look at the economics of natural management systems (1987) in tropical mixed forests. Unasylva 39 (N9 155/1): 46-58. FAO, Rome (Italy). Logan, W.E.M. Timber exploitation in the Gold Coast. Empire Forestry (1947) Review 26:30-53. Lovejoy, T.E., Birregaard, R.O., Rankin, J. and Schubart, H.O.R. (1983) Ecological dynamics of tropical forest fragments. In; Tropical Rainforest: Ecology and Management (Eds. S.L. Sutton, T.C. Whitmore and A.C. Chadwick) . Blackwell, Oxford (U.K.). MacKinnon, J. Guidelines for the development of conservation buffer zones (1981) and enclaves. PPA/WWF/FAO Nature Conservation Workshop, Bogor (Indonesia) . MacKinnon, J., MacKinnon, K., Child, G. and Thorsell, J. (Eds). Managing (1988) protected areas in the tropics. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland) . Maitre, H.F. Silvicultural interventions and their effects on forest (1991) dynamics and production in some rainforests of Cote d'lvoire. In: Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa, T.C. Whitmore and M. Hadley). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). Martin, C. The rainforests of West Africa. Birkhauser Verlag, Basel (1991) (Switzerland). Masson, J.L. Management of tropical mixed forests: preliminary assessment (1983) of present status. FO:MISC/83/17. FAO, Rome (Italy). McCarter, P.S. The evaluation of the international provenance trials of (1988) Cordia alliodora and Cedrela spp. Final Report, Overseas R4101, Development Administration. Project Institute, Oxford (U.K.). Oxford Forestry McNeely, J.A. Climate change and biological diversity: Policy (1990) implications. In: Landscape-Ecological Impact of Climatic Change (Eds. M.M. Boer and R.S. de Groot) . IOS Press, Amsterdam (Netherlands) . McNeely, J.A., Miller, K.R., Reid, W.V., Mittermeier, R.A. and (1990) Werner, T.B. (Eds). Conserving the world's biological diversity. IUCN, World Resources Institute, World Wildlife Fund-US and the World Bank. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland). 100 Mendoza, V. Estudio de algunas caracteristicas de la biologia floral de (1965) Cordia alliodora (Ruiz & Pav.)Cham. PhD Thesis, IICA, Turrialba (Costa Rica), (Unpublished). Mergen, F. and Vincent, J.R. (Eds). Natural management of tropical moist (1987) forests. Yale University, New Haven, Connecticut (U.S.A.). Monbiot, G. Amazon Watershed - the new environmental investigation. (1991) Michael Joseph, London (U.K.). Myers, N. Threatened biotas: "hot spots 11 in tropical forests. (1988) The Environmentalist 8 (3): 1-20. Namkoong, 6. Biological diversity: its conservation and use for (1990) sustainable agricultural, forestry and fisheries development. FAO Working Paper of the Inter-Departmental Sub-Group on Biological Diversity. FAO, Rome (Italy). Namkoong, G. Conservation and protection of ecosystems and genetic re- (1991) sources. Proc. 10th World Forestry Congress. Rev. For.Francaise, Hors Serie NQ 6. Ng, F.S.P. Germination ecology of Malaysian woody plants. Malayan (1980) Forester 63:606-637. Ng, F.S.P. Ecological principles of tropical lowland rain forest (1983) conservation. In; Tropical Rain Forest Ecology and Management (Eds. S.L. Sutton, T.C. Whitmore and A.C. Chadwick) . Blackwell, Oxford (U.K.). Nicholson, D.I. The effects of logging and treatment on the mixed diptero- (1979) carp forests of Southeast Asia. FO:MISC/79/8. FAO, Rome (Italy). NRC: National Research Council (U.S.A.). Managing global genetic resources: (1991) forest trees. National Academy Press, Washington D.C. (U.S.A.). Nwoboshi, L.C. Regeneration success of natural management, enrichment (1987) planting, and plantations of native species in West Africa. In; Natural Management of Tropical Moist Forests (Eds. F. Mergen and J.R. Vincent). Yale University, New Haven, Connecticut (U.S.A.). OFI: Oxford Forestry Institute. Incentives in producer and consumer (1991) countries to promote sustainable development of tropical forests. Pre-Project Report to the International Tropical Timber Organization NS PPR 22/91-M.F.I. ITTO, Yokohama (Japan). Ofosu-Asiedu, A. and Ampong, F.K. Preservation of wood as a means of (1990) extending the further utilization and using a wider range of species of tropical hardwood in Africa. In: Further Processing of Tropical Timber in Africa. International Tropical Timber Organization, Yokohama (Japan). Okafor, J.C. Varietal delimitation in Irvingia gabonensis (Irvingianceae) . (1975) Bull.Jard.Bot.Nat.Belg. 45:211-221. Oliver, C.D. and Larson, B.C. Forest stand dynamics. McGraw Hill (1990) Publishers (U.K.). 101 Opler, P. A., Baker, H.G. and Frankie, G.W. Reproductive biology of some (1975) Costa Rican Cordia species (Boraginaceae) . Biotropica 7(4): 234-247. Padua, M.T.J. Zoning and conservation units. In: Amazonia: Facts, (1989) Problems and Solutions. Proc. Symposium University of Sao Paulo, Aug. 1989, University of Sao Paulo, S.P. (Brazil). Palmer, J. Management of natural forest for sustainable timber production: (1989) a commentary. In: No Timber Without Trees (Ed. D. Poore). Earthscan Publications Ltd., London (U.K.). Palmberg, C. and Esquinas-Alcazar, J.T. The role of the United Nations (1990) agencies and other international organizations in the conservation of plant genetic resources. Forest Ecology and Management 35:171-197. Parant, B. Further processing and marketing of lesser known timber species (1990) in Africa. In; Further Processing of Tropical Timber in Africa. International Tropical Timber Organization, Yokohama (Japan). Pascal, J.P. Forest maps of South India. Institut Francais de Pondicherry, (1982) Pondicherry (India). Peters, C.M., Gentry, A.M. and Mendelsohn, R.O. Valuation of an Amazonian (1989) rain forest. Nature 339:655-656. Philip, M.S. Management systems in the tropical moist forests of the (1986) anglophone countries of Africa with conclusions. Paper prepared for the Forest Resources Division, FAO, Rome (Italy). (Unpublished; synthesized in FAO Forestry Paper nQ 88, see FAO 1989b). Poore, D. and Sayer, J.A. The management of tropical moist forest lands: (1987) ecological guidelines. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland). Poore, M.E.D. No timber without trees. Earthscan Publications Ltd., (1989) London (U.K.). Prance, G.T. The phytogeographic divisions of Amazonia and their influence (1977) on the selection of biological reserves. In: Extinction is Forever (Eds. G.T. Prance and T. Elias). New York Bot. Garden, New York N.Y. (U.S.A.). Prance, G.T. (Ed). Biological diversification in the tropics. Colombia (1982) University Press, New York N.'Y. (U.S.A.). Prance, G.T. The pollination of Amazonian plants. In: Key Environments: (1985) Amazonia (Eds. G.T. Prance and T.E. Lovejoy). Pergamon Press (U.S.A.). Prance, G.T. and Brown, K.S. The principal vegetation types of the (1987) Brazilian Amazon. In: Biogeography and Quaternary History in Tropical America (Eds. T.C. Whitmore and G.T. Prance). Clarendon Press, Oxford (U.K.). 102 Prance, G.T. Future of the Amazonian rainforest* Futures 22(9). (1990) Prescott-Allen, R. National conservation strategies and biological (1986) diversity. Report to the IUCN. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland) . Rai, S.N. and Proctor, J. Ecological studies on four rainforests in (1986) Karnataka, India. I. Environment, structure, floristics and biomass. Journal of Ecology 74(2): 439-454. Reid, W.V. and Miller, K.R. Keeping options alive: the scientific basis (1989) for conserving biodiversity. World Resources Institute, Washington D.C. (U.S.A.). Repetto, R. and Gillis, M. Public policies and the misuse of forest (1988) resources. Cambridge University Press, Cambridge (U.K.). Riggs, L.A. Conserving genetic resources on- site in forest ecosystems. (1990) Forest Ecology and Management 35:45-68. Roche, L. and Dourojeanni, M.J. A Guide to in situ conservation of genetic (1984) resources of tropical woody species. FORGEN/MISC/84/2. FAO, Rome (Italy). Rylands, A.B. Situajao atual das unidades de conserva^ao na Amazonia (1990) Brazileira. Paper to "Workshop 90", Manaus, January 1990. IBDF, Brasilia (Brazil). Salati, E. The forest and the hydrological cycle. In; The Geophysiology (1987) of Amazonia (Ed. R.E. Dickinson). Wiley, New York N.Y. (U.S.A.). Sayer, J.A. The management of tropical forest buffer zones. The World (1991) Conservation Union, Gland (Switzerland). Sayer, J.A., McNeely, J.A. and Stuart, S.N. The conservation of tropical (1990) forest vertebrates. In: Vertebrates in the Tropics (Eds. G. Peters and R. Hutterer). Museum Alex. Koenig, Bonn (Germany). Sayer, J.A. and Whitmore, T.C. Tropical moist forests: destruction and (1991) species extinction. Biological Conservation 55:199-213. Schmidt, R.C. Tropical rain forest management. Unasylva 3j)(NQ 156/2); (1987) 2-17. FAO, Rome (Italy). Schmidt, R.C. Current tropical moist forest management activities in (1989) Brazil. FO:MISC/89/4. FAO, Rome (Italy). Schmidt, R.C. Tropical rain forest management: a status report. In: (1991) Rain Forest Regeneration and Management (Eds. A. Gomez-Pompa, T.C. Whitmore and M. Hadley). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France). 103 Schubart, H.O.R. Commercial forestry and the conservation of biodiversity (1990) in tropical forests: the situation in Brazil. Realistic Strategies for the Conservation of Biodiversity in Tropical Moist Forest. IUCN Workshop 5; IUCN General Assembly, Perth, Australia. The World Conservation Union, Gland (Switzerland). Sedjo, R.A. and Lyon, K.S. The long-term adequacy of world timber supply. (1990) Resources for the Future, Washington B.C. (U.S.A.). Shuttleworth, W.J. Evaporation from Amazonian rainforest. Proc. Roy. (1988) Soc. London, Biological Sciences 233(321-346). London (U.K.). Shyamsunder, S. and Parameswarappa, S. Forestry in India - the forester's (1987) view. Ambio ^6:332-7. Simberloff, D. Do species-area curves predict extinction in fragmented (1992) forest? In; Tropical Deforestation and Species Extinction (Eds T.C. Whitmore and J.A. Sayer). IUCN Forest Conservation Programme. Chapman and Hall, London (U.K.). Singh, S. Protected areas in India. In: Conserving Asia's Natural (1985) Heritage (Ed. W.J. ThorselTJ. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland) . Siqueira, J.D.P. Sustained forest management in the Amazon: need versus (1989) research. In: Amazonia: Facts, Problems and Solutions. Proc. Symp. University of Sao Paulo, Aug. 1989. University of Sao Paulo, S.P. (Brazil). Soul6, M.E. (Ed). Conservation biology: the science of scarcity and (1986) diversity. Sinauer Associates, Sunderland, Massachusetts (U.S.A.). Stead, J. Commonwealth Forestry Institute international provenance trials (1980) of Cordia alliodora (R & P) Oken. Paper presented at the Eleventh Commonwealth Forestry Conference, Trinidad and Tobago and Jamaica, September 1980. (Unpublished). Swaine, M.D. and Whitmore, T.C. On the definition of ecological species (1988) groups in tropical rain forests. Vegetation 7J>:81-86. Tangley, L. Cataloguing Costa Rica's diversity. BioScience, 40:633-636. (1990) Terborgh, J. Keystone plant resources in the tropical forest. In; (1986) Conservation Biology (Ed. M. Soul6). Sinauer, Sunderland (U.S.A.). Tufuor, K. Status of forest areas allocated to timber production and (1990a) their contribution to the conservation of biological diversity. Ghana Country Paper, IUCN Workshop on Realistic Strategies for the Conservation of Biodiversity in Tropical Moist Forests, Perth, Australia. The World Conservation Union, Gland (Switzerland). 104 Tufuor, K. Cross-country issues in forest conservation and management. (1990b) Paper presented to the Conference on the Conservation and Rational Utilization of the Closed Forest of Central and West Africa, Abidjan, 5-9 November, 1990. IUCN, African Development Bank/World Bank. The World Conservation Union, Gland (Switzerland). UNDP/FAO. Management Plan for the Tapajos National Fores t.FO: BRA/78/003, (1980) Technical Report 2. UNDP/FAO/IBDF-Brasilia. FAO, Rome (Italy). UNESCO: United Nations Educational, Scientific and Cultural Organization. (1984) Action plan for biosphere reserves. Nature and Resources 20(4). Unesco, Paris (France). Vanclay, J.K. A growth model for north Queensland rainforests. Forest (1989) Ecology and Management 27(3-4) :245-271 . Vanniere, B. Les possibilities d ' amenagement de la foret dense Africaine. (1975) FAO/CTFT, Nogent-sur-Marne. FAO, Rome (Italy). Viana, V.M. Seed and seedling availability as a basis for management of (1990) natural forest regeneration. In: Alternatives to Deforestation: Steps Toward the Sustainable Use of Amazon Rain Forest. Columbia University Press, New York N.Y. (U.S.A.). Vivien, J. and Faure, J.J. Arbres des forSts denses d'Afrique Cent rale. (1985) Agence de Cooperation Culturelle et Technique, Ministere des Relations Exterieures Cooperation et Developpement, Paris (France) Wadsworth, F.H. Natural forests in the development of the humid American (1975) tropics. In: Proc. Meeting on the Use of Ecological Guidelines for Development in the American Humid Tropics. Caracas (Venezuela) 20-22 February 1974. International Union for Conservation of Nature and Natural Resources, Gland (Switzerland). WCMC: World Conservation Monitoring Centre. Pre-project study on the (1991) conservation status of tropical timbers in trade. Final Report to the International Tropical Timber Organization, February 1991. ITTO, Yokohama (Japan). Webb, L.J. The human face in forest management. In: Socio-Economic Effects (1982) and Constraints in Tropical Forest Management (Ed. E.G. Hallsworth). Wiley, Chichester (U.K.). Wells, M., Brandon, K. and Hannah. L. People and parks: linking protected (1990) area management with local communities. World Bank, Washington, D.C. (U.S.A.) (Unpublished). Whitmore, T.C. Tropical rain forests. Clarendon Press, Oxford (U.K.). (1990) Whitmore, T.C. Tropical rain forest dynamics and its implications for (1991) for management. In: Rain Forest Regeneration and management (Eds. A. Gomez-Pompa et al). Parthenon, Carnforth (U.K.) and UNESCO, Paris (France!"! 105 Whitmore, T.C. and Sayer, J.A. (Eds). Tropical deforestation and species (1992) extinction. IUCN Forest Conservation Programme. Chapman and Hall, London (U.K.). Wilcox, B.A. Requirements for the establishment of a global network of (1990) in situ conservation areas for plants and animals. Forest Resources Division. FAO, Rome (Italy) (Unpublished). Wilson, E.O. (Ed). Biodiversity. National Academy Press, Washington D.C. (1988) (U.S.A.). Wong, J.L.G. (Ed). Forestry inventory project. Seminar Proceedings, (1989) 29-30 March, 1989. Forestry Department, Accra (Ghana). WRI: World Resources Institute. World resources 1990-91. WRI/UNEP/UNDP. (1990) Oxford University Press, Oxford (U.K.) and New York N.Y (U.S.A.). WRI Colloquium on sustainability in natural tropical forest (1991) management summary report. World Resources Institute, Washington D.C. (U.S.A.). WRI Global biodiversity strategy: guidelines for action to save, (1992) study and use earth's biotic wealth sustainably and equitably. World Resources Institute, the World Conservation Union and the United Nations Environment Programme. WRI, Washington (USA). WWF: Worldwildlife Fund/U.S.A. Views from the forest: natural forest (1991) management initiatives in Latin America. Worldwildlife Fund, Washington D.C. (U.S.A.). Wyatt-Smith, J. Manual of Malayan silviculture for inland forests. (1963) Malay. For. Rec. NQ 23. Forestry Department, Malaysia. Wyatt-Smith, J. The management of tropical moist forest for the sus~ (1987a) tained production of timber: some issues. IUCN/IIED Tropical Forest Policy Paper No 4. International Institute for Environment and Development (IIED), London (U.K.). Wyatt-Smith, J. Problems and prospects for natural management of tropical (1987b) moist forests. In: Natural Management of Tropical Moist Forests (Eds. F. Mergen and J.R. Vincent). Yale University, New Haven, Connecticut (U.S.A.). ; 111403 FAO TECHNICAL PAPERS FAO FORESTRY PAPERS 1 Forest utilization contracts on public land, 1977 30 (E F S) 31 2 Planning forest roads and harvesting systems, 32 1977(EFS) 3 World liat of forestry schools, 1 977 (E/F/S) 33 3 Rev. 1 . World liat of forestry schools, 1 98 1 (E/F/S) 34 3 Rav. 2. World list of forestry schools, 1 986 (E/F/S) 35 4/1 World pulp and paper demand, supply and trade 36 -Vol. 1,1977 (EFS) 37 4/2 World pulp and paper demand, supply and trade -Vol.2, 1977 (EFS) 38 5 The marketing of tropical wood, 1 976 (E S) 39 6 National parks planning, 1 976 (E F S ) 40 7 Forestry for local community development, 1978 41 (Ar EFS) 8 Establishment techniques for forest plantations, 42 1978(ArCE* F S) 9 Wood chips - production, handling, transport, 43 1976(CES) 10/1 Assessment of logging costs from forest 44/1 inventories in the tropics - 1 . Principles and methodology, 1978 (E F S) 44/2 1 0/2 Assessment of logging costs from forest inventories in the tropics - 2. Data collection and calculations, 1978 (E F S) 44/3 1 1 Savanna afforestation in Africa, 1 977 (E F) 12 China: forestry support for agriculture, 1978 (E) 45 1 3 Forest products prices 1 960-1 977, 1 979 (E/F/S) 46 1 4 Mountain forest roads and harvesting, 1 979 (E) 47 1 4 Rev. 1 . Logging and transport in steep terrain, 1 985 (E) 1 5 AGRIS forestry - world catalogue of information 48 and documentation services, 1979 (E/F/S) 49 16 China: integrated wood processing industries, 1979 (E F S) 50 17 Economic analysis of forestry projects, 1979 50/1 (EFS) 17 Sup. 1 . Economic analysis of forestry projects: 51 /I case studies, 1979(ES) ' 17 Sup. 2. Economic analysis of forestry projects: readings, 52/1 1980(CE) 1 8 Forest products prices 1 960- 1 978, 1 980 (E/F/S) 52/2 1 9/1 Pulping and paper-making properties of fast-growing plantation wood species 53 -Vol. 1,1 980 (E) 19/2 Pulping and paper-making properties of 54 fast-growing plantation wood species 55 - Vol. 2, 1980 (E) 20 Forest tree improvement, 1985 (C E F S) 56 20/2 A guide to forest seed handling, 1 985 (E S ) 57 21 Impact on soils of fast-growing species in lowland 58 humid tropics, 1 980 (E F S) 59 22/1 Forest volume estimation and yield prediction - Vol. 1 . Volume estimation, 1980 (C E F S) 60 22/2 Forest volume estimation and yield prediction - Vol. 2. Yield prediction, 1980 (C E F S) 61 23 Forest products prices 1961-1980, 1981 (E/F/S) 62 24 Cable logging systems, 1 981 (C E) 25 Public forestry administrations in Latin America, 63 1981 (E) 64 26 Forestry and rural development, 1981 (EFS) 65 27 Manual of forest inventory, 1 981 (E F) 28 Smalt and medium sawmills in developing countries, 66 1981 (ES) 67 29 World forest products, demand and supply 1 990 and 2000, 1982 (EFS) 68 Tropical forest resources, 1982 (E F S) Appropriate technology in forestry, 1982 (E) Classification and definitions of forest product! 1982(Ar/E/F/S) Logging of mountain forests, 1982 (E F S) Fruit-bearing forest trees, 1982 (E F S) Forestry in China, 1982 (C E) Basic technology in forest operations, 1982 (E Conservation and development of tropical fore resources, 1982 (EFS) Forest products prices 1962-1981, 1982 (E/F< Frame saw manual, 1982 (E) Circular saw manual, 1983 (E) Simple technologies for charcoal making, 198 (EFS) Fuel wood supplies in the developing countries 1983(ArEFS) Forest revenue systems in developing countri* 1983 (EFS) Food and fruit-bearing forest species - 1. Examples from eastern Africa, 1983 (E F Food and fruit-bearing forest species - 2. Examples from southeastern Asia, 1984 (EFS) Food and fruit-bearing forest species - 3. Exi from Latin America, 1986(ES) Establishing pulp and paper mills, 1983 (E) Forest products prices 1963-1982, 1983 (E/l Technical forestry education - design and implementation, 1984 (E F S) Land evaluation for forestry, 1984 (C E F S) Wood extraction with oxen and agricultural tractors, 1986 (EFS) Changes in shifting cultivation in Africa, 1 98 Changes in shifting cultivation in Africa - seven case-studies, 1 985 (E) Studies on the volume and yield of tropical f stands - 1 . Dry forest formations, 1 989 (E I Cost estimating in sawmilling industries: gui 1984(E) Field manual on cost estimation in sawmillin industries, 1985 (E) Intensive multiple-use forest management ir 1984 (EFS) Planificaci6n del desarrollo forestal, 1984 (S Intensive multiple-use forest management ir tropics, 1985 (EFS) Breeding poplars for disease resistance, 1 9! Coconut wood - processing and use, 1 985 Sawdoctoring manual, 1 985 (E S) The ecological effects of eucalyptus, 1985 (C E F S) Monitoring and evaluation of participatory f projects, 1985 (EFS) Forest products prices 1965-1984, 1985 (! World list of institutions engaged in forestr forest products research, 1985 (E/F/S) Industrial charcoal making, 1985 (E) Tree growing by rural people, 1985 (Ar E F Forest legislation in selected African count 1986(EF) Forestry extension organization, 1986 (C E Some medicinal forest plants of Africa and America, 1986 (E) Appropriate forest industries, 1986 (E) 69 Management of forest Industrie*, 1 986 (E) 70 Wildland fire management terminology, 1986 (E/F/S) 71 World compendium of forestry and forest products research institutions, 1986 (E/F/S) 72 Wood gas as engine fuel, 1 986 (E S) 73 Forest products: world outlook projections 1985-2000, 1986 (E/F/S) 74 Guidelines for forestry information processing, 1986 (E) 75 An operational guide to the monitoring and evaluation of social forestry in India, 1986 (E) 76 Wood preservation manual, 1986 (E) 77 Databook on endangered tree and shrub species and provenances, 1986 (E) 78 Appropriate wood harvesting in plantation forests, 1987(E) 79 Small-scale forest-based processing enterprises, 1987(EFS) 80 Forestry extension methods, 1987 (E) 81 Guidelines for forest policy formulation, 1987 (C E) 82 Forest products prices 1967-1986, 1988 (E/F/S) 83 Trade in forest products: a study of the barriers faced by the developing countries, 1 988 (E) 84 Forest products: world outlook projections 1987-2000 - product and country tables, 1988 (E/F/S) 85 Forestry extension curricula, 1988 (E/F/S) 86 Forestry policies in Europe, 1988 (E) 87 Small-scale harvesting operations of wood and non-wood forest products involving rural people, 1988 (EPS) 88 Management of tropical moist forests in Africa, 1989(EFP) 89 Review of forest management systems of tropical Asia, 1989 (E) 90 Forestry and food security, 1989 (Ar E S) 91 Design manual on basic wood harvesting technology, 1989 (EPS) (Published only as FAO Training Series, No. 18) 92 Forestry policies in Europe - an analysis, 1989 (E) 93 Energy conservation in the mechanical forest industries, 1990 (ES) 94 Manual on sawmill operational maintenance, 1990(E) 95 Forest products prices 1 969- 1 988, 1 990 (E/F/S) 96 Planning and managing forestry research: guidelines for managers, 1990(E) 97 Non-wood forest products: the way ahead, 1991 (ES) 98 Las plantations a vocation de bois d'osuvre en Afrique intertropicale humide, 1991 (F) 99 Cost control in forest harvesting and road construction, 1992 (E) 100 Introduction to ergonomics in forestry in developing countries, 1992(E) 101 Amtnagement et conservation des forto denses en Ame*rique tropicele, 1992 (F) 102 Research management in forestry, 1991 (E) 103 Mixed and pure forest plantations in the tropics and subtropics, 1992 (E) 104 Forest products prices 1971-1990, 1992 (E) 105 Compendium of pulp and paper training and research institutions, 1 992 (E) 106 Economic assessment of forestry project impacts, 1992(E) 107 Conservation of genetic resources in tropical forest management: Principles and concepts, 1 993 (E) 108 A decade of wood energy activities within the Nairobi Programme of Action, 1 993 (E) Availability: February 1993 Ar C E F P S Arabic Chinese English French Portuguese Spanish Muttil - Multilingual Out of print " * In preparation The FAO Technical Papers are available through the authorized FAO Sales Agents or directly from Distribution and Sales Section, FAO, Viale delle Terme di Caracalla, 00100 Rome, Italy.